Psi Chi Journal of Psychological Research – Summer 2023

ISSN: 2325-7342

Published by Psi Chi, The International Honor Society in Psychology ®

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH SUMMER 2023 | VOLUME 28, NUMBER 2

EDITOR

STEVEN V. ROUSE, PhD

Pepperdine University

Telephone: (310) 506-7959

Email: steve.rouse@psichi.org

ASSOCIATE EDITORS

JENNIFER L. HUGHES, PhD

Agnes Scott College

STELLA LOPEZ, PhD

University of Texas at San Antonio

TAMMY LOWERY ZACCHILLI, PhD Saint Leo University

ALBEE MENDOZA, PhD

Delaware State University

KIMBERLI R. H. TREADWELL, PhD University of Connecticut

ROBERT R. WRIGHT, PhD

Brigham Young University-Idaho

EDITOR EMERITUS

DEBI BRANNAN, PhD

Western Oregon University

MANAGING EDITOR

BRADLEY CANNON

DESIGNER

JANET REISS

EDITORIAL ASSISTANT

EMMA SULLIVAN

ADVISORY EDITORIAL BOARD

GLENA ANDREWS, PhD

RAF Lakenheath USAF Medical Center

AZENETT A. GARZA CABALLERO, PhD Weber State University

MARTIN DOWNING, PhD Lehman College

HEATHER HAAS, PhD

University of Montana Western

ALLEN H. KENISTON, PhD

University of Wisconsin–Eau Claire

MARIANNE E. LLOYD, PhD Seton Hall University

DONELLE C. POSEY, PhD

Washington State University

LISA ROSEN, PhD

Texas Women's University

CHRISTINA SINISI, PhD

Charleston Southern University

PAUL SMITH, PhD

Alverno College

ABOUT PSI CHI

Psi Chi is the International Honor Society in Psychology, founded in 1929. Its mission: "recognizing and promoting excellence in the science and application of psychology." Membership is open to undergraduates, graduate students, faculty, and alumni making the study of psychology one of their major interests and who meet Psi Chi’s minimum qualifications. Psi Chi is a member of the Association of College Honor Societies (ACHS), and is an affiliate of the American Psychological Association (APA) and the Association for Psychological Science (APS). Psi Chi’s sister honor society is Psi Beta, the national honor society in psychology for community and junior colleges.

Psi Chi functions as a federation of chapters located at over 1,180 senior colleges and universities around the world. The Psi Chi Headquarters is located in Chattanooga, Tennessee. A Board of Directors, composed of psychology faculty who are Psi Chi members and who are elected by the chapters, guides the affairs of the Organization and sets policy with the approval of the chapters. Psi Chi membership provides two major opportunities. The first of these is academic recognition to all inductees by the mere fact of membership. The second is the opportunity of each of the Society’s local chapters to nourish and stimulate the professional growth of all members through fellowship and activities designed to augment and enhance the regular curriculum. In addition, the Organization provides programs to help achieve these goals including conventions, research awards and grants competitions, and publication opportunities.

JOURNAL PURPOSE STATEMENT

The twofold purpose of the Psi Chi Journal of Psychological Research is to foster and reward the scholarly efforts of Psi Chi members, whether students or faculty, as well as to provide them with a valuable learning experience. The articles published in the Journal represent the work of undergraduates, graduate students, and faculty; the Journal is dedicated to increasing its scope and relevance by accepting and involving diverse people of varied racial, ethnic, gender identity, sexual orientation, religious, and social class backgrounds, among many others. To further support authors and enhance Journal visibility, articles are now available in the PsycINFO®, EBSCO®, Crossref®, and Google Scholar databases. In 2016, the Journal also became open access (i.e., free online to all readers and authors) to broaden the dissemination of research across the psychological science community.

JOURNAL INFORMATION

The Psi Chi Journal of Psychological Research (ISSN 2325­7342) is published quarterly in one volume per year by Psi Chi, Inc., The International Honor Society in Psychology. For more information, contact Psi Chi Headquarters, Publication and Subscriptions, 651 East 4th Street, Suite 600, Chattanooga, TN 37403, (423) 756­2044. www.psichi.org; psichijournal@psichi.org

Statements of fact or opinion are the responsibility of the authors alone and do not imply an opinion on the part of the officers or members of Psi Chi.

Advertisements that appear in Psi Chi Journal do not represent endorsement by Psi Chi of the advertiser or the product. Psi Chi neither endorses nor is responsible for the content of third­party promotions. Learn about advertising with Psi Chi at http://www.psichi.org/Advertise

COPYRIGHT

Permission must be obtained from Psi Chi to reprint or adapt a table or figure; to reprint quotations exceeding the limits of fair use from one source, and/or to reprint any portion of poetry, prose, or song lyrics. All persons wishing to utilize any of the above materials must write to the publisher to request nonexclusive world rights in all languages to use copyrighted material in the present article and in future print and nonprint editions. All persons wishing to utilize any of the above materials are responsible for obtaining proper permission from copyright owners and are liable for any and all licensing fees required. All persons wishing to utilize any of the above materials must include copies of all permissions and credit lines with the article submission.

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

91 Editorial: The Use of Mechanical Turk Data in Psychological Research

Stella G. Lopez1, Steven V. Rouse2, and Kimberli Treadwell3

1Department of Psychology, University of Texas at San Antonio

2Social Sciences Division, Pepperdine University

3Department of Psychological Sciences, University of Connecticut

96 Swiping Away Your Well-Being? Examining Well-Being Indicators Among TikTok Account Holders

Christian Nienstedt, Nathan Smith, Hannah Braithwaite, Ben Gilbert, and Robert R. Wright*

Department of Psychology, Brigham Young University – Idaho

107 The Impact of Age and Gender on Trust in the U.S. Federal Government

Kailey A. Schroeder, Christopher A. Creecy*, Emily L. Perkins, and K. E. Carter

Department of Behavioral Sciences, Freed-Hardeman University

114 The Effects of Imagining and Experiencing Ostracism on Pain Perception

Jennifer Zwolinski

Department of Psychological Sciences, University of San Diego

123 The Role of Parental Pressure and Warmth in the Relationship Between Parental Involvement, Parental Expectations, and Child Academic Success

Madison T. Weir, Janet P. Trammell*, and Jennifer Harriger*

Social Science Division, Seaver College, Pepperdine University

132 Minority Stress and Psychological Distress Among Asexual Transgender and Gender Nonconforming Individuals

Jared W. Boot Haury

Department of Clinical Psychology, Michigan School of Psychology

142 Role of Social Gaze on Visual Search in an Eye-Tracking Paradigm

Jonathan K. Kroeger, Erin A. Conway, and Ralph G. Hale*

Department of Psychological Science, University of North Georgia

149 Threatened Social Needs After Exclusion in Undergraduate Students With Varying Degrees of Attention Switching Difficulties

Jessica C. Reich and Richard S. Pond, Jr.*

Department of Psychology, University of North Carolina Wilmington

157 Like Mother, Like Daughter: Excessive Reassurance Seeking Across Mother–Daughter and Romantic Relationships

Jaidyn K. Charlton, Elizabeth Grassia, Alexandra D. Popowich*, and Aislin R. Mushquash*

Department of Psychology, Lakehead University

Editorial: The Use of Mechanical Turk Data in Psychological Research

Stella G. Lopez1, Steven V. Rouse2, and Kimberli Treadwell3

1Department of Psychology, University of Texas at San Antonio

2Social Sciences Division, Pepperdine University

3Department of Psychological Sciences, University of Connecticut

ABSTRACT. As psychological researchers increasingly collect data from Workers on Amazon’s Mechanical Turk (MTurk), we examine the history of this data collection platform, reasons for both confidence and caution in the use of the data, and factors that differentiate MTurk data from data collected using more traditional methods. This discussion includes issues of recruitment, responder bias, the psychometric quality of the data, and ethical issues relating to incentives and compensation.

Keywords: crowdsourcing platforms, Mechanical Turk, data collection strategies, nontraditional methods, psychological research.

Who would have guessed that an 18th century hoax could have a big effect on psychological research in the 21st century? The original “Mechanical Turk” was seen as a marvel of early artificial intelligence (Standage, 2002). Its creator, Wolfgang von Kempelen (an engineer who developed many steampowered inventions), created a chess­playing machine that he unveiled to impress the Austrian empress. Although the actual movements of the pieces were performed by a robotic mannequin dressed in Turkish clothes (representing beliefs about the historical origins of the game of chess), von Kempelen was willing to show his audiences the power behind the machine by opening the cabinet under the mannequin to reveal complicated whirring gears and spinning wheels, all steam powered. However, he refused to explain exactly how his machine functioned. Von Kempelen and his machine wowed audiences during their tour of Europe, besting several chess masters and such notable people as Benjamin Franklin, Napoleon Bonaparte, and Charles Babbage (the mathematician who later created the first mechanical mathematical computers). It wasn’t until many years after von Kempelen’s death that his invention was revealed as a hoax; behind the ineffectual gears and wheels was a compartment large enough for a human chess master to sit, operating the mannequin’s hands with a system of levers. Since then, the term “mechanical Turk” has been used to describe functions that appear to be performed by computers

but are, in reality, performed by people. In other words, a “mechanical Turk” is an artificial form of artificial intelligence. It wasn’t until the latter half of the 20th century that digital computers became strong enough to play chess against humans, culminating in the 1997 defeat of chess Grandmaster Garry Kasparov by the IBM Deep Blue computer.

A few years after Kasparov’s defeat, the online retailer Amazon revived the term “Mechanical Turk.” As the online store was beginning to sell a broad range of products, they needed to find a way to categorize their inventory quickly and accurately. One possibility would be to create an artificial intelligence system that could be programmed, for example, to categorize a toaster oven as a kitchen appliance. Many such machine­learning applications do exist, but their utility is based heavily on extensive programming and constant maintenance. It is more expedient to hire a team of human Workers and pay each one a few cents each time they categorize a product (a “Human Intelligence Task” or “HIT”). For example, simply categorizing the new toaster oven as a kitchen appliance might be a HIT that requires 5 seconds and results in a payment of $0.01. As Amazon began to employ thousands of Workers, remotely receiving payment for the HITs they completed, this system began to be known as Amazon’s Mechanical Turk, or MTurk. In 2005, Amazon realized that MTurk was not only a way to solve their own practical productcategorization challenges, but they could also profit

from it. Other people could get accounts to become “Requesters,” allowing them to pay the Workers (with an additional fee paid to Amazon) for a variety of internetbased HITs. For example, a website designer could get a Requester account and then hire hundreds of Workers, paying them a few cents each, to look at a website and rate its aesthetic appeal or navigability.

In 2010, psychological researchers began utilizing MTurk Workers as an inexpensive and easily accessible source of research data. The next few years saw rapid growth in the publication of MTurk­based research. Today, MTurk’s impact on the field of psychological research cannot be underestimated. For example, of the articles published in the 2022 volume of the Psi Chi Journal of Psychological Research, 20% used data collected from MTurk or Prolific (a newer competitor to MTurk). Is this a desirable trend? Are there reasons to be concerned as MTurk ­ based data represents a substantial proportion of published research studies? Or is this an appropriate method of gathering data for psychological research?

The primary appeals of MTurk as a source of subject recruitment and data collection are the relatively inexpensive nature of compensating Workers and the rapidness of data collection (Buhrmeister et al., 2011). This article addresses these benefits, as well as concerns about the motivations of Workers, the quality of data collected, and whether the results from MTurk­based studies can be generalized.

Ease of Recruitment

During the pandemic years beginning in 2020, the traditional face­to­face recruitment of study participants was considered risky. Therefore, many researchers used the MTurk system at this time as a relatively “safe” way to recruit participants for studies during the pandemic years. MTurk made it relatively easy for psychological researchers to join the platform as Requestors, recruit Workers as study participants, and then have them generate data by completing online surveys or other types of online assessments.

Workers are paid for tasks they complete within a designated timeframe. Compensation involves money deposited in an account by the Requester. The compensation amount is set prior to posting the task (Buhrmester et al., 2011). Workers have an Amazon payment account, and the monies are paid either in cash that is deposited into a bank account or are used as credits in the Amazon website (Sheehan & Pittman, 2016). Payment rates can vary starting from a penny for a HIT (Buhrmester et al., 2011). Thus, workers attend to posted HITs, respond to the recruitment call, complete the task or tasks, and are compensated. Workers have relative freedom to choose

which HITs to do, how many to do, and can quit at any time (Paolacci & Chandler, 2014).

There are numerous reasons why MTurk has become a popular platform for at least some psychological researchers. There are always thousands of Workers available at any time of the day or night, so data collection or research task completion can occur almost continuously without stopping, unlike using local college students for participants. Completion of datagathering can occur in a matter of hours, not days. Also, HITs can be set at a low, affordable wage when research funding is lacking or large numbers of participants are needed (Paolacci & Chandler, 2014). Workers are able to work with a standardized user interface, and there is supposedly a high level of trust with Amazon regarding compensation (Sheehan & Pittman, 2016).

Worker Diversity and Generalizability

As Sheehan and Pittman (2016) summarized, MTurk Worker samples more adequately reflect the diversity of a population relative to college samples. There is greater diversity among Workers in terms of age, gender, ethnic background, marital status, socioeconomic level, and other important demographic characteristics (Difallah et al., 2018). Most “Turkers” are reportedly high on education level and are technically literate (Paolacci & Chandler, 2014; Sheehan & Pittman, 2016). They are from different places worldwide, though it is dominated by residents from the United States (Difallah et al., 2018; Paolacci & Chandler, 2014; Sheehan & Pittman, 2016).

The greater diversity of Workers may translate into better external validity for research results generated via MTurk. Generalizing a study’s results is thus more straightforward considering the vast access to a population for the investigation of a psychological phenomenon using MTurk. Thus, MTurk samples are here to stay and likely will lead to the development of other crowdsourcing platforms for participant recruitment and data collection.

Although Workers are diverse, can we be confident that the results obtained from MTurk respondents generalize to the broader population? After all, most MTurk respondents indicate that they are completing surveys to make money, and MTurk HITs are a primary source of income for many Workers (Keith et al., 2019; Vanhove et al., 2021). This raises the question of whether or not this specific subset of people is representative of the general population. Research has suggested that an MTurk sample has far more diversity than one might find in a typical college sample, especially regarding age, race, ethnicity, and gender (Sheehan, 2018). Nevertheless, U.S. ­ based MTurk samples tend to differ from the U.S. population on several important demographic

Lopez, Rouse, and Treadwell | Use of Mechanical Turk Data

Use of Mechanical Turk Data | Lopez, Rouse, and Treadwell

characteristics, including a greater likelihood of identifying as atheist or agnostic (Burnham et al., 2018), having more children (Jensen­Doss et al., 2022), being more highly educated (Ogletree & Katz, 2021), and being unemployed or underemployed (Keith et al., 2019; Vanhove et al., 2021). Despite these demographic differences, Merz et al. (2022) did not find significant differences with a community comparison sample on any broad cognitive ability, although Ogletree and Katz (2021) found higher verbal analogy abilities and higher verbal fluency abilities among older adults on MTurk than in a comparison group. McCredie and Morey (2019) compared an MTurk sample with normative data for a prominent measure of personality pathology; of the 47 pathology­related scales on the personality test, there were only two scales (social detachment and depression) for which the MTurk mean was outside the normative sample’s 95% confidence interval. In short, research has suggested that MTurk samples are more diverse than many college samples. Although these samples should not be construed as being representative of the U.S. population on all demographic characteristics, they do not differ from community samples on most psychological characteristics. Other studies that have compared data collected in the traditional method of college samples with data collected from MTurk report that they are not that different (Sheehan & Pittman, 2016). In some cases, data collection through digital or virtual means have been found to have lower biases than the traditional student samples (Buhrmester et al., 2011). In addition, attention issues are reportedly similar between Workers and other participants (Paolacci & Chandler, 2014).

Worker Motivations

Online platforms such as MTurk provide one means of obtaining samples, as does the long­standing use of institutional psychology department undergraduate research pools. However, the ethics surrounding the use of each of these recruitment methods brings with it a need for careful attention to issues, including the use of compensation and coercion in accessing these potentially vulnerable populations.

Compensation

Kaufmann and colleagues (2011) described both extrinsic and intrinsic motivations for participation in studies by MTurk Workers. The main extrinsic reason is obviously the monetary compensation. Even though money is a major motivator, monetary compensation can range in any amount starting with a penny for completing a single HIT. Although Workers may recognize that the money earned from MTurk is not a major source of

income, it is still considered a significant and positive supplement to one’s budget (Kaufmann et al., 2011).

A criticism of MTurk is that a minimum wage is not defined. The MTurk competitor Prolific Academic, created by Oxford University in 2014, criticizes MTurk for this, and sets a minimum of € 6.00 per hour (Prolific, 2022). However, it is difficult to compare wages across international boundaries as well as various geographic areas in the United States. Hence, there is widespread variability in pay rates to research participants across these platforms, and the establishment of a base rate does not necessarily indicate a payment commensurate with time and effort exerted. On any given platform, researchers set the wage for employment of participants and can take into account demographics sought and ethical wages.

Another consideration for researchers when determining a wage for participants is the estimate of hourly wage based on the length of time to complete the online study. Researchers may tend to underestimate the time for completion given the competing motivation of more participants for available funds. It may be useful to have undergraduates or persons similar to the target sample complete the experimental task on a pilot basis in order to obtain as realistic an estimate as possible for establishing an hourly wage compensation. Another consideration for online platforms is that, unlike in­person jobs, persons working on an online platform take time between studies to seek the next work opportunity, and this work­seeking time is unpaid. This inherent unpaid time between jobs is typically not part of the calculation for hourly wages.

Intrinsic Motivations

A number of intrinsic motivations may drive Workers completing psychological research (Antin & Shaw, 2012; Kaufmann et al., 2011). Some Workers report that the work is fun, pleasant, passes the time, and contains challenging tasks. Other reasons they may participate include a flexible work schedule (i.e., it can be done at any time a Worker has time to do a HIT), the opportunity to learn something new, the opportunity to attain networking prospects when meeting fellow Turkers in online forums, and to utilize Workers’ skills and knowledge in computers and the Internet.

Quality of Data

Because Workers can complete any number and variety of HITs at any time, the question arises as to how this arrangement may affect the quality of data gathered. Are the data biased in any way, relative to those data collected via more traditional methods?

With MTurk participants, investigators have noted

certain factors that may affect the quality of data. Workers may be distracted while they are completing a HIT, some may even be multitasking. Some Workers may not be paying attention to instructions, while others may be careless in responding or completing HITs to complete as many HITs as possible for the monetary compensation (Sheehan & Pittman, 2016). Workers’ responses have been reported to be more careless than student groups (e.g., Aruguete et al., 2019). Due to the anonymity that MTurk Workers have, there may be a perception of low to no negative consequences for being careless. Nevertheless, researchers can adopt a number of precautions to control this carelessness. They can use screening processes much like what is done for college samples. MTurk also allows researchers to reject incomplete responses which, again, is a precaution similarly used on student samples (Aruguete et al., 2019). MTurk also has established the idea of a “Worker reputation,” which is based on the ratio of HITs accepted and completed. Requesters can then verify the “reputation” of a Worker through this ratio (Sheehan & Pittman, 2016). One approach to examining the appropriateness of MTurk­based data is to examine the quality of the data. That is, are the responses reliable? Are they valid? And can the results be generalized to other populations? The answer to each question appears to be a cautious “yes,” but with caveats.

Reliability

One important question regarding the psychometric quality of MTurk­based data relates to the reliability of the responses. After all, if MTurk Workers respond to surveys less attentively and more haphazardly than traditional respondents, there would be an increase in the proportion of error variance in the measurement, which would be associated with lower reliability estimates. Although many studies have reported reliability estimates calculated for MTurk­based data sets, most of these studies simply make general statements regarding reliability such as “the reliability level was high.” To the best of our knowledge, the only study that has examined the statistical significance of the difference between reliability estimates obtained for MTurk samples and traditional samples was conducted by Rouse (2015), who compared coefficient alpha values obtained for MTurk samples with those obtained when scales were standardized with a normative sample. When best practices were followed (specifically, asking the Worker to indicate whether or not they were attentive to the task), the MTurk responses were no less reliable than those obtained for a community­based sample. However, Rouse (2020) found that MTurk Workers who had been granted “Master” status based on positive performance in the past (and

therefore eligible for higher paying HITs) were no more likely to produce reliable data than standard Workers, which suggests that researchers may not benefit by paying more for high­status Master Workers.

Validity

Many MTurk researchers use one of three methods to identify responses that might be invalid: (a) the overt acknowledgment that a participant has not been paying attention; (b) incorrect responses to attention checks, and (c) unrealistically speedy task completion. The literature suggests that it is important for a researcher to use one or more of these methods because most studies that use these methods have identified a significant number of Workers whose responses should be eliminated from analyses. For example, Kim and Hodgins (2017) eliminated 10% of their participants either because their responses were unrealistically speedy or because they did not indicate that they had been attentive to the task. However, when analyses were performed with the remaining sample, the correlations between scales were comparable to those obtained for traditionally collected data sets. Furthermore, the remaining respondents in that study indicated that they found it easier to be honest in answering sensitive questions on MTurk than in an interview, suggesting that the anonymity of the MTurk format might actually increase validity especially when addressing potentially stigmatizing topics such as addictive behaviors. Nevertheless, these results highlight the importance of including a method by which a researcher can screen out invalid responses (Aruguete et al., 2019).

Trends in the Psi Chi Journal of Psychological Research

The growing presence of MTurk is seen within the pages of the Psi Chi Journal of Psychological Research. Ten years ago, only one study used data collected from MTurk, whereas 59% used data from a more traditional source of data—undergraduate research participation pools. In 2021, the use of MTurk increased to 10%, while it has increased to 20% in 2022 with 15% of articles utilizing another online platform, Qualtrics Panel. The increased use in MTurk and other online platforms may be due to its growing ubiquity and may also be an artifact of the pandemic.

Conclusions

Efficient recruitment of research participants from large sampling resources such as online platforms and undergraduate research pools provides opportunities as well as challenges for psychology researchers. Online platforms expand the scope and efficiency of data collection while reducing financial and personnel costs.

SUMMER 2023

Use of Mechanical Turk Data | Lopez, Rouse, and Treadwell

Yet with these benefits, the challenges presented include the quality of data, payment rate for participants, and representativeness of particular populations. Careful consideration of the time and effort needed for a typical person in the target population to complete the study can recommend adequate compensation. Considering that consent and data privacy and confidentiality with the IRB can increase protection of human subjects, while maintaining current understandings of online behavior and community standards (Kraut et al., 2004). These considerations have not been typical of previous domains for human subject data collection and require thoughtful consideration and consultation when embarking with internet data collection.

References

American Psychological Association. (2017). Ethical principles of psychologists and code of conduct. https://www.apa.org/ethics/code/

Antin, J., & Shaw, A. D. (2012). Social desirability bias and self-reports of motivation: A study of Amazon Mechanical Turk in the US and India. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2925–2934. https://doi.org/10.1145/2207676.2208699

Aruguete, M. S., Huynh, H., Browne, B. L., Jurs, B., Flint, E., & McCutcheon, L. E. (2019). How serious is the “carelessness” problem on Mechanical Turk? International Journal of Social Research Methodology, 22(5), 441–449. https://doi.org/10.1080/13645579.2018.1563966

Buhrmester, M., Kwang, T., & Gosling, S. (2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6(1), 3–5. https://doi.org/10.1177/1745691610393980

Burnham, M. J., le, Y. K., & Piedmont, R. L. (2018). Who is MTurk? Personal characteristics and sample consistency of these online workers. Mental Health, Religion, & Culture, 21(9), 934–944. https://doi.org/10.1080/13674676.2018.1486394

Difallah, D., Filatova, E., & Ipeirotis, P. (2018, February 5). Demographics and dynamics of Mechanical Turk Workers. Proceedings of WSDM 2018: The Eleventh ACM International Conference on Web Search and Data Mining, Marina Del Rey, CA., USA, 135–143. https://doi.org/10.1145/3159652.3159661

Jensen-Doss, A., Patel, Z. S., Casline, E., Mora Ringle, V. A., & Timpano, K. R. (2022). Using Mechanical Turk to study parents and children: An examination of data quality and representativeness. Journal of Clinical Child and Adolescent Psychology, 51(4), 428–442.

https://doi.org/10.1080/15374416.2020.1815205

Kaufman, N., Schulze, T., & Veit, D. (2011). More than fun and money: Worker motivation in crowdsourcing—A study on Mechanical Turk. AMCIS 2011 Proceedings. Submission 340.

https://aisel.aisnet.org/amcis2011_submissions/340

Keith, M. G., Harms, P., & Tay, L. (2019). Mechanical Turk and the gig economy: Exploring differences between gig workers. Journal of Managerial

Psychology, 34(4), 286–306. https://doi.org/10.1108/JMP-06-2018-0228

Kim, H. S., & Hodgins, D. C. (2017). Reliability and validity of data obtained from alcohol, cannabis, and gambling populations on Amazon’s Mechanical Turk. Psychology of Addictive Behaviors, 31(1), 85–94. http://doi.org/10.1037/adb0000219

Kraut, R., Olson, J., Banaji, M., Bruckman, A., Cohen, J., & Couper, M. (2004). Psychological research online: Report of Board of Scientific Affairs’ Advisory Group on the Conduct of Research on the Internet. American Psychologist, 59(2), 105–117. https://doi.org/10.1037/0003-066X.59.2.105

McCredie, M. N., & Morey, L. C. (2019). Who are the Turkers? A characterization of MTurk Workers using the Personality Assessment Inventory. Assessment, 26(5), 759–766. https://doi.org/10.1177/1073191118760709

Merz, Z. C., Lace, J. W., & Eisenstein, A. M. (2022). Examining broad intellectual abilities obtained within an MTurk internet sample. Current Psychology, 41, 2241–2249. https://doi.org/10.1007/s12144-020-00741-0

Ogletree, A. M., & Katz, B. (2021). How do older adults recruited from MTurk differ from those in a national probability sample? International Journal of Aging and Human Development, 93(2), 700–721. https://doi.org/10.1177/0091415020940197

Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23(3), 184–188. https://doi.org/10.1177/0963721414531598

Prolific. (2022). Conduct your best research. Prolific. https://www.prolific.co/researchers#pricing

Rouse, S. V. (2015). A reliability analysis of Mechanical Turk data. Computers in Human Behavior, 45, 304–307. https://doi.org/10.1016/j.chb.2014.11.004

Rouse, S. V. (2020). Reliability of MTurk data from Masters and Workers. Journal of Individual Differences, 41(1), 30–36. https://doi.org/10.1027/1614-0001/a000300

Sheehan, K. B. (2018). Crowdsourcing research: Data collected within Amazon’s Mechanical Turk. Communication Monographs, 85(1), 140–156. https://doi.org/10.1080/03637751.2017.1342043

Sheehan, K. B., & Pittman, M. (2016). Amazon’s Mechanical Turk for Academics. Melvin & Leigh Publishers.

Standage, T. (2002). The Turk: The life and times of the famous eighteenth century chess-playing machine. Walker.

Vanhove, A. J., Miller, A. D., & Harms, P. D. (2021). Understanding subpopulations on Mechanical Turk: A comparison of three employment status groups. Journal of Personnel Psychology, 20(4), 176–186. https://doi.org/10.1027/1866-5888/a000281

Author Note. Stella G. Lopez https://orcid.org/0000­0003­1669­8485

Steven V. Rouse

http://orcid.org/0000­0002­1080­5502

Kimberli Treadwell

https://orcid.org/0000­0002­6595­4193

All authors contributed equally to this editorial. Correspondence concerning this editorial should be addressed to Stella G. Lopez, One UTSA Circle, Department of Psychology, University of Texas at San Antonio, San Antonio, Texas 78249. Email: stella.lopez@utsa.edu

Swiping Away Your Well-Being? Examining Well-Being Indicators Among TikTok Account Holders

Christian Nienstedt, Nathan Smith, Hannah Braithwaite, Ben Gilbert, and Robert R. Wright* Department of Psychology, Brigham Young University – Idaho

ABSTRACT. Social media platforms continue to increase in popularity and number. Although many associations between social media and user health interactions have been explored, little to no research has investigated TikTok, a recently released social media platform, and its individual interactions with user mental and physical well­being variables. To address this gap, we administered an online questionnaire to a group of college students (n = 407) at Brigham Young University – Idaho about their social media use, mental, and physical well­being. We observed potential differences among TikTok account holders and non­TikTok users using a series of independent­samples t tests with Bonferroni correction. Although TikTok account holders reported higher negative mental well­being indicators than their nonuser counterparts in almost all measured areas, these relationships were only consistent among female TikTok account holders. Conversely, our findings of significantly higher reported consumption of sugary drinks across our entire sample, t(403) = 3.41, p = .001, d = 0.33, only maintained significance among male participants, t(141) = 3.04, p < .001, d = 0.50. As such, female TikTok account holders reported consistently worse mental health and well­being indicators than non­TikTok users. Building on our findings, we call for future research to better understand the nature of these relationships for both TikTok and other social media users’.

Keywords: social media, TikTok, health and wellness, behavior

Within the last few decades, technology use has greatly increased among individuals and institutions all over the world. One of the most popular technologies available is social media, which offers individuals the opportunity to connect online, share content in various formats, and browse material posted by others (Shao 2009). Prior research has demonstrated how social media use across a wide range of social media platforms (e.g., Facebook, Instagram, Snapchat) is related to user health and well­being, with some notable differences (e.g., Liu et al., 2019; Wright et al., 2020; Wright et al., 2021). Although the utility and entertainment value of social media technology is high, current scientific literature suggests that a potential relationship may exist between deleterious health and well­being outcomes, and social media use, particularly with prolonged use or exposure among college students (e.g., Wright et al., 2023). The continued release and availability of popular social media platforms such as TikTok warrants continued research efforts. As such,

we review the current literature regarding connections between social media use, mental health, and physical health and then provide an overview of the research conducted to­date on TikTok below.

Social Media Use and Well-Being Indicators

First, within the extant literature, there have been several connections made between social media use and poor mental health among users. This list includes, but is not limited to, such notable well­being indicators as negative mood (Fardouly et al., 2015), poor body image (Eckler et al., 2017; Meier & Gray, 2014), loneliness (Reer et al., 2019; Wright et al., 2017), anxiety (Almarzouki et al., 2022; Reer et al., 2019), and depressive symptoms (Perlis et al., 2021; Reer et al., 2019). Additionally, studies have identified harmful mental health outcomes resulting from social comparison, or one’s tendency to compare oneself to others they see on social media (Stapelton et al., 2017; Vogel et al., 2014; Vogel et al., 2015). Moreover, it has been demonstrated that social media may facilitate

these deleterious social comparisons especially among individuals high in social comparison orientation, which then, in turn, may result in increased negative well­being outcomes such as increased negative mood, depressive symptoms and loneliness (Stapelton et al., 2017; Twenge et al., 2018; Vogel et al., 2015; Yang, 2016).

Related to this, negative mood and social comparison have been associated with exposure to unrealistic body images presented in social media especially among female users (Eckler et al., 2017; Meier & Gray, 2014). Interestingly, even taking short­term breaks from social media lasting as short as a week have led to significantly improved outcomes in these domains for individual well­being (Tromholt, 2016; Vanman et al., 2018). Although the strength of these relationships may be moderated by individual personality traits (Stead & Bibby, 2017; Vogel et al., 2015), these negative health indicators associated with social media use may be widespread as the overall increase and accessibility of social media and technology has positively correlated with rising rates of suicide and depression in the United States (Twenge et al., 2017). Additionally, addictions related to technology and social media have been associated with symptoms of other severe psychological disorders such as psychosis, obsessive­compulsive disorder (Huang, 2010) and maladaptive eating behaviors (Murray et al., 2016). Finally, other research has pointed to an observed association between elevated perceived loneliness and depressive symptomology with higher daily use of social media (Song et al., 2014; Wright et al., 2017), suggesting that prolonged social media use may have some unintended negative consequences for user mental health and well­being. As such, although causality remains difficult to firmly establish, the relationship between prolonged social media use and poor mental health indicators seems consistent and robust. Research on the effects of technology and social media on physical and behavioral health outcomes is less extensive, but suggests a similar relationship. Indeed, as individuals place higher value on and spend more time on social media, several physical and behavioral well­being indicators seem to decline (e.g., Dibb, 2019; Wright et al., 2021). For instance, increased social media and technology use has been associated with increased sugar and caffeine intake (Bradbury et al., 2019; Fomby et al., 2021; Wright et al., 2021), physical complaints (Wright et al., 2021), decreased sleep duration (Kelly et al., 2018; Reynolds et al., 2019) and overall poorer sleep quality among users (Fomby et al., 2021; Hamilton et al., 2020; Woods & Scott, 2016). This relationship is especially poignant when social media and technology usage is localized around user bedtime (Levenson et al., 2017; Reynolds et al., 2019).

Little research has examined the relationship between social media use and physical activity of users. However,

general technology usage and media consumption have been associated with deficits in physical exercise and increased sedentary behaviors (Singh et al., 2008; Tandon et al., 2012), suggesting a potential similar relationship. Increases in media consumption have been observed to take up time that could be spent engaging in physical exercise while also promoting sedentary behavior and unhealthy eating among children and adolescents (Cox et al., 2012; Rosen et al, 2014). Researchers have argued that the displacement of healthy behaviors, such as sleeping and exercise, are the true culprits of negative health outcomes rather than media usage itself (Huang, 2010).

Known as the displacement hypothesis, technology and media usage, such as watching television, playing videogames, or using social media, is thought to displace other activities that may better contribute to well­being (see Putnam, 1995). Traditionally, this framework has been applied to the displacement of quality social interactions (i.e., face­to­face conversations) by technology usage, affecting individual social well­being (Hall et al., 2019b; Liu et al., 2019; Valkenburg & Peter 2007). Recent investigations, however, have discovered that different forms of media and technology may vary in their potential displacement of activities related to social well­being (Liu et al., 2019). Although claims suggesting the displacement of quality social interactions by social media have been previously unsupported (Hall et al., 2019b), social media usage has been observed to displace day­to­day activities such as work and other technology usage (Hall et al., 2019a). Moreover, displacement of these activities by social media also positively correlated with poor mental health indicators such as negative mood (Hall et al., 2019a).

Many studies have examined observed trends of poor physical and mental health indicators between men and women and have identified discrepancies reported between the two (Alt, 2015; Dibb, 2019; Przybylski & Weinstein, 2017). Blomfield & Barber (2014) found higher levels of reported negative mental health indicators from women than men but neither frequency nor investment in social media usage were substantially different. Additionally, women have reported higher rates of psychosocial risk factors related to social pressures from media and social networks than men (Ata et al., 2007; Beyens et al., 2016). Poor body image related to social comparison, in particular, has been identified more strongly among female social media users than male users (Eckler et al., 2017; Meier & Gray, 2014). More recently, female social media users reported experiencing more intense feelings of stress and anxiety during the COVID19 pandemic when compared to male users (Hou et al., 2020). These findings suggest gender to be an important factor when considering potential relationships between social media usage and user health and well­being.

Nienstedt, Smith, Braithwaite, Gilbert, and Wright | TikTok and Well-Being

Although evidence supporting a relationship between social media usage and poor mental and physical well­being exist, not all research supports this claim. Prior longitudinal literature has failed to find any significant relationship between social media screen time and negative physical or mental health indicators (Coyne et al., 2020; Houghton et al., 2018). While most of the prior literature has claimed that higher amounts of screen time predict higher amounts of negative wellbeing among users (Tandon et al., 2012; Twenge et al., 2017; Twenge et al., 2018), longitudinal findings propose that potential relationships between social media usage and negative wellbeing may be more complex (Coyne et al. 2020; Houghton et al., 2018). Moderate amounts of social media use has been linked to positive outcomes such as increased social participation in extracurricular activities among adolescents (Blomfield & Barber, 2014; Romer et al., 2013). Indeed, the “Goldilocks hypothesis” proposed by Przybylski & Weinstein (2017) suggests that negative well­being is associated with extreme amounts of social media use, however, moderate amounts of social media usage yield positive user well­being outcomes (Przybylski & Weinstein, 2017). Berryman et al. (2018) proposes that social media may serve more as an outlet for poor mental well­being rather than acting as the cause of negative well­being. These observed discrepancies within the literature suggest that potential relationships between social media usage and user well­being are complicated rather than straightforward.

Along these lines, another area of research has uncovered potential differential impacts of social media platforms on user health and well­being. Indeed, different types of social media have been shown to be related differentially to user health and well­being indicators (Frison & Eggemont, 2017; Limniou et al., 2021; Masciantonio et al., 2021; Perlis et al., 2021; Wright et al., 2020; 2021). For instance, two studies, Wright et al. (2020; 2021) reported that users of image­based platforms, such as Snapchat, had poorer health and well­being compared to those who did not use these platforms. Furthermore, those who use video­based or more professional platforms, such as MarcoPolo or LinkedIn, had better well­being profiles. However, these studies were unable to include TikTok, a newer social media platform, in their analyses. Moreover, although the precise mechanism behind these differential observations is not clear, these findings suggest a focused examination of specific forms of social media and their relationship with user health and well­being are necessary to understand how the use of a particular social media platform may be related to health and well­being indicators.

TikTok

TikTok is a social media platform that was originally named Musical.ly, founded in September 2016, and then later renamed to TikTok when it was acquired by Beijing Bytedance Technology (Vaterlaus & Winter, 2021). TikTok is an application that enables individuals to create short videos to share with others and perform playback videos to different songs that are distributed on the platform and can then be recreated by other users as well. As of December 2021, TikTok had 1.2 billion active monthly users worldwide, the United States being the second most active region with 105 million registered users (Iqbal, 2022). As such, TikTok is a popular social media platform.

Little research exists currently examining TikTok as an individual social media platform. The few published studies that do examine TikTok primarily focus on user motivations for engaging with TikTok as a platform (e.g., Bossen & Kottasz, 2020; Montag et al., 2021). Individual motives for TikTok usage have primarily been studied through the Uses and Gratifications framework, which suggests that individuals use specific platforms of social media to gratify specific needs or desires (Pelletier et al., 2020). Analyses summarized by Montag et al. (2021) of motivations for TikTok usage have identified self­expression, entertainment, and affective management as the primary uses and gratifications fulfilled by TikTok participation and consumption. Bossen & Kottasz (2020) found entertainment to be the primary motivation of TikTok consumption (e.g., scrolling and viewing), and searching for, and engaging with new social networks motivated TikTok users to create content on the platform. Perlis et al. (2021) analyzed the relationships between depressive symptoms and social media usage on various platforms and found that TikTok usage predicted higher rates of reported depressive symptoms in US adults 35 and over, coinciding with findings from studies examining usage of other social media platforms (Wright et al., 2017). However, with the relative lack of studies, little is known about the potential health and well­being trends among individuals with a TikTok account. Moreover, the popularity of TikTok and the previous literature’s identification of differential associations between health and wellbeing indicators unique to specific social media platforms warrants further in­depth study surrounding TikTok.

Current Study

In the current study, we measured well­being indicators among college student individuals who have a TikTok account compared to individuals who do not have a TikTok account in order to elucidate potential trends in well­being of TikTok account holders. First, consistent with general findings in the literature, we expected TikTok account holders to report poorer

mental well­being (i.e., mood, body image, stress, anxiety, loneliness, depressive symptoms, interpersonal conflict and self­regulation) relative to those who do not use TikTok. Second, we suspected, in accordance with the displacement hypothesis, that TikTok account holders may exhibit worse physical well­being due to a potential increase in displaced time being spent on social media platforms such as TikTok rather than engaging in health behaviors such as diet, exercise and sleep. Third, we examined reported differences of mental and physical well­being indicators between TikTok account holders and non­ TikTok users between men and women. In line with previous research, we expected to see more significant differences between female TikTok account holders and non­TikTok users than male TikTok account holders and non­TikTok users.

TABLE 1

Demographic Information

Methods

Participants and Procedure

After acquiring institutional review board approval, we surveyed a convenience sample regarding their social media practices and well­being. The sample was comprised of students taking a general psychology course at Brigham Young University – Idaho who were offered class credit for their volunteered participation in our survey. Data was collected via online survey in March and July 2021. Although we received 433 survey responses, we removed 26 responses due to participants reporting their age below 18 years old, or denying permission to use their data for publication purposes, which brought our final total sample size to 407. Participants had an average age of 25.68 (SD = 10.17), consisted of women (n = 263; 63.8%), and men (n = 141; 35.6%) and had an average credit enrollment of 10.82 (SD = 3.73) credits. Ethnicity included 78% White, 8% Hispanic, 5% Asian American, 2% Black or African American, and the remaining 7% were more than one race or other. Our sample was comprised of 44% first­year students, 26% sophomores, 18% juniors, and 10% seniors. Forty­nine percent were not in a committed relationship and 31% were married. The majority of our sample was reportedly unemployed (n = 173; 42%), though 39% held part­time jobs, and 17% held full­time jobs (see Table 1).

Measures

Although order effects were not expected, the questionnaire contained measures that were presented in this order: general health, health behaviors, physical health, emotional health, cognitive health, social health, technology use, personality, and demographics. Demographics were collected purposefully at the end to encourage full completion of the questionnaire.

Social Media Use and Demographic Information

We assessed the average number of hours spent on all social media platforms per day over the last 30 days using a sliding scale ranging from 0 to 10 hours. Participants also indicated the number of social media platforms they used. Holding a TikTok account was measured by a single dichotomous question asking participants to indicate if they had a TikTok account. Participants provided their sex, ethnicity, relationship status, employment status, and current level of college education along with age and current credits enrolled. Biological sex, specifically, was assessed using a single dichotomous variable in which participants were asked by the following prompt, “What is your biological sex?” with the dichotomous choice of Male or Female.

Mental Well-Being

Mental well­being was assessed using several measures.

Nienstedt, Smith, Braithwaite, Gilbert, and Wright | TikTok and Well-Being

Participants responded to an 8 ­ item scale assessing general mood where participants chose among a 5­point scale between 1 (not at all) and 5 (extremely) how much particular positive (α = .66) and negative (α = .62) mood descriptions agreed with their own general mood over the last 30 days (Wright et al., 2017). Positive affect was assessed by the adjectives enthusiastic, happy, relaxed, and alert whereas negative affect was captured by sad, nervous, irritable, and bored. Although previous studies using these scales have noted similar low alpha levels (e.g., Wright et al., 2016, 2017), the low alpha level for both positive and negative affect is likely attributable to the low number of items used to measure each construct while still existing within the same scale (see Tavakol & Dennick, 2011). Moreover, these mood descriptions may also represent extremes of emotion on two different axes (e.g., pleasantness, arousal), which could play a role (see Russell, 1980).

The 13­item Body Appreciation Scale was used to assess participant perceived body image; responses ranged from a 7­point truth scale from 1 (not at all true) to 7 (very true; Avalos et al., 2005; α = .94). For example, “Please indicate how true each is for you: My feelings for my body are positive for the most part.” Participants reported their perceived stress over the past 90 days through 7 items on the Perceived Stress Scale reporting on a 5­point frequency scale ranging from 1 (never) to 5 (very often; Cohen et al., 1983; α = .88). For example, “Please indicate how often you felt or thought a certain way about your life in general during the past month: Felt that you were unable to control the important things in your life?” Anxiety was measured over the past 90 days through a 4­item 5­point frequency measure from 1(never) to 5 (very often; Butz & Yogeeswaran, 2011; α = .85). For example: “In the past 3 months, how often have you been anxious?”

The Short Loneliness Scale provided 3­items which assessed frequency of feelings of loneliness among participants using a 5­point scale between 1 ( never) and 5 (all the time; Hughes et al., 2004; α = .90). For example, “How often do you feel isolated from others?” Participants reported occurrence of depressive symptoms over the last 30 days on a 4­point scale from 1 (rarely or none of the time) to 4 (most or all of the time; α = .73) using the 5 items on the Center for Epidemiological Studies Depression Scale­5 (Bohannon et al., 2003). For example, “Please indicate how you have felt during the past 4 weeks: I felt depressed.”

Interpersonal conflict was measured using a 6­item frequency scale; each item a 5­point scale ranging from 1 (never) to 5 (very often; Wright et al., 2017; α = .90). For example, “In the past 3 months, how often have you: Had a disagreement with other people over the work that you do?” Self ­ regulation was analyzed through

10­items on the Self­Regulation Scale on a 4­point scale ranging from 1 (not at all) to 4 (completely true; Diehl et al., 2010; α = .82). For example, “Please indicate how accurately each of the following statements describe you: It is difficult for me to suppress thoughts that interfere with what I need to do.”

Physical Well-Being

Physical well­being was also measured across several dimensions. A single item from the EuroQol­ 5 Dimension (EQ­5D) was used to measure perceived physical wellbeing among participants using a rating from 0 (worst physical health) to 100 (best physical health; Kind et al., 2005). Physical symptoms occurring over the past 30 days were measured through the 18­item Physical Symptom Inventory (Spector & Jex, 1998), which was administered as a simple dichotomous checklist (yes or no).

Dietary intake of health food servings (e.g., fruits and vegetables) and unhealthy foods (e.g., sugary snacks, sugary drinks, fast food) were measured using a 10­point scale indicating the number of servings consumed over the past month ranging from 0 (never) to 10 (5 or more servings per day) using one item each (Buxton et al., 2009). Participants were informed about what qualifies as a serving of fruits and vegetables as recommended by the U.S Department of Agriculture (U.S. Department of Agriculture, 2020). Water consumption was measured using a slider scale between 1 and 12 indicating how many 10oz glasses of water participants consumed on each day of the week (Wright et al., 2016).

To assess average weekly aerobic exercise, 5 items from the Stanford Patient Education Research Center measure were used to assess participant aerobic exercise over the last 30 days (Lorig et al., 1996). For example, “In a typical week during the last 30 days, how much total time for the entire week did you spend on each of the following: bicycling (including stationary exercise bikes)?” Sedentary behavior during the past month was measured using the 10­item Sedentary Behavior Questionnaire (Rosenberg et al., 2010) which queries length of time spent being sedentary on a 9­point scale ranging from 1 (none) to 9 (6 or more hours). For example: “On a typical weekday, how much time do you spend (from when you wake up until when you go to bed) doing the following: Watching television (including TV shows, movies on DVD)?” Participants reported average hours spent sleeping per night over the past 30 days and reported average sleep quality on a 5­point scale ranging from 1 (very poor) to 5 (very good; Buysse et al., 1998).

Data Analysis

We examined our data by following the methods of previous studies with similar objectives to our study (e.g., Wright et al., 2020; 2021). First, we investigated descriptive

Nienstedt, Smith, Braithwaite, Gilbert, and Wright | TikTok and Well-Being

statistics of those who held a TikTok account in order to describe a typical college student TikTok account holder. Second, using a series of independent­samples t tests, we examined the data for systematic differences in both mental and physical well­being variables between those with a TikTok account and non­TikTok users. Furthermore, due to the concern for inflated risk of Type I error while running multiple t tests, we applied a Bonferroni correction to all analyses. This adjusts the alpha level to p = .005 for our analyses of mental well­being indicators, and p = .004 for our analyses of

TABLE 2

Mental Well-Being Among TikTok Account Holders and Non-TikTok Users

physical well­being indicators based on the number of independent t tests conducted. Additionally, we analyzed the strength of any observed statistical significance using Cohen’s d effect sizes. Finally, we explored sex as another influential variable on the potential relationship between holding a TikTok account and our mental and physical well­being variables.

Results

Across the entire sample, participants recorded their total time spent on all social media platforms at an average of 182 minutes (SD = 170.71) and reported an average total number of social media platforms of 4.21 (SD = 1.78). Thirty­six percent of respondents (n = 145) indicated they had a current TikTok account. TikTok account holders were significantly younger (M = 23.12, SD = 6.61) than those who did not have a TikTok account ( M = 27.23, SD = 11.37), t (402) = 4.60, p < .001. Moreover, TikTok account holders reported using significantly more social media platforms (M = 5.26, SD = 1.65) than their non ­ TikTok counterparts (M = 3.64, SD = 1.56), t(400) = 9.76, p < .001 (see Table 1). Finally, our sample seemed relatively healthy with an average reported score of 78.38 (SD = 14.65) out of 100 on their subjective physical health.

Mental and Physical Well-Being

Physical Well-Being Among TikTok and Non-TikTok Users

Comparative analysis of mental well­being outcomes among TikTok account holders versus non­TikTok users yielded several significant differences. Across nearly all the mental well­being outcomes, TikTok account holders reported significantly worse mental well­being including higher amounts of negative mood, perceived stress, anxiety, interpersonal conflict, and depressive symptoms along with lower levels of self­regulation compared to non­TikTok users (see Table 2). Interestingly, the largest effect size was observed among perceived stress, suggesting that TikTok account holders were reporting significantly (p < .001) and substantially (d = 0.47) greater stress than their non­TikTok user counterparts. Although TikTok account holders reported a lower average of positive mood (M = 3.27, SD = 0.67) than non­TikTok users (M = 3.37, SD = 0.66) the difference did not reach significance between the two groups, t (403) = –1.39, p = .16, d = 0.15. Thus, those who had a TikTok account reported significantly worse mental well­being than those without a TikTok account across the variables we assessed.

Note. * p < .05 with Bonferroni correction; ∆ refers to the difference between TikTok and non-TikTok Users

Comparing TikTok account holders to non­TikTok users yielded significant differences in only one of the twelve physical well­being indicators, namely sugary drink consumption (see Table 3). Although TikTok account holders reported consuming significantly more sugary drinks than non­TikTok users, t(403) = 3.41,

TikTok and Well-Being | Nienstedt, Smith, Braithwaite, Gilbert, and Wright

p = < .001, d = 0.33, no significant relationship was found in the difference of consuming sugary snacks, fast food or any other diet related health outcome. No significant differences emerged between the two groups in terms of physical activity, physical health complaints, or overall subjective health.

Sex Differences and TikTok

Finally, we explored the potential relationship between these well ­being outcomes and biological sex. Most notably is the number of significant differences in reported well­being outcomes among female respondents, which is much higher than those of male respondents (see Tables 4 and 5). In terms of our mental well­being variables, female TikTok account holders reported significantly poorer mental health in almost all measured variables compared to nonTikTok users (see Table 4). All significant differences in mental well ­ being indicators disappeared when examining only male participants. Interestingly, Sugary drink consumption was only found to be significantly different among male TikTok account holders, t (141) = 3.04, p < .001, d = 0.50, while no physical well­being indicators showed any significant differences among female TikTok account holders and non­TikTok users (see Tables 4 and 5).

Discussion

Social media, with all its many benefits to consumers, has become a form of technology that is commonly used. However, recent studies have discovered potential links between increased use of this technology and many health and well­being indicators (e.g., Wright et al., 2023), giving rise to some concerns regarding how social media use may influence well­being. Investigating this relationship between well­being and TikTok, a new and popular social media platform, our study observed several differences between those who have a TikTok account and those who do not. TikTok account holders reported worse mental well­being in several areas (i.e., stress, depressive symptoms, body image, negative mood) and with only one observed physical health deficit (i.e., higher sugary drink intake) relative to their non­TikTok user counterparts. These findings surrounding mental health, however, were only consistent among female TikTok account holders, who reported worse mental health outcomes in almost all measured variables comparative to non­TikTok users, whereas no significant differences were observed among male participants. Notably, we found many nonsignificant differences between the two groups within each sex, especially among physical well­being indicators related to sleep, diet, and exercise with the exception of sugary drink

intake, which was only significantly different among male participants. Overall, these findings highlight an area of social media exploration that has had little prior investigation and points to directions for future research regarding TikTok as an individual social media platform. First, participants with a TikTok account reported significantly worse mental health in nearly all variables examined. Even more interesting is the loss of all significant differences in these reported indicators when examining male participants (which will be discussed in more detail below). Although further investigation is needed to determine the nature of the relationship between TikTok and these various indicators (e.g., causation), our findings are consistent with that of previous studies noting poorer comparative mental well­being among those who use social media, in general, and those who use social media for longer durations of time (e.g.,

TABLE 4

Reported Well-Being Among Women

Note. * p < .05, with Bonferroni Correction; ∆ refers to the difference between TikTok and non-TikTok Users

Nienstedt, Smith, Braithwaite, Gilbert, and Wright | TikTok and Well-Being

Almarzouki et al., 2022; Fardouly et al., 2015; Meier & Gray, 2014; Reer et al., 2019; Wright et al., 2017).

The usage and structure of the TikTok platform may be an important factor to consider. Although TikTok is primarily a video ­ based platform rather than an image­based platform, possible underlying similarities between platforms may further elucidate the relationship between particular social media platforms, such as TikTok, and user mental well­being. Indeed, Wright et al. (2021) observed differences in user health profiles based on usage of image versus video­based platforms suggesting that user interaction with platform modalities rather than specific platform brands may be a mediating factor in social media’s relationship to user mental wellbeing. However, in that study, the video­based platform of Marco Polo seemed to be associated with better rather than poorer health and well­being profiles as we

5

Reported Well-Being Among Men

observed in the current study with TikTok. Although specific modalities may be important, it may be that the motive for using these particular social media platforms has a stronger moderating effect (e.g., Yang, 2016). For instance, TikTok is often used for pleasure, recreation, and entertainment (Bossen & Kottasz, 2020; Montag et al., 2021) whereas Marco Polo is regularly used to stay socially connected with family and friends. Ultimately, however, it remains unclear the exact causal mechanism behind this observation.

Second, physical and behavioral well­being differences between TikTok account holders and nonTikTok users, though few, are noteworthy. As previously observed by Bradbury et al. (2019), technology usage has been successful in predicting an increase in sugary drink and caffeine intake. The several nonsignificant differences between TikTok account holders and non­TikTok users are important to note, suggesting that TikTok may not be related to these well­being indicators. In any case, it seems that TikTok use may not be strongly related to diet, physical activity, or physical health indicators.

Finally, the differences in reported deleterious well­being outcomes between women and men is an especially important observation. Several differences in mental well­being indicators among female TikTok account holders were observed, but none were observed among males. These findings are reminiscent of previous literature which has consistently observed female social media users reporting more intense and frequent negative health and well ­ being outcomes than male counterparts (Beyens et al., 2016; Blomfield & Barber, 2014; Dibb, 2019; Hou et al., 2020). Moreover, these results may extend into maladaptive behaviors that differ between the sexes such as maladaptive eating behaviors which have been associated with media use among women (Alta et al., 2007), although no significant dietrelated differences were observed among female TikTok account holders within our sample.

Note. * p < .05, with Bonferroni Correction; ∆ refers to the difference between TikTok and non-TikTok Users

The similarity in findings between the present study and prior literature regarding female account holders may indicate that TikTok resembles other platforms in potential effects on user mental well­being. For instance, reports of increased poor body image by female TikTok account holders may reflect a higher exposure to unrealistic body images that facilitate social comparison as seen in prior studies (Eckler et al., 2017; Meier & Gray, 2014) especially among women (Ata et al., 2007; Beyens et al., 2016). As such, potential higher social comparison among female TikTok account holders may be the culprit for other reported deficits in mental well­being indicators compared to non­TikTok account holders, such as reported levels of increased negative mood, perceived stress, anxiety, and depressive symptoms (Stapleton et

TABLE

al., 2017; Vogel et al., 2015; Yang et al., 2016).

Also notable was the significant difference in sugary drink intake observed only among male participants, while no differences in physical well­being indicators were observed among female participants whatsoever. Although some research has found relationships between the consumption of sugary and caffeinated drinks, and screen time activities such as video game and social media usage (Bradbury et al., 2019; Larson et al., 2014) no major discrepancies have been found between sexes. Furthermore, more research into the types of sugary drinks consumed by TikTok account holders and social media users in general is needed to better understand this observation and the potential relationship between sugary drink consumption and social media engagement (Bradbury et al., 2019).

Limitations and Future Research

Several potential limitations exist within the present study. Most notable are the self ­ report nature and cross­sectional design of the study, which prevents any causal conclusions from being drawn. Additionally, we did not measure the amount of TikTok screen time nor that of any specific social media platform; groups were separated based on TikTok account status. Moreover, participants within our sample were college students, predominantly White, and mostly women which may limit the generalizability of these findings to individuals of whom these descriptors do not apply. The assessment of biological sex through a single dichotomous variable using “male” and “female” should also be noted as a potential limitation as these findings may not be generalizable to transgender, intersex, or non­binary individuals. Furthermore, characteristics of other universities and their respective student pools may further limit generalizability of these findings to populations that might otherwise be similar. Other characteristics, such as the average age difference between TikTok account holders and non ­ TikTok users, may have influenced our findings. Additionally, we were unable to independently control for use of other social media platforms, which may have inadvertently biased the results. Not all measures involved in the study measured their respective health indicators across the same time frame which may limit precision in the interpretations of findings within the study. Survey fatigue, due to the extensive measures involved in the study, is another possibility, which may potentially bias participant responses. However, the rather large sample size mitigates this concern somewhat. The use of multiple t tests raises concern for Type 1 error, though we adjust for this risk using a Bonferroni correction based on the number of t tests conducted. Finally, although our study did examine differences in health outcomes among women

and men, further analysis is needed to directly compare those outcomes. Despite these shortcomings, this study makes important contributions as it is one of the first examinations of the relationship between TikTok use and user health and well­being.

The widespread popularity of TikTok and potential trends in user mental and physical health reported in this analysis warrant further study into use of the social media platform itself. Further investigations measuring screen time on the TikTok platform and more in­depth TikTok use (such as consuming versus creating content) may elucidate potential relationships between the app and user well­being. Particularly, comparisons between user health outcomes associated with TikTok use and other platforms of similar and different modalities (i.e., video versus image­based user interactions) may provide better understanding of social media usage and user health, and the effects of individual social media platforms, such as TikTok, on user health outcomes. Controlling for factors such as age, further comparisons of mental health indicators between sexes regarding TikTok users may also point out potential trends and relationships unique to TikTok as a platform. The relationship between number of social media platforms reported by respondents with a TikTok account versus non­TikTok users also provides an interesting avenue for further study. Investigations into whether or not making an account for the TikTok platform precedes or follows subscription to other social media platforms, especially among younger users, may be insightful. In conclusion, although these findings suggest that TikTok account holders are generally not as healthy as their counterparts, further research is needed to ascertain the causal mechanisms and implications for future recommendations of social media use.

References

Almarzouki, A. F., Alghamdi, R. A., Nassar, R., Aljohani, R. R., Nasser, A., Bawadood, M., & Almalki, R. H. (2022). Social media usage, working memory, and depression: An experimental investigation among university students. Behavioral Sciences, 12(1), 16. https://doi.org/10.3390/bs12010016

Alt, D. (2015). College students’ academic motivation, media engagement and fear of missing out. Computers in Human Behavior, 49, 111–119. https://doi.org/10.1016/j.chb.2015.02.057

Ata, R. N., Ludden, A. B., & Lally, M. M. (2007). The effects of gender and family, friend, and media influences on eating behaviors and body image during adolescence. Journal of Youth and Adolescence, 36(8), 1024–1037. https://doi.org/10.1007/s10964-006-9159-x

Avalos, L., Tylka, T. L., & Wood-Barcalow, N. (2005). The Body Appreciation Scale: Development and psychometric evaluation. Body Image, 2(3), 285–297. https://doi.org/10.1016/j.bodyim.2005.06.002

Berryman, C., Ferguson, C. J., & Negy, C. (2018). Social media use and mental health among young adults. Psychiatric Quarterly, 89(2), 307–314. https://doi.org/10.1007/s11126-017-9535-6

Beyens, I., Frison, E., & Eggermont, S. (2016) “I don’t want to miss a thing”: Adolescents’ fear of missing out and its relationship to adolescents’ social needs, Facebook use, and Facebook related stress. Computers in Human Behavior 64, 1–8. https://doi.org/10.1016/j.chb.2016.05.083

Blomfield Neira, C. J., & Barber, B. L. (2014). Social networking site use: Linked

SUMMER 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

Nienstedt, Smith,

to adolescents' social self‐concept, self‐esteem, and depressed mood. Australian Journal of Psychology, 66(1), 56–64.

https://doi.org/10.1111/ajpy.12034

Bohannon, R. W., Maljanian, R., & Goethe, J. (2003). Screening for depression in clinical practice: Reliability and validity of a five item subset of the CES depression. Perceptual and Motor Skills, 97(3), 855–861.

https://doi.org/10.2466/pms.97.7.855-861

Bossen, C. B., & Kottasz, R. (2020). Uses and gratifications sought by preadolescent and adolescent TikTok consumers. Young Consumers, 21(4). https://doi.org/10.1108/YC-07-2020-1186

Bradbury, K. M., Turel, O., & Morrison, K. M. (2019). Electronic device use and beverage related sugar and caffeine intake in US adolescents. PLoS ONE, 14(10), 1–18.

https://doi.org/10.1371/journal.pone.0223912

Butz, D. A., & Yogeeswaran, K. (2011). A new threat in the air: Macroeconomic threat increases prejudice against Asian Americans. Journal of Experimental Social Psychology, 47(1), 22–27. https://doi.org/10.1016/j.jesp.2010.07.014

Buxton, O. M., Quintilliani, L. M., Yang, M. H., Ebbeling, C. B., Stoddard, A. M., Pereira, L. K., & Sorensen, G. (2009). Association of sleep adequacy with more healthful food choices and positive workplace experiences among motor freight workers. American Journal of Public Health, 99(S3). https://doi.org/10.2105/AJPH.2008.158501

Buysse, D. J., Reynolds C. F. III, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1998). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193–213. https://doi.org/10.1016/0165-1781(89)90047-4

Cohen, S., Kamarck, T., Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396.

Cox, R., Skouteris, H., Rutherford, L., Fuller‐Tyszkiewicz, M., Dell'Aquila, D., & Hardy, L. L. (2012). Television viewing, television content, food intake, physical activity and body mass index: A cross‐sectional study of preschool children aged 2–6 years. Health Promotion Journal of Australia, 23(1), 58–62. https://doi.org/10.1071/HE12058

Coyne, S. M., Rogers, A. A., Zurcher, J. D., Stockdale, L., & Booth, M. (2020). Does time spent using social media impact mental health?: An eight year longitudinal study. Computers in Human Behavior, 104, 106160.

https://doi.org/10.1016/j.chb.2019.106160

Dibb, B. (2019). Social media use and perceptions of physical health. Heliyon, 5(1). https://doi.org/10.1016/j.heliyon.2018.e00989

Diehl, M., Semegon, A. B., & Schwarzer, R. (2010). Assessing attention control in goal pursuit: A component of dispositional self-regulation. Journal of Personality Assessment, 86(3), 306–317.

https://doi.org/10.1207/s15327752jpa8603_06

Eckler, P., Kalyango, Y., & Paasch, E. (2017). Facebook use and negative body image among US college women. Women & Health, 57(2), 249–267. https://doi.org/10.1080/03630242.2016.1159268

Fardouly, J., Diedrichs, P. C., Vartanian, L. R., and Halliwell, E. (2015). Social comparisons on social media: The impact of Facebook on young women’s body image concerns and mood. Body Image, 13, 38–45.

https://doi.org/10.1016/j.bodyim.2014.12.002

Fomby, P., Goode, J. A., Truong-Vu, K.-P., & Mollborn, S. (2021). Adolescent technology, sleep, and physical activity time in two U.S. cohorts. Youth & Society, 53(4), 585–609, https://doi.org/10.1177/0044118X19868365

Frison, E., & Eggemont, S. (2017). Browsing, posting, and liking on Instagram: The reciprocal relationships between different types of Instagram use and adolescents’ depressed mood. Cyberpsychology, Behavior, and Social Networking, 20(10), 603–609. https://doi.org/10.1089/cyber.2017.0156

Hall, J. A., Johnson, R. M., & Ross, E. M. (2019a). Where does the time go? An experimental test of what social media displaces and displaced activities’ associations with affective well-being and quality of day. New Media & Society, 21(3), 674–692. https://doi.org/10.1177/1461444818804775

Hall, J. A., Kearney, M. W., & Xing, C. (2019b). Two tests of social displacement through social media use. Information, Communication & Society, 22(10), 1396–1413. https://doi.org/10.1080/1369118X.2018.1430162

Hamilton, J. L., Chand, S., Reinhardt, L., Ladouceur, C. D., Silk, J. S., Moreno, M., Franzen, P. L., & Bylsma, L. M. (2020). Social media use predicts later sleep timing and greater sleep variability: An ecological momentary assessment study of youth at high and low familial risk for depression. Journal of Adolescence, 83, 122–130. https://doi.org/10.1016/j.adolescence.2020.07.009

Hou, F., Bi, F., Jiao, R., Luo, D., & Song, K. (2020). Gender differences of depression and anxiety among social media users during the COVID-19 outbreak in

China: A cross-sectional study. BMC Public Health, 20(1), 1–11. https://doi.org/10.1186/s12889-020-09738-7

Houghton, S., Lawrence, D., Hunter, S. C., Rosenberg, M., Zadow, C., Wood, L., & Shilton, T. (2018). Reciprocal relationships between trajectories of depressive symptoms and screen media use during adolescence. Journal of Youth and Adolescence, 47(11), 2453–2467. https://doi.org/10.1007/s10964018-0901-y

Huang, C. (2010). Internet use and psychological well-being: A meta-analysis. Cyberpsychology, Behavior, and Social Networking, 13(4), 241–249. https://doi.org/10.1089/cyber.2009.0217

Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). A short scale for measuring loneliness in large surveys: Results from two populationbased studies. Research on Aging, 26(6), 655–672. https://doi.org/10.1177/0164027504268574

Iqbal, M. (2022, April 20) TikTok Revenue and Usage Statistics (2022). businessofapps. https://www.businessofapps.com/data/tik-tok-statistics/ Kelly, Y., Zilanawala, A., Booker, C., & Sacker, A. (2018). Social media use and adolescent mental health: Findings from the UK Millennium Cohort Study. EClinicalMedicine, 6, 59–68. https://doi.org/10.1016/j.eclinm.2018.12.005

Kind, P., Brooks, R., & Rabin, R. (2005). EQ-5D concepts and methods. A Developmental History.

https://link.springer.com/content/pdf/10.1007/1-4020-3712-0.pdf

Larson, N., DeWolfe, J., Story, M., & Neumark-Sztainer, D. (2014). Adolescent consumption of sports and energy drinks: linkages to higher physical activity, unhealthy beverage patterns, cigarette smoking, and screen media use. Journal of Nutrition Education and Behavior, 46(3), 181–187. https://doi.org/10.1016/j.jneb.2014.02.008

Levenson, J. C., Shensa, A., Sidani, J. E., Colditz, J. B., & Primack, B. A. (2017). Social media use before bed and sleep disturbance among young adults in the United States: A nationally representative study. Sleep, 40(9). https://doi.org/10.1093/sleep/zsx113

Limniou, M., Ascroft, Y., & McLean, S. (2021). Differences between Facebook and Instagram usage in regard to problematic use and well-being. Journal of Technology in Behavioral Science, 1–10. https://doi.org/10.1007/s41347-021-00229-z

Liu, D., Baumeister, R. F., Yang, C. C., & Hu, B. (2019). Digital communication media use and psychological well-being: A meta-analysis. Journal of Computer-Mediated Communication, 24(5), 259–273. https://doi.org/10.1093/jcmc/zmz013

Lorig, K., Stewart, A., Ritter, P., González, V., Laurent, D., & Lynch, J. (1996). Outcome measures for health education and other health care interventions. Sage Publications, 25, 37–38.

Masciantonio, A., Bourguignon, D., Bouchat, P., Balty, M., & Rimé, B. (2021). Don’t put all social network sites in one basket: Facebook, Instagram, Twitter, TikTok, and their relations with well-being during the COVID-19 pandemic. PLoS ONE, 16(3), 1–14. https://doi.org/10.1371/journal.pone.0248384

Meier, E. P., & Gray, J. (2014). Facebook photo activity associated with body image disturbance in adolescent girls. Cyberpsychology Behavior and Social Networking, 17, 199–206. https://doi,org/10.1089/cyber.2013.0305

Montag, C., Yang, H., & Elhai, J. D. (2021). On the psychology of TikTok use: A first glimpse from empirical findings. Frontiers in Public Health, 9(62). https://doi.org/10.3389/fpubh.2021.641673

Murray, M., Maras, D., & Goldfield, G. S. (2016). Excessive time on social networking sites and disordered eating behaviors among undergraduate students: Appearance and weight esteem as mediating pathways. Cyberpsychology Behavior and Social Network, 19, 709–715.

https://doi.org/10.1089/cyber.2016.0384

Pelletier, M. J., Krallman, A., Adams, F. G., & Hancock, T. (2020). One size doesn’t fit all: A uses and gratifications analysis of social media platforms. Journal of Research in Interactive Marketing, 14(2), 269–284. https://doi.org/10.1108/JRIM-10-2019-0159

Perlis, R. H., Green, J., Simonson, M., Ognyanova, K., Santillana, M., Lin, J., Quintana, A., Chwe, H., Druckman, J., Lazer, D., Baum, M. A., & Della Volpe, J. (2021). Association between social media use and self-reported symptoms of depression in US adults. JAMA Network Open, 4(11). https://doi.org/10.1001/jamanetworkopen.2021.36113

Przybylski, A. K., & Weinstein, N. (2017). A large-scale test of the goldilocks hypothesis: Quantifying the relations between digital-screen use and the mental well-being of adolescents Psychological Science, 28(2), 204–215. https://doi.org/10.1177/0956797616678438

Putnam, R. (1995). Tuning in, tuning out: The strange disappearance of social capital in America. PS: Political Science & Politics, 28(4), 664–683. https://doi.org/10.2307/420517

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

Braithwaite, Gilbert, and Wright | TikTok and Well-Being

TikTok and Well-Being | Nienstedt, Smith, Braithwaite, Gilbert, and Wright

Reer, F., Tang, W. Y., and Quandt, T. (2019). Psychosocial well-being and social media engagement: The mediating roles of social comparison orientation and fear of missing out. New Media & Society, 21, 1486–1505. https://doi.org/10.1177/1461444818823719

Reynolds, A. C., Meltzer, L. J., Dorrian, J., Centofanti, S. A., & Biggs, S. N. (2019). Impact of high-frequency email and instant messaging (E/IM) interactions during the hour before bed on self-reported sleep duration and sufficiency in female Australian children and adolescents. Sleep Health, 5(1), 64–67. https://doi.org/10.1016/j.sleh.2018.10.008

Romer, D., Bagdasarov, Z., & More, E. (2013). Older versus newer media and the well-being of United States youth: Results from a national longitudinal panel. Journal of Adolescent Health, 52(5), 613–619. https://doi.org/10.1016/j.jadohealth.2012.11.012

Rosen, L. D., Lim, A. F., Felt, J., Carrier, L. M., Cheever, N. A., Lara-Ruiz, J. M., Mendoza, J. M., & Rokkum, J. (2014). Media and technology use predicts ill-being among children, preteens and teenagers independent of the negative health impacts of exercise and eating habits. Computers in Human Behavior, 35, 364–375. https://doi.org/10.1016/j.chb.2014.01.036

Rosenberg, D. E., Norman, G. J., Wagner, N., Patrick, K., Calfas, K. J., & Sallis, J. F. (2010). Reliability and validity of the sedentary behavior questionnaire (SBQ) for adults. Journal of Physical Activity and Health, 7, 697–705. https://doi.org/10.1123/jpah.7.6.697

Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. https://doi.org/10.1037/h0077714

Shao, G. (2009). Understanding the appeal of user‐generated media: A uses and gratification perspective. Internet Research, 19(1), 7–25. https://doi.org/10.1108/10662240910927795

Singh, G. K., Kogan, M. D., Siahpush, M., & Van Dyck, P. C. (2008). Independent and joint effects of socioeconomic, behavioral, and neighborhood characteristics on physical inactivity and activity levels among US children and adolescents. Journal of Community Health, 33(4), 206–216. https://doi.org/10.1007/s10900-008-9094-8

Stead, H., & Bibby, P. A. (2017). Personality, fear of missing out and problematic internet use and their relationship to subjective well-being. Computers in Human Behavior, 76, 534–540. https://doi.org/10.1016/j.chb.2017.08.016

Song, H., Zmyslinski-Seelig, A., Kim, J., Drent, A., Victor, A., Omori, K., & Allen, M. (2014). Does Facebook make you lonely?: A meta-analysis. Computers in Human Behavior, 36, 446–452. https://doi.org/10.1016/j.chb.2014.04.011

Spector, P. E., & Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: Interpersonal conflict at work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory. Journal of Occupational Health Psychology, 3, 356–367. https://doi.org/10.1037/1076-8998.3.4.356

Stapleton, P., Luiz, G., & Chatwin, H. (2017). Generation validation: The role of social comparison in use of Instagram among emerging adults. Cyberpsychology, Behavior, and Social Networking, 20(3), 142–149. https://doi.org/10.1089/cyber.2016.0444

Tandon, P. S., Zhou, C., Sallis, J. F. (2012). Home environment relationships with children’s physical activity, sedentary time, and screen time by socioeconomic status. International Journal of Behavioral Nutrition and Physical Activity, 9(88). https://doi.org/10.1186/1479-5868-9-88

Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach's alpha. International Journal of Medical Education, 2, 53–55. https://doi.org/10.5116/ijme.4dfb.8dfd

Tromholt, M. (2016). The Facebook experiment: Quitting Facebook leads to higher levels of well-being. Cyberpsychology, Behavior, and Social Networking, 19, 661–666. https://doi.org/10.1089/cyber.2016.0259

Twenge, J. M., Joiner, T. E., Rogers, M. L., & Martin, G. N. (2017). Increases in depressive symptoms, suicide-related outcomes, and suicide rates among U.S. adolescents after 2010 and links to increased new media screen time. Clinical Psychological Science, 6, 3–17. https://doi.org/10.1177/2167702617723376

Twenge, J. M., Martin, G. N., & Campbell, W. K. (2018). Decreases in psychological wellbeing among American adolescents after 2012 and links to screen

time during the rise of smartphone technology. Emotion, 18(6), 765–780. http://dx.doi.org/10.1037/emo0000403

U.S. Department of Agriculture. (2020). What is MyPlate? https://www.myplate.gov/eat-healthy/what-is-myplate

Valkenburg, P. M., & Peter, J. (2007). Online communication and adolescent well-being: Testing the stimulation versus the displacement hypothesis. Journal of Computer-Mediated Communication, 12(4), 1169–1182. https://doi.org/10.1111/j.1083-6101.2007.00368.x

Vanman, E. J., Baker, R., & Tobin, S. (2018). The burden of online friends: The effects of giving up Facebook on stress and well-being. The Journal of Social Psychology, 158, 496–507. https://doi.org/10.1080/00224545.2018.1453467

Vaterlaus, J. M., & Winter, M. (2021). TikTok: An exploratory study of young adults’ uses and gratifications. The Social Science Journal, 1–20. https://doi.org/10.1080/03623319.2021.1969882

Vogel, E. A., Rose, J. P., Okdie, B. M., Eckles, K., & Franz, B. (2015). Who compares and despairs? The effect of social comparison orientation on social media use and its outcomes. Personality and Individual Differences, 86, 249–256. https://doi.org/10.1016/j.paid.2015.06.026

Vogel, E. A., Rose, J. P., Roberts, L. R., & Eckles, K. (2014). Social comparison, social media, and self-esteem. Psychology of Popular Media Culture, 3(4), 206–222. https://doi.org/10.1037/ppm0000047

Woods, H. C., & Scott, H. (2016). #Sleepyteens: Social media use in adolescence is associated with poor sleep quality, anxiety, depression and low selfesteem. Journal of Adolescence, 51, 41–49. https://doi.org/10.1016/j.adolescence.2016.05.008

Wright, R. R., Broadbent, C., Graves, A., & Gibson, J. (2016). Health behavior change promotion among Latter-day Saint college students. Psi Chi Journal of Psychological Research, 21(3), 200–215. https://doi.org/10.24839/b21.3.200

Wright, R. R., Hardy, K., Shuai, S. S., Egli, M., Mullins, R., & Martin, S. (2017). Loneliness and social media use among religious Latter-day Saint college students: An exploratory study. Journal of Technology in Behavioral Science, 3, 12–25. https://doi.org/10.1007/s41347-017-0033-3

Wright, R. R., Larson, J., Richards, S., Hoffmann, S., & Nienstedt, C. (2023). The COVID-19 Pandemic: Electronic media use and health among U.S. college students. Journal of American College Health https://doi.org/10.1080/07448481.2022.2155463

Wright, R. R., Schaeffer, C., Mullins, R., Evans, A., & Cast, L. (2020). Comparison of student health and wellbeing profiles and social media use. Psi Chi Journal of Psychological Research, 25(1), 14–21. https://doi.org/10.24839/2325-7342.JN25.1.14

Wright, R. R., Evans, A., Schaeffer, C., Mullins, R., & Cast, L. (2021). Social networking site use: Implications for health and wellness. Psi Chi Journal of Psychological Research, 26(2), 165–175. https://doi.org/10.24839/2325-7342.JN26.2.165

Yang, C. C. (2016). Instagram use, loneliness, and social comparison orientation: Interact and browse on social media, but don't compare. Cyberpsychology, Behavior, and Social Networking, 19(12), 703–708 https://doi.org/10.1089/cyber.2016.0201

Yang, Z., Heizhati, M., Wang, L., Li, M., Pan, F., Wang, Z., Abudureyimu, R., Hong, J., Yao, L., Yang, W., Liu, S., & Li, N. (2021). Subjective poor sleep quality is associated with higher blood pressure and prevalent hypertension in general population independent of sleep disordered breathing. Nature and Science of Sleep, 13, 1759–1770. https://doi.org/10.2147/NSS.S329024

Author Note. Robert R. Wright https://orcid.org/0000­0002­4101­7840

We declare no conflict of interest. Support for this study came from the Department of Psychology at Brigham Young University – Idaho. Special thanks to Stephie L. San Diego and Ashley M. Shin for their input in early stages of this study. Correspondence concerning this article should be sent to Christian Nienstedt, Department of Psychology, Brigham Young University – Idaho, 525 South Center St., MS 2140 Rexburg, ID 83460. Telephone: 480­206­4774. Email: cnienstedt97@gmail.com

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

The Impact of Age and Gender on Trust in the U.S. Federal Government

Kailey A. Schroeder, Christopher A. Creecy*, Emily L. Perkins, and K. E. Carter Department of Behavioral Sciences, Freed-Hardeman University

ABSTRACT. The purpose of this study was to measure differences in governmental trust levels among adult Americans according to gender (male vs. female) and age group (20–39, 40–59, and over 60) as well as how those 2 variables interact. It was hypothesized that the interaction of age and gender would have a significant impact on the level of trust adult Americans place in their political institutions. A sample of 262 participants responded to 8 questions on the Trust Index Scale, and a 3 x 2 between­groups factorial ANOVA was used to analyze the results. No main effect on governmental trust was found for the 2 independent variables, but a significant interaction between age and gender was discovered, F(2, 256) = 4.64, p = .01, with a 95% CI [17.61, 18.81]. However, the effect size of this interaction was quite small (partial η2 = .04). Implications for the perceptual and relational approaches to trust are discussed.

Keywords: gender, age, trust, federal government, Trust Index Scale

In the midst of the COVID ­ 19 pandemic, 80.9% of Americans trusted information from the CDC, 75.6% trusted information from state health departments, 41.2% trusted mainstream media, and least of all, only 20.9% of Americans trusted information from the White House (Latkin et al., 2020). The formation of nonpeaceful groups such as the Proud Boys, Antifa, and the Boogaloo Boys, as well as events such the riot at the nation's capital, are all demonstrative of the lack of confidence that the citizens of the United States have in the ability of their federal government to make good decisions, create equality, and keep its people safe (Dave et al., 2022; Eberly, 2022). Trust in the federal government has been declining since the 1960s with the only exceptions being periods of time when the nation experienced economic growth (Public Trust, 2021) suggesting that the current global inflation rates will likely continue the erosion of trust that the United States has experienced for over half a century (Maeng & Aggarwal, 2022).

Trust is a complex emotion that is context­specific, and trust must be studied in the environment in which

it is produced, because it is affected by numerous situational factors such as a person's perception of an event or the entity in which or whom they are placing their trust (Kappmeier et al., 2021). Rousseau et al. (1998) suggested that trust can be defined as the willingness to take risks and be vulnerable based upon positive expectations of another person’s behavior and intentions. The first characteristics that individuals generally take note of as they are deciding who to trust involve personalities, cultural characteristics, and assumed or known experiences of the trustee (Hurley, 2006). Stereotypes can influence how people form trusting relationships, and although the information assumed about other individuals does influence the extent to which someone may be willing to trust them, learned information is less influential when there is a time constraint (Rudoy & Paller, 2009)—the quicker an interaction, the more trustworthiness is based on visual perceptions rather than relations.

Gender and Trust

Previous research has suggested that men and women have different perspectives on trust and their trust in

Diversity badge earned for conducting research focusing on aspects of diversity.

the government, and they make decisions differently as a result (Oh et al., 2020; Rotter, 1967). Mayer et al. (1995) proposed two major approaches regarding trust—the relational approach and the perceptual approach. The relational approach explains how trust is enabled by identifying with other members of a community; race, gender, education, and socioeconomic status have a strong impact on how people relate to, and therefore trust, those around them (Crepaz et al., 2014). In the perceptual approach, trust is based on situational cues and past experiences, and people base their trust on the degree to which they perceive another person as being trustworthy (Mayer et al., 1995).

From a relational standpoint, women tend to automatically trust other women much more than they do men (Zhao & Zhang, 2016). Yet, women are also more likely to maintain trust in others after repeated untrustworthy actions and regain trust in a previously untrustworthy counterpart because of a tendency to place more emphasis on physical appearance (Haselhuhn et al., 2015). Relationally, men tend to be less sensitive to social risks and have less gender bias when determining where to place their trust (Wu et al., 2020; Zhao & Zhang, 2016). Women are often trusted more by others than their male counterparts (Riedl et al., 2010), and when women hold leadership positions, people tend to have higher levels of trust in their leadership and expect to be treated more fairly (Joshi & Diekman, 2022).

Perceptually, women tend to attribute more emphasis to physical appearance, culture, and level of perceived kindness when deciding whom to trust (Golesorkhi, 2006). They are also more likely to rank the trustworthiness of an individual based on attractiveness—likely because attractive individuals are automatically seen to be more healthy, honest, and intelligent (Li & Liu, 2021; Zebrowitz, 2017). Trustworthy facial expressions seem to be particularly influential for men, and they tend to trust conventionally attractive individuals more (Kovacs­Balint et al., 2013), likely because a high level of attractiveness is considered a more feminine characteristic associated with devalued performance and overall competency (Braun et al., 2012; Sczesny et al., 2006). Perceptions of trustworthy faces seem to adapt over time for men as interaction with and knowledge about others grow, but the same is not true for women (Wincenciak et al., 2013). Finally, men in leadership roles often tend to elicit lower instinctual levels of trust in women because of past prejudice and the domination of men in high positions in the workplace (Joshi & Diekman, 2022).

Age and Trust

Caregivers begin teaching children about the world and the concept of trust by implementing social values they

consider to be essential such as not talking to strangers (Hooghe & Stolle, 2004). As a result, young children typically do not blindly place their trust in authority figures when they are able to distinguish between reliable and unreliable sources (Clément et al., 2004), although young children can be incredibly trusting (Vanderbilt et al., 2014). Stolle and Nishikawa (2011) noted that as middle­aged children develop, they often trust authority figures and unknown situations significantly less than their caregivers. During adolescence trust enters a transitional period when the ability to evaluate trust in others and base social relationships on trust becomes more important—particularly as increasing self­awareness becomes further established by late adolescence (Kragel & LaBar, 2015).

Young adults ages 18–24 in the 1960s and 1970s often had initial faith in the government, but young adults presently have a sense of skepticism and doubt (Dalton et al., 2005). Young adults have been shown to have less overall trust than any other age group (Chen et al., 2019), and this age group also has the fastest declining trust levels (McLeigh, 2015; Wu et al., 2020). This likely stems from the way that younger adults evaluate personal impact above overall well­being when making decisions (Korn et al., 2021). Low trust levels in the federal government among young adults contributed to the disregard for government­mandated safety precautions during the COVID­19 pandemic (Schaal, 2004), and the group of rioters at the U.S. Capitol on January 6, 2021, was mainly comprised of people in this age group (Jacobs, 2022). When younger adults begin to form their political views and political trust, better­educated youth are more likely to exhibit greater concern for new quality of life issues that often put them in conflict with the dominant political parties and existing government priorities (Dalton et al., 2005). However, the more that young adults participate in public events and form relationships, the more likely they are to place trust in other people and institutions (McLeigh, 2015).

Trust develops gradually from middle age childhood through middle age adulthood and then levels off (Robinson & Jackson, 2001). As people have more life experiences and more interactions with others, they become more trusting, and then they maintain this higher level of trust over the remaining lifespan (Clark & Einstein, 2013). As people move into the life situations of middle age that evoke or require civic engagement, general inclination to trust or distrust others becomes increasingly important, and socioeconomic characteristics can become increasingly important predictors of trust and engagement (Jennings & Stoker, 2004).

Older age is significantly associated with a higher level of generalized trust when compared to younger

Age, Gender, and Governmental Trust | Schroeder, Creecy, Perkins, and Carter

Creecy, Perkins, and Carter | Age, Gender, and Governmental Trust

adults (Bailey & Leon, 2019; Li & Fung, 2013), because there is a notable association between heightened levels of emotional well­being and trust (Lang & Carstensen, 2002). Older adults tend to place their trust in people and institutions based on the deeper level of connection for which this age group is searching (Bailey et al., 2015; Poulin & Haase, 2015). As a result, it can be more difficult for the older population to tell when they are being deceived or directly misled (Slessor et al., 2012), and older adults are more likely to remember, pay attention to, and trust negative information (Jefferies & Reed, 2000). Even so, older adults seem to be more accurate in assessing trustworthy characteristics than younger people and are less likely to be swayed by superficial characteristics when determining who to trust (Castle et al., 2012).

Trust in the Federal Government

Political trust refers to the confidence people have in their government, institutions, and the policies enforced by these entities; this notion of trust hinges on beliefs that institutions will competently and fairly deliver their services (Hudson, 2006). Though the general population expresses a general, overall distrust in the government, most have a strong sense of trust as it pertains to implicit trust in the government (Intawan & Nicholson, 2018). Freitag and Ackermann (2015) suggested that an individual’s first reaction is to trust the government and to be more agreeable with its policies, and Sibley et al. (2020) noted that an act keeping the nation's best interest and safety first strengthens trust in the government and its decisions. Unique to governmental trust, trust levels tend to fluctuate substantially in response to who the elected officials are in accordance with the political party individuals identify with (Listhaug, 1995). Trust influences public opinion and thereby the incentive structure for political leaders (Justwan et al., 2018), and there are significant gender differences in political trust, as women tend to express higher levels of political trust than men (Schoon & Cheng, 2011). Different personal factors can also affect personal trust in the government. Education, cognitive ability, socioeconomic status, long­term unemployment, and heightened levels of fear can all influence political trust levels (Hudson, 2006; Rindermann, 2008; Vasilopoulos et al., 2022).

Visual judgments can determine the level of trust individuals have in others as well as institutions, suggesting that individuals use the perceptual approach not only to establish trust on a daily basis but also for very large and impactful decisions regarding trust (Duarte et al., 2012). Regardless of political orientation, people tend to trust male politicians more because male faces tend to have more dominance­related features, which people tend to attribute to good leadership qualities (Maeng &

Aggarwal, 2022). The relational approach implies that governmental trust can be improved by establishing relationships through with the general population; that is, the more government and citizen interaction occurs, the more the people of the United States will trust the government and its elected officials (Cole, 2021). Given that most individuals in the United States have accounts on multiple social media platforms, more positive and recurrent interactions between the government and the community should inevitably lead to higher trust the more frequently citizens are permitted to voice their opinions about daily decisions that affect themselves and their families (Evans et al., 2018).

Current Study

In 1964, 75% of Americans trusted the government to consistently do the right thing for its citizens (Nye, 1997). Since then, the extent to which people trust the government has continued to plummet to the point where even the topic of government officials often makes many Americans angry (Webster, 2017). Since 2020, trust in the federal government appears to be declining at an even more rapid pace as a result of the COVID­19 pandemic, sociopolitical events such as the disputed 2020 election results, regular riots, and violent political groups. This is problematic because trust in authority figures is how citizens cooperate and how society functions efficiently (Kritzinger et al., 2021). As noted, previous research has suggested that men and women have different perspectives on trust and their trust in the government, and they make decisions differently as a result (Oh et al., 2020). It is also known that individuals’ views on the government depend heavily on their age (Lie et al., 2009). Little research, however, has addressed how gender and age interact to impact trust. The purpose of this study was to measure differences in governmental trust levels among adult Americans according to gender (male vs. female) and age group (20–39, 40–59, and over 60) as well as the interaction between these two variables with the hypothesis that the interaction of age and gender will have a significant impact on the level of trust adult Americans place in the U.S. federal government.

Method

Sample

Participants were recruited using ads posted on three different social media sites (Facebook, Instagram, and Twitter). Initially, the ad was posted on the personal social media sites of the three researchers, and viewers of the recruiting ad were asked to share the post on their own personal social media sites in order to recruit more participants through snowball sampling. Anyone age 20 or older was eligible to participate in the study. A total of

Schroeder,

266 people responded by clicking the link to a Google Form included with the recruiting ad. Four participants were removed from the study because they either did not meet the age criterion, did not consent to the given form, or did not complete all survey questions resulting in a final sample of 262 participants (M = 82, F = 182). Participants were divided into groups based on gender (male and female) and age. Age group 1 was defined as 20–39­year­olds (n = 122, M = 30, F = 90); age group 2 was defined as 40–59­year­olds (n = 101, M = 38, F = 63); and age group 3 was defined as 60 years or older (n = 41, M = 14, F = 27). These age groups were chosen because (a) they are representative of equal proportions of the current population (Statista, 2020) and (b) they are consistent with other studies in which age was a categorical variable and participants were divided into similar age groupings (Korn et al., 2021; Seijts et al., 2021). Table 1 shows the percentage of each group represented in both the total population and the final sample. Data was collected over a six­week period in January and February 2022 after the research project was approved by the Freed­Hardeman University Institutional Review Board.

Design and Procedure

A between­groups 3 x 2 factorial ANOVA design was used to examine how the interaction of gender and age impact governmental trust. Data were collected using a Google Form which required participants to provide informed consent and answer two demographic questions (age and gender) which were presented as categorical variables. No other demographic information was collected from participants as the scope of this study was simply limited to age and gender. Participants completed a slightly modified version of the Trust Index Scale (Sharoni, 2012; α = .81) in which they were asked a series of eight questions such as “How much trust and confidence do you have in our federal government in Washington when it comes to handling international problems?” and “How much trust and confidence do you have at the time in the U.S. House of Representatives?” Participants chose a number on a Likert­type scale that best represented their level of trust on each item where 1 represented full trust and 5 represented no trust at all. All eight items were reversed scored and summed to produce a total score between 8 and 40 for each participant that indicated an individual’s level of trust in the federal government where higher scores indicate higher levels of trust. The data was analyzed using a two­way ANOVA to determine whether any main effects for gender and age were present or whether an interaction between these two variables occurred which may have impacted levels of government trust. We used SPSS for all analyses and ran a Bonferroni post hoc test to examine any statistically significant differences that may have been present among the three age groups.

Results

We hypothesized that the interaction of age and gender would have a significant impact on the level of trust adult Americans place in the U.S. federal government. Table 2 shows the mean scores on the Trust Index Scale for each gender, the three age groups, and the six pairings for gender by age. Equal variance across groups was supported by Levene’s Test of Equality of Error Variances, F(5, 256) = 0.49, p = .79. A main effect analysis revealed that neither age, F(1, 256) = 0.06, p = .94, nor gender, F(1, 256) = 0.00, p = .97, had a statistically significant effect on the level of trust individuals have in the U.S. federal government. A statistically significant interaction between age and gender was discovered, F(2, 256) = 4.64, p = .01, with a 95% CI [17.61, 18.81], but the effect size of this interaction was quite small (partial η2 = .04) as shown by the group means in Table 2.

Discussion

With the erosion of trust in the U.S. federal government (Webster, 2017), especially since the beginning of the COVID­19 crisis through the present (Kritzinger et al., 2021), examining variables which impact trust is timely and relevant. Trust levels are an important part of society because higher levels of trust are indicative of socioemotional states (Curşeu & Schruijer, 2010). Although other variables could certainly have been considered

TABLE 1

Population and Sample Demographics

Note. Populations are representative of American citizens in the given groups, and samples are representative of respondents within that criterion. Each unit represents 1 million people.

TABLE 2

Means and Standard Deviations by Age and Gender

Note. Scores on The Trust Index Scale show trust levels in the United States federal government for the six groups of participants with possible scores ranging from 8 to 40.

as part of this study, the current research was limited to just two—gender and age—as an exploratory study on interaction. Research findings on trust with regard to both gender and age (as well as other variables) are prevalent in scientific literature, and the purpose of our study was to examine a simple interaction as a first step toward understanding how multiple variables may work in tandem to impact trust in the federal government. Our results did not indicate that gender and age independently have a significant impact on Americans' trust in the federal government, but we did find that trust levels fluctuate in very opposite directions between genders across the lifespan as shown in Figure 1. Younger women tend to have more trust in the U.S. federal government, but this trust significantly diminishes as women grow older. The effect is reversed for men who tend to trust the federal government less when they are younger, with trust increasing significantly over the lifespan. Thus, the amount of trust one gender places in the federal government depends on the age category in which an individual of that gender is classified. While the results of the present study support our hypothesis, the effect size is relatively small.

It is interesting to note that, while not a significant difference, women did indicate an overall higher level of trust in the federal government than men contradicting previous research which suggests that women have a tendency to trust other women more than men (Zhao & Zhang, 2016). Further, the overall means between the age groups—also not statistically significant—do decrease with age, which contrasts with earlier findings that trust increases over the lifespan through middle age, levels off, and remains at a higher level for the duration of the

Interaction Between Age and Gender on Trust

lifespan (Clark & Einstein, 2013; Robinson & Jackson, 2001). What is consistent with prior research is that the total trust rating for all six groups is relatively low as noted in Table 2, and differences in priorities between genders (Oh et al., 2020) and among age groups (Li et al., 2009) likely influence the criteria individuals choose when deciding what and whom to trust (Carstensen, 2006).

Older men reported a significantly higher amount of trust in the federal government than older women. As men age, they see more of their peers in the government who are predominantly of the same gender as themselves suggesting that older men can more easily relate to government officials (Nakano and Yamamoto, 2022) than older women. Consequently, men are more likely to place trust in the government when the majority of the current elected officials meet a subconscious standard of what men perceive to be typical or normal. It may be that younger men have not yet formed a perception of what a typical face in the federal government looks like, but as they grow into middle age, their life experiences more definitively shape their perception (Kovacs­Balint et al., 2013). Because men tend to trust more conventionally attractive individuals, it could also be that the lack of diversity among elected officials in the U.S. federal government has formed a representative face for these positions among older men.

Conversely, women may become less trusting as they age because there is less gender representation in the federal government for older women than younger ones (Fredrickson & Carstensen, 1990). Of the 100 current members of the Senate, 76 are men and 24 are women (3:1 ratio), and only nine of those 24 women are age 60 or older. Although representation of women in each political party is increasing, this is occurring mainly in the younger age groups, which may explain why younger women indicated the most trust in the federal government among the six groups. Further, as women age, they gain a better understanding of the inherent limitations of their gender. Older women face more barriers than men in general due to the glass ceiling effect (Mandel & Rotman, 2022). That older women put less trust in the federal government for this reason is consistent with Joshi and Diekman (2022) who suggested that women tend to put less trust in men who leadership roles because of their past experiences.

There are at least three limitations that must be noted when considering the results. First, a factorial ANOVA was used to analyze the data. In hindsight, a multiple regression analysis would (a) more readily show the relationship between either predictor variables or covariates and the trust outcome variable, and (b) identify the impact of each predictor variable or covariate. Second, analyzing age as a continuous variable instead of a categorical variable

Schroeder, Creecy, Perkins, and Carter | Age, Gender, and Governmental Trust

FIGURE 1

Note. Mean scores for trust are shown by gender and age indicating a cross-over effect.

hindered a more robust interpretation of the results. Third, the manner in which data was collected including (a) the use of snowball sampling, (b) exclusively collecting data online, and (c) the use of limited social media platforms to recruit participants all have inherent biases that may have impacted the results of this study.

Despite these limitations, this exploratory research did establish a relatively small interaction effect between age and gender that, to our knowledge, have not been directly established in previous research. The results of this study provide some support for an interaction between age and gender in relation to trust in the U.S. federal government, but more research is needed before a full understanding of how trust in the federal government is established as a function of these variables. Future research on trust might also benefit from the addition of other demographic variables such as education level, socioeconomic status, race, and ethnicity in order to increase generalizability, as this study sheds no light on how additional demographic variables interact with or predict trust in the federal government. Replication of this study in the future when a different political party is in power or at a time when there is more female representation in the White House might also help clarify whether these additional variables influence trust among genders and between age groups.

References

Bailey, P. E., & Leon, T. (2019). A systematic review and meta-analysis of agerelated differences in trust. Psychology and Aging, 34(5), 674–685. https://doi.org/10.1037/pag0000368

Bailey, P. E., Slessor, G., Rieger, M., Rendell, P. G., Moustafa, A. A., & Ruffman, T. (2015). Trust and trustworthiness in young and older adults. Psychology and Aging, 30(4), 977–986. https://doi.org/10.1037/a0039736

Braun, S., Peus, C., & Frey, D. (2012). Is beauty beastly? Gender-specific effects of leader attractiveness and leadership style on followers’ trust and loyalty. Zeitschrift Für Psychologie, 220(2), 98–108. https://doi.org/10.1027/2151-2604/a000101

Carstensen, L. L. (2006). The influence of a sense of time on human development. Science, 312(5782), 1913–1915. https://doi.org/10.1126/science.1127488

Castle, E., Eisenberger, N. I., Seeman, T. E., Moons, W. G., Boggero, I. A., Grinblatt, M. S., & Taylor, S. E. (2012). Neural and behavioral bases of age differences in perceptions of trust. PNAS Proceedings of the National Academy of Sciences of the United States of America, 109(51), 20848–20852. http://dx.doi.org/10.1073/pnas.1218518109

Chen, L. H., Gramlich, J., & Houser, K. A. (2019). The effects of board gender diversity on a firm's risk strategies. Accounting & Finance, 59(2), 991–1031. https://doi.org/10.1111/acfi.12283

Clark, A. K., & Eisenstein, M. A. (2013). Interpersonal trust: An age–period–cohort analysis revisited. Social Science Research, 42(2), 361–375. https://doi.org/10.1016/j.ssresearch.2012.09.006

Clément, F., Koenig, M., & Harris, P. (2004). The ontogenesis of trust. Mind and Language, 19(4), 360–379. https://doi.org/10.1111/j.0268-1064.2004.00263.x

Cole, T. M. (2021). The role of social media in supporting US citizens’ trust in government. The Sciences and Engineering, 82(5), 12–36. https://scholarworks.waldenu.edu/dissertations

Crepaz, M., Polk, J., Bakker, R., & Singh, S. (2014). Trust matters: The impact of ingroup and outgroup trust on nativism and civicness. Social Science Quarterly, 95(4), 938–959. https://doi.org/10.1111/ssqu.12082

Curşeu, P. L., & Schruijer, S. G. L. (2010). Does conflict shatter trust or does trust obliterate conflict? Revisiting the relationships between team diversity,

conflict, and trust. Group Dynamics: Theory, Research, and Practice, 14(1), 66–79. https://doi.org/10.1037/a0017104

Dave, D., McNichols, D., & Sabia, J. J. (2022). Political violence, risk aversion, and population health: Evidence from the US Capitol riot. Journal of Population Economics, 35(4), 1345–1384. https://doi.org/10.1007/s00148-022-00914-0

Dalton, K. M., Nacewicz, B. M., Johnstone, T., Schaefer, H. S., Gernsbacher, M. A., Goldsmith, H. H., & Davidson, R. J. (2005). Gaze fixation and the neural circuitry of face processing in autism. Nature Neuroscience, 8(4), 519–526. https://doi.org/10.1038/nn1421

Duarte, J., Siegel, S., & Young, L. (2012). Trust and credit: The role of appearance in peer-to-peer lending. The Review of Financial Studies, 25(8), 2455–2485. https://doi.org/10.1093/rfs/hhs071

Eberly, T. E. (2022). Trust and anger in the Trump era. In S. E. Schier & T. E. Eberly (Eds.), The Trump effect: Disruption and its consequences in US politics and government (pp. 1–20). Rowan & Littlefield.

Evans, L., Franks, P., & Chen, H. M. (2018). Voices in the cloud: Social media and trust in Canadian and US local governments. Records Management Journal, 28(1), 18–46. https://doi.org/10.1108/RMJ-11-2016-0041

Fredrickson, B. L., & Carstensen, L. L. (1990). Choosing social partners: How old age and anticipated endings make people more selective. Psychology & Aging, 5(3), 335–347. https://doi.org/10.1037/0882-7974.5.3.335

Freitag, M., & Ackermann, K. (2015). Direct democracy and institutional trust: Relationships and differences across personality traits. Political Psychology, 37(5), 707–723. https://doi.org/10.1111/pops.12293

Golesorkhi, B. (2006). Gender differences and similarities in judgments of trustworthiness. Women in Management Review, 21(3), 195–210. https://doi.org/10.1108/09649420610657380

Haselhuhn, M. P., Kennedy, J. A., Kray, L. J., Van Zant, A. B., & Schweitzer, M. E. (2015). Gender differences in trust dynamics: Women trust more than men following a trust violation. Journal of Experimental Social Psychology, 56, 104–109. https://doi.org/10.1016/j.jesp.2014.09.007

Hooghe, M., & Stolle, D. (2004). Good girls go to the polling booth, bad boys go everywhere: Gender differences in anticipated political participation among American fourteen-year-olds. Women & Politics, 26(3–4), 1–23. https://doi.org/10.1300/J014v26n03_01

Hudson, J. (2006). Institutional trust and subjective well-being across the EU. Kyklos, 59(1), 43–62. https://doi.org/10.1111/j.1467-6435.2006.00319.x

Hurley, R. F. (2006, September). The decision to trust. Harvard Business Review, 84(9), 55–62. https://hbr.org/2006/09/the-decision-to-trust

Intawan, C., & Nicholson, S. P. (2018). My trust in government is implicit: Automatic trust in government and system support. The Journal of Politics, 80(2), 601–614. https://doi.org/10.1086/694785

Jacobs, R. A. (2022). Elite impact on the Capitol Hill riot: The straw that stirs the drink. [Undergraduate Honors Thesis, Yeshiva University].

Jeffries, F. L., & Reed, R. (2000). Trust and adaptation in relational contracting. Academy of Management Review, 25(4), 873–882. https://doi.org/10.5465/amr.2000.3707747

Jennings, M. K., & Stoker, L. (2004). Social trust and civic engagement across time and generations. Acta Politica, 39, 342–379. https://doi.org/10.1057/palgrave.ap.5500077

Joshi, M. P., & Diekman, A. B. (2022). My fair lady? Inferring organizational trust from the mere presence of women in leadership roles. Personality and Social Psychology Bulletin, 48(8), 1220–1237. https://doi.org/10.1177/01461672211035957

Justwan, F., Bakker, R., & Berejikian, J. D. (2018). Measuring social trust and trusting the measure. The Social Science Journal, 55(2), 149–159. https://doi.org/10.1016/j.soscij.2017.10.001

Kappmeier, M., Guenoun, B., & Fahey, K. H. (2021). Conceptualizing trust between groups: An empirical validation of the five-dimensional intergroup trust model. Peace and Conflict: Journal of Peace Psychology, 27(1), 90–95. https://doi.org/10.1037/pac0000537

Korn, L., Siegers, R., Eitze, S., Sprengholz, P., Taubert, F., Böhm, R., & Betsch, C. (2021). Age differences in COVID-19 preventive behavior: A psychological perspective. European Psychologist, 26(4), 359–372. https://doi.org/10.1027/1016-9040/a000462

Kovács‐Bálint, Z., Bereczkei, T., & Hernádi, I. (2013). The telltale face: Possible mechanisms behind defector and cooperator recognition revealed by emotional facial expression metrics. British Journal of Psychology, 104(4), 563–576. https://doi.org/10.1111/bjop.12007

Kragel, P. A., & LaBar, K. S. (2015). Multivariate neural biomarkers of emotional states are categorically distinct. Social Cognitive and Affective Neuroscience,

SUMMER 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

113

Schroeder, Creecy, Perkins, and Carter | Age, Gender, and Governmental Trust

10(11), 1437–1448. https://doi.org/10.1093/scan/nsv032

Kritzinger, S., Foucault, M., Lachat, R., Partheymüller, J., Plescia, C., & Brouard, S. (2021). ‘Rally round the flag’: The COVID-19 crisis and trust in the national government. West European Politics, 44(5–6), 1–27.

https://doi.org/10.1080/01402382.2021.1925017

Lang, F. R., & Carstensen, L. L. (2002). Time counts: Future time perspective, goals, and social relationships. Psychology and Aging, 17(1), 125–139. http://dx.doi.org/10.1037/0882-7974.17.1.125

Latkin, C. A., Dayton, L., Strickland, J. C., Colon, B., Rimal, R., & Boodram, B. (2020). An assessment of the rapid decline of trust in US sources of public information about COVID-19. Journal of Health Communication, 25(10), 764–773. https://doi.org/10.1080/10810730.2020.1865487

Li, T., & Fung, H. H. (2013). Age differences in trust: An investigation across 38 countries. The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 68, 347–355. https://doi.org/10.1093/geronb/gbs072

Li, N., & Liu, N. (2021). The nonlinear and gender-related relationships of face attractiveness and typicality with perceived trustworthiness. Frontiers in Psychology, 12(65), 60–84. https://doi.org/10.3389/fpsyg.2021.656084

Lie, M., Baines, S., & Wheelock, J. (2009). Citizenship, volunteering and active aging. Social Policy & Administration, 43(7), 702–718. https://doi.org/10.1111/j.1467-9515.2009.00690.x

Listhaug, O. (1995). The dynamics of trust in politicians. In H. D. Klingemann & D. Fuchs (Eds.), Citizens and the state (pp. 261–297). Oxford University Press.

Maeng, A., & Aggarwal, P. (2022). The face of political beliefs: Why gender matters for electability. Journal of the Association for Consumer Research, 7(3), 360–370. https://doi.org/10.1086/719579

Mandel, H., & Rotman, A. (2022). The stalled gender revolution and the rise of top earnings in the United States, 1980 to 2017. Sociological Science, 9, 136–158. https://doi.org/10.15195/v9.a6

Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. The Academy of Management Review, 20(1), 709–734. https://doi.org/10.5465/amr.1995.9508080335

McLeigh, J. D. (2015). Creating conditions that promote trust and participation by young people And why it matters. American Journal of Orthopsychiatry, 85(6), S67–S69. https://doi.org/10.1037/ort0000134

Nakano, T., & Yamamoto, T. (2022). You trust a face like yours. Humanities and Social Sciences Communications, 9(1), 1–6.

https://doi.org/10.1057/s41599-022-01248-8

Nye, J. (1997). In government we don't trust. Foreign Policy, 108, 99–111.

https://doi.org/10.2307/1149092

Oh, D., Dotsch, R., Porter, J., & Todorov, A. (2020). Gender biases in impressions from faces: Empirical studies and computational models. Journal of Experimental Psychology: General, 149(2), 323–342. https://doi.org/10.1037/xge0000638

Poulin, M. J., & Haase, C. M. (2015). Growing to trust: Evidence that trust increases and sustains well-being across the life span. Social Psychological and Personality Science, 6(6), 614–621. http://dx.doi.org/10.1177/1948550615574301

Public Trust in Government: 1958-2022. (2022, June 6). Pew Research Center - U.S. Politics & Policy. https://www.pewresearch.org/politics/2022/06/06/publictrust-in-government-1958-2022/

Riedl, R., Hubert, M., & Kenning, P. (2010). Are there neural gender differences in online trust? An fMRI study on the perceived trustworthiness of eBay offers. MIS Quarterly, 34(2), 397–428. https://doi.org/10.2307/20721434

Rindermann, H. (2008). Relevance of education and intelligence for the political development of nations: Democracy, rule of law and political liberty. Intelligence, 36(4), 306–322. https://doi.org/10.1016/j.intell.2007.09.003

Robinson, R. V., & Jackson, E. F. (2001). Is trust in others declining in America? An age-period-cohort analysis. Social Science Research, 30(1), 117–145. https://doi.org/10.1006/ssre.2000.0692

Rotter, J. B. (1967). A new scale for the measurement of interpersonal trust. Journal of Personality, 35(4), 651–655. https://doi.org/10.1111/j.1467-6494.1967.tb01454.x

Rousseau, D. M., Sitkin, S. B., Burt. R. S., & Camerer, C. (1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23(1), 393–404. https://doi.org/10.5465/amr.1998.926617

Rudoy, J. D., & Paller, K. A. (2009). Who can you trust? Behavioral and neural differences between perceptual and memory-based influences. Frontiers

in Human Neuroscience, 3(1), 2–5. https://doi.org/10.3389/neuro.09.016.2009

Schaal, G. S. (2004). Vertrauen und demokratie [Trust, constitution, and democracy]. In Vertrauen, Verfassung und Demokratie (pp. 153–187). VS Verlag für Sozialwissenschaften, Wiesbaden. https://doi.org/10.1007/978-3-663-12436-8_4

Schoon, I., & Cheng, H. (2011). Determinants of political trust: A lifetime learning model. Developmental Psychology, 47(3), 619–631. https://doi.org/10.1037/a0021817

Sczesny, S., Spreemann, S., & Stahlberg, D. (2006). Masculine = competent? Physical appearance and sex as sources of gender-stereotypic attributions. Swiss Journal of Psychology, 65, 15–23. https://doi.org/10.1024/1421-0185.65.1.15

Seijts, G. H., de Clercy, C., & Mohan, G. (2021). Trust as a mediator of the relationship between character and perceptions of leader effectiveness during the COVID-19 crisis. Canadian Journal of Behavioural Science / Revue Canadienne Des Sciences Du Comportement, 53(3), 358–364. https://doi.org/10.1037/cbs0000214

Sharoni, S. (2012). E-Citizenship: Trust in government, political efficacy, and political participation in the internet era. Electronic Media & Politics, 1(8), 119–135. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3342574

Sibley, C. G., Greaves, L. M., Satherley, N., Wilson, M. S., Overall, N. C., Lee, C. H. J., Milojev, P., Bulbulia, J., Osborne, D., Milfont, T. L., Houkamau, C. A., Duck, I. M., Vickers-Jones, R., & Barlow, F. K. (2020). Effects of the COVID-19 pandemic and nationwide lockdown on trust, attitudes toward government, and wellbeing. American Psychologist, 75(5), 618–630.

https://doi.org/10.1037/amp0000662

Slessor, G., Phillips, L. H., Bull, R., Venturini, C., Bonny, E. J., & Rokaszewicz, A. (2012). Investigating the “deceiver stereotype”: Do older adults associate averted gaze with deception? The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 67B(2), 178–183. http://dx.doi.org/10.1093/geronb/gbr087

Statista. (2020). Resident population of the United States by sex and age as of July 1, 2020. https://www.statista.com/statistics/241488/population-of-the-us-by-sex-and-age Stolle, D., & Nishikawa, L. (2011). Trusting others—How parents shape the generalized trust of their children. Comparative Sociology, 10(2), 281–314. https://doi.org/10.1163/156913311X566599

Vanderbilt, K. E., Heyman, G. D., & Liu, D. (2014). In the absence of conflicting testimony young children trust inaccurate informants. Developmental Science, 17(3), 443–451. https://doi.org/10.1111/desc.12134

Vasilopoulos, P., McAvay, H., Brouard, S., & Foucault, M. (2022). Emotions, governmental trust and support for the restriction of civil liberties during the covid‐19 pandemic. European Journal of Political Research. Advance online publication. https://doi.org/10.1111/1475-6765.12513

Webster, S. W. (2017). Anger and declining trust in government in the American electorate. Political Behavior, 40(4), 933–964. https://doi.org/10.1007/s11109-017-9431-7

Wincenciak, J., Dzhelyova, M., Perrett, D. I., and Barraclough, N. E. (2013). Adaptation to facial trustworthiness is different in female and male observers. Vision Research. 87(19), 30–34. https://doi.org/10.1016/j.visres.2013.05.007

Wu, Y., Hall, A. S., Siehl, S., Grafman, J., & Krueger, F. (2020). Neural signatures of gender differences in interpersonal trust. Frontiers in Human Neuroscience, 14(1), 1–11. https://doi.org/10.3389/fnhum.2020.00225

Zhao, N., & Zhang, J. (2016). Gender differences in trusting strangers: Role of the target's gender. PsyCh Journal, 5(2), 83–91. https://doi.org/10.1002/pchj.120

Zebrowitz, L. A. (2017). First impressions from faces. Current Directions in Psychological Science, 26(3), 237–242. https://doi.org/10.1177/0963721416683996

Author Note. Kailey A. Schroeder https://orcid.org/0009­0000­4651­0266

Christopher A. Creecy

https://orcid.org/0009­0008­4218­1301

K. E. Carter

https://orcid.org/0009­0005­8919­3666

This research was not sponsored by any organization, and we have no conflicts of interest to disclose.

Correspondence concerning this article should be addressed to Kailey A. Schroeder. Email: kaileyschroeder01@gmail.com

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

The Effects of Imagining and Experiencing Ostracism on Pain Perception

Jennifer Zwolinski Department of Psychological Sciences, University of San Diego

ABSTRACT. This article included 2 experiments to investigate the short­term pain recovery following imagined and directly experienced social threats in healthy adult college samples. Experiment 1 showed that, when participants imagined ostracism, as opposed to inclusion, they expected more immediate physical pain outcomes, although they expected this physical discomfort to dissipate within a few minutes (ps < .05). Experiment 2 found that, when individuals directly experienced ostracism, they reported more pain than when they experienced inclusion; however, this pain reverted to baseline levels within 20 minutes of the social threat (ps < .05). Overall, data from these 2 experiments provide support of the physical pain­social pain overlap literature that social pain does indeed “hurt” but that the pain resolves within 20 minutes.

Keywords: ostracism, rejection, pain, persistence

Over the last two decades, there has been a surge of pharmacological, neuropsychological, and neuroimaging research interest on the socialphysical pain overlap system (Eisenberger, 2014). These studies have shown that both physical pain and social pain share parts of a common underlying neurological system that alerts individuals for potential harm and initiates affective distress once harm has been detected in order to motivate recovery and safety (Eisenberger, 2012). Ferris and colleagues (2019) provided a review of this literature research as well as a framework for studying the social­physical pain overlap, including areas of convergence and divergence in both forms of pain. The evolutionary explanation for the development of a common neurological system is that the social attachment system that keeps people close to others for survival might have co­opted with the pain system. By using the pain signal to signify and identify social threats, individuals are able to quickly manage associated pain and prevent social separation (Eisenberger & Lieberman, 2005; Nelson & Panksepp, 1988).

Williams’ (2009) Temporal Needs Model provides a framework to understand this social­physical pain overlap by examining how individuals respond to the social threats, and the nature of these responses. In the first, reflexive, stage individuals respond quickly to ostracism with distress such as thwarted fundamental needs (e.g., belonging, control) and pain. Next, the individual copes in the reflective stage by attending to

and/or appraising the social threat in order to manage this distress. If ostracism is persistent, the individual’s coping resources can become depleted, in the resignation stage, leading to longer term psychosocial problems such as depression and alienation (Williams, 2009). Without intervention, although individuals show immediate emotional distress following social threat, these outcomes tend to dissipate within 20 minutes following mild social threats (Sethi et al., 2013; Zwolinski, 2012, 2014). Meta­analytic research has shown that individuals recover from emotional distress outcomes within minutes of playing Cyberball, a virtual ball toss game designed to induce ostracism; still, it is difficult to predict when that recovery occurs (Hartgerink et al., 2015). Given the physical pain­social pain overlap model, could it be the case that physical pain reactivity to social threats would also dissipate within 20 minutes?

Although individuals reflexively respond with a pain signal immediately following mild social threats, such as imagining a future life alone (Bernstein & Claypool, 2011; DeWall & Baumeister, 2006), reliving a past rejection or experiencing Cyberball ostracism (Donate et al., 2017; MacDonald, 2008), the persistence of this physical pain is unclear. This information is necessary to provide insight into how, when, and whether support interventions are needed to mitigate this pain. By intervening (e.g., distraction, prayer, affirmation) in the reflexive or reflective stages of the Temporal Needs Model, it is possible to quickly mitigate

the negative emotional distress impact of short­term ostracism (Williams & Nida, 2022). Knowing how long any physical pain lasts following mild social threats provides a better understanding of when to best apply early interventions for pain exacerbated by social distress. Also, an understanding of the persistence of pain following social threats can help to provide insight into the potential for chronic pain for vulnerable individuals. This is important given that, when individuals experience social threats, victimization, and trauma in childhood and adulthood, they are at an increased risk of numerous pain disorders, including migraine headaches, interstitial cystitis or painful bladder, and fibromyalgia (for a review, see Lumley at al., 2011).

The current article presents two experiments to address how quickly individuals recover from pain following ostracism, and whether it matters if the experience of ostracism is real or imagined. In Experiment 1, participants were asked to imagine being ostracized or included in order to evaluate differences in expected physical pain before, during, and after the social scenario. Participants were also asked to predict the time that it would take for them to feel better, physically, following the social scenario. In Experiment 2, participants were directly ostracized or included to determine pain responses before, immediately after, and 20 minutes after the social experience. In summary, this paper sought to examine whether imagined (Experiment 1) or directly experienced (Experiment 2) ostracism causes immediate and short­term pain reactivity.

Experiment 1

Experiment 1 tested the hypothesis that imagining ostracism would cause more short­term physical pain reactivity than imagining inclusion. Further, ostracism was expected to cause longer recovery from pain reactivity, relative to inclusion. Social threats are so personally distressing that they do not even have to be directly experienced to result in physical pain. Neuroimaging research found that the extent to which individuals perceived other players to be rejecting during Cyberball was correlated with pain intensity, even if the rejection was not actually occurring (Landa et al., 2020). Simply visualizing the social threat is enough to increase distress, even more so than visualizing physical pain. For example, when individuals were asked to relive (i.e., imagine a past social betrayal; Chen et al., 2008) or prelive (i.e., imagine a future betrayal by a romantic partner; Chen & Williams, 2012) an experience of social pain, they found it easier to imagine these socially painful experiences than to imagine physically painful experiences (e.g., physical injury). Regardless of whether participants were imagining ostracism, reliving a prior

experience of ostracism, or concurrently experiencing ostracism via Cyberball, they were so distressed that they had an increase in suicidal thoughts (Chen et al., 2020). Given that it is adaptive for individuals to identify early cues to potential social threat to mitigate distress, they should also be more likely to anticipate or expect more pain to follow social threats. Individuals’ expectations about social threats and associated pain have implications for whether and how quickly they will seek out social support to mitigate any pain outcomes when actual social threats do occur.

Although participants are able to more easily imagine social threat in the immediate short­term, it is unclear how long participants expect that any associated pain will last. Further, it is unclear whether the expected pain changes during or after social threat. For Experiment 1, the first hypothesis was that ostracism would cause more pain from baseline to immediate poststressor, relative to inclusion. Further, participants were hypothesized to have a longer pain recovery following ostracism, relative to inclusion.

Method

Participants

A total of 60 undergraduates attending a liberal arts college participated in the study called “Perceptions of Social Interactions” for partial course credit. Participants’ average age was 18.67 (SD = 0.84) years. A total of 63.3% identified as women (n = 38) and 36.7% (n = 22) identified as men. The breakdown for ethnicity was as follows: European American (48.3%), Hispanic (18.3%), Asian American (13.3%), African American (8.3%), and Other (11.7%). A total of 61.7% were first­year students, 31.7% were sophomores, and the remaining 6.7% were juniors. Participants represented a range of majors but most had not yet declared a major (58.3%).

Procedure

After receiving Institutional Review Board approval from the University of San Diego, the study was administered using Qualtrics, an online data collection system. Qualtrics randomly assigned participants to a Cyberball condition, and participants completed a series of preCyberball and post ­ Cyberball questionnaires. Only those questionnaires described in the current experiment will be reported here. After completing a consent form, participants completed a demographics measure, a validity check, and a baseline pain rating. Next, they read a Cyberball condition scenario and imagined that they were experiencing that social scenario. After, they rated their expected pain during and immediately following the social scenario. Next, they estimated physical recovery to baseline. Finally, participants received a message thanking them for their participation.

Social Threat Manipulation

The Cyberball paradigm is a commonly used experimental approach to study ostracism (Williams & Nida, 2011). Two study conditions were observed. In the ostracism condition, participants were asked to imagine that they were playing an online ball tossing game with two other students at the same university. Participants were told that the game was composed of 30 tosses that would last approximately 4 minutes. Participants were told to imagine that they would receive the ball twice, and thereafter, never again. In the inclusion condition, participants were asked to imagine that they received the ball 33% of the time and that all three players receive the ball an equal number of times. Participants in each condition were also provided with Cyberball image of three “players” that shows the direction of the ball tosses, as a visual model representing each of the two unique conditions. After being modified by the principle investigator, items from the Aversive Impact Index (Williams et al., 2000) were used to measure perceived social threat. To measure rejection, participants’ responded to three items, and an average score was created for these items. The three items (i.e., “I would feel rejected,” “I would feel excluded,” and “I would feel ignored”) were rated using a 5­point scale from 1 (not at all true) to 5 (extremely true) immediately after imagining the game condition. For perception of ball tosses received, participants responded to an open­ended question:

How many ball tosses were you asked to imagine that you received by the two players in the ball toss game above? For example, if you were not asked to imagine receiving any ball tosses, enter 0 below. If you were asked to imagine receiving a total of 15 ball tosses from the two other players, enter 15 below. The total number of ball tosses passed among all players before the game ended was 30 tosses.

This composite rejection score was found to be nonnormally distributed as assessed by a Shapiro­Wilk test (ostracism p = .007, inclusion p < .001). A Mann Whitney test indicated that participants perceived that they would feel less rejected in the inclusion condition (Mdn = 1.00, SD = 0.79) than the ostracism condition (Mdn = 4.00, SD =1.10; U = 58.50, Z = –5.97, p < .001, n = 60). This perceived ball toss proportion score was also found to be nonnormally distributed as assessed by a Shapiro ­ Wilk test (ostracism and inclusion ps < .001). A Mann Whitney test indicated that participants perceived that they would receive more ball tosses in the inclusion condition (Mdn = 10.00, SD = 4.08) as opposed to the ostracism condition ( Mdn = 2.00, SD = 5.65; U = 55.50, Z = –6.24, p < .001, n = 58).

Pain

The Numeric Rating Scale for Pain (NRS­11; McCaffery & Beebe, 1989) was used to evaluate momentary pain. Participants rated their expected pain level on an 11­point scale from 0 (no pain at all) to 10 (pain as bad as it could be). This single item rating has been shown to be both a valid and reliable measure of pain especially in a comparison with the two other most commonly used pain assessments (i.e., Verbal Rating Scale and the Visual Analogue Scale; Williamson & Hogart, 2005). There is strong interscale agreement with the NRS with the Verbal Rating Scale and Visual Analogue Scale (Spearman ρ = .94 for both) and NRS with the Color Analogue Scale (Spearman ρ = .95; Bahreini et al., 2015). Interclass correlation for 24­hour test retest correlation for NRS = .95 (Alghadir et al., 2018). The benefits of an NRS include sensitivity of treatment effects (Paice & Cohen, 1997), and it has been reported to be the most precise, replicable, and predictively valid measure of pain (Jensen et al. Braver, 1986). Participants were asked to imagine pain at baseline (i.e., What number would you give your pain right now?), pain during the game (i.e., What number would you give your pain during the social interaction you just imagined?) and pain immediately after the game (i.e., What number would you give your pain immediately after the social interaction you just imagined?). Next, participants responded to:

After reading about the social situation that you were just asked to imagine, how long do you think that it would take for you to feel the same way, physically, that you felt immediately before starting this study? If you would feel physically the same after as soon as the study ended as you did before the study started, enter 0. If it would take you 10 minutes to feel physically the same way that you felt when you started the study, then enter 10. Enter value for minutes here:__.

Results

Data Exploration and Analyses

Due to violations of normality in the outcome pain data and time to recovery, nonparametric statistics were used. Pain across the three time points were determined to be nonnormally distributed as assessed by Shapiro Wilk tests of normality at baseline (ostracism and inclusion ps < .001); the distribution was only normally distributed for ostracism but not inclusion during Cyberball (ostracism p = .25, inclusion p < .001) and after Cyberball (ostracism p = .16, inclusion p < .001). Time to recovery was shown to be nonnormally distributed by

Zwolinski | Pain Persistence Following Ostracism

a Shapiro Wilk test of normality (ostracism and inclusion ps < .001). Effect sizes (r) were calculated using r = Z /√N (Rosenthal, 1991).

Cyberball

Pain. Wilcoxon signed rank tests found that, for ostracism, there was a change in pain from Baseline to Ostracism (T1–T2; Z = –2.11, p = .04, n = 27, r = –.40), and from Baseline to Immediate Post­Ostracism (T1–T3; Z = –2.60, p = .009, n = 30, r = –.48), but not from Ostracism to Immediate Post­Ostracism (T2–T3; Z = –.82, p = .41, n = 30, r = –.14). There was no change in pain for inclusion for any of the three comparison time points (ps </=.18). Raw scores for each time period are shown in Figure 1. Time to Recovery. Compared to included participants (Mdn = 0.00, SD = 0.61), ostracized participants (Mdn = 0.00, SD = 4.33) expected a higher time to physical recovery from Immediate Post­Cyberball to Baseline (U = 289.50, Z = –2.88, p = .004, n = 59, r = –.37).

Discussion

Experiment 1 examined whether participants’ expectations of ostracism would cause more pain and lead to a longer recovery time following ostracism, relative to inclusion. As expected, participants did imagine that ostracism would cause more physical pain from baseline through immediate post­Cyberball, whereas this trend was not observed for included participants who did not expect a change in pain over time. Relative to included participants, ostracized participants imagined that it would take longer to recover from physical pain following ostracism, although they expected a full recovery to occur within a couple minutes. This study makes a unique contribution to the existing literature by being one of the only published studies to examine expected

pain persistence and recovery immediately following imagined ostracism. Is this imagined tendency for shortterm pain reactivity following ostracism also observed for participants who directly experience ostracism (Experiment 2)? And, if so, how long does the pain sensitivity last (Experiment 2)?

Experiment 2

Experiment 2 tested the hypothesis that directly experiencing ostracism would result in more pain immediately following ostracism, relative to inclusion; also, any pain was expected to revert to baseline within 20 minutes. Ostracism causes immediate imagined pain reactivity (Experiment 1) as well as directly experienced physical pain (Bernstein & Claypool, 2012; Riva et al., 2011, Study 2). Although a few published studies have investigated the emotional pain of ostracism from baseline, poststressor and again at delay assessment points to gain an understanding of the natural history of emotional distress (Zwolinski, 2012, 2014), less published research attention has been given to the natural history (i.e., within 20 minutes) of physical pain beyond baseline to immediate poststressor reactivity to ostracism. Given the physicalsocial pain neural overlap (Eisenberger, 2014), does the short­term natural history of the persistence of physical pain to match the reactivity and recovery trajectory on the short­term natural history of emotional pain after ostracism? Although ostracism is emotionally painful, this emotional pain starts to diminish during ostracism (Wesselmann et al., 2012) reverting to baseline levels within 20 minutes (Sethi et al., 2013; Zwolinski, 2012, 2014) for most individuals.

Based on this work above showing the short­term impact of emotional distress following ostracism, the current study examined whether subjective physical pain outcomes would be similarly short­term in duration? Given that participants tend to underestimate the negative painful effects of ostracism when asked to imagine ostracism (Nordgren et al., 2011) yet participants still rated imagined ostracism to cause immediate pain reactivity after the imagined experience (Experiment 1), Experiment 2 tested the hypothesis that, relative to inclusion, ostracism would cause more immediate reported physical pain, and that this pain would not persist beyond 20 minutes after the social threat.

Method Participants

A total of 99 undergraduates from a liberal arts university participated in the study called “Hormones and Social Relationships” for partial course credit. During the debriefing interview, 14 participants (n = 11 in the Ostracism group) reported some awareness of the study

FIGURE 1

Mean Expected Pain Rating by Cyberball Condition at Baseline, During Cyberball, and Immediately After Cyberball.

hypotheses, so those participants were not included in the final study pool. The final sample included 85 participants (n = 45 in the Ostracism group). In the final sample, participants were traditional age college students (M = 18.80, SD = 1.34). A total of 51.8% (n = 44) identified as women and 48.2% identified as men (n = 41). A total of 60% were European American, 10.6% Hispanic, 12.9% Asian American, 2.4% African American, and 12.9% Other, with 1.2% not reporting. The breakdown for year in college was 71.8% first­year students, 15.3% sophomores, 9.4% juniors, and 1.2% seniors with 2.4% not reporting. The sample included participants from a range of majors; 30.6% of participants had not yet declared a major. A total of 27.1% were in an intimate romantic relationship at the time of the study. Prior to the study, participants were randomly assigned to participate in one of two conditions of Cyberball.

Procedure

Data recruitment and collection began following Institutional Review Board approval from the University of San Diego, and all participants provided written informed consent before participating. For purposes of these study hypotheses, only those assessments described in the current experiment will be reported here. This experiment was completed using paper and pencil assessments. After completing the consent form, participants waited in a private room for 5 minutes. After providing a baseline pain rating, they played Cyberball, and then they completed a second immediate post ­ Cyberball pain rating. Approximately 20 minutes after Cyberball, participants completed a third pain rating. On the day of the study administration, participants completed a partial debriefing interview with the researcher and they were thanked for their participation before being excused. A full review debriefing via email was done at the end of the semester, in order to decrease risk of hypothesis awareness for future participants.

Social Threat Manipulation

The researcher escorted each participant into a private room with a computer to play Cyberball. The participant was given instructions on how to play the game and then was left alone in the room, with the instruction to open the door to the room when they were finished with the game. In the game, the participant was led to believe that they were playing an onscreen version of catch with two other people at the same university who are using computers linked to their own. The other two players were actually virtual confederates. Cyberball begins with one of the “players” throwing the ball to the participant. By mouse­clicking on the appropriate player’s icon, the participant then passes the ball to that player. In the ostracism condition, the participant

received the ball twice, and thereafter, never received it again. In the inclusion condition, the participant received the ball approximately 33% of the time. In both conditions, there was a total of 30 throws, and the game lasted approximately four minutes.

Two manipulation checks were based on the Aversive Impact Index (Williams et al., 2000) to confirm participants’ perception of their rejection status during the Cyberball game. A composite rejection score was created by averaging participants’ responses to three items (i.e., “I was rejected,” “I was ignored,” and “I was excluded”), measured on a 5­point scale used in Experiment 1. This composite score was found to be nonnormally distributed as assessed by KomogorovSmirnov test of normality (ostracism p = .03, inclusion p < .001), so a Mann Whitney nonparametric test was used to evaluate this manipulate check. Ostracized participants (Mdn = 4.00, SD = 0.92) reported higher feelings of the composite rejection variable than included participants (Mdn = 1.67, SD = 0.99), U = 138.50, Z = –6.61, n = 83, p < .001. The second item asked, “Assuming that 33% of the time you would receive the ball if everyone received it equally, what percent of throws did you receive?” This score was found to be normally distributed as assessed by Komogorov­Smirnov’s test for inclusion ( p = .09) but not normally distributed for ostracism (p < .001); a Mann Whitney nonparametric test was used to evaluate this manipulate check. Ostracized participants reported receiving an average of 7.86% ( Mdn = 5.00, SD = 7.93) of the ball tosses, whereas included participants reported receiving an average of 30.04% ( Mdn = 30, SD = 9.06) of the ball tosses, U = 77.50, Z = –7.71, n = 83, p < .001. These values were consistent with reports from other research (Sethi et al., 2013).

Pain

The 11­point NRS for Pain described in Experiment 1 was also used in this study. Participants were asked to rate their pain level (i.e., "What number would you give your pain right now?) at baseline, immediately after the game, and 20 minutes after playing the game.

Results

Data Exploration

Outcome data (i.e., change in pain across the three time points) were determined to be nonnormally distributed as assessed by Komogorov­Smirnov tests of normality (ostracism ps <.001, inclusion ps < .001), so nonparametric tests were used. Effect sizes (r; Rosenthal, 1991) are noted below. Given the data loss due to hypothesis awareness, statistical analyses were also conducted with the full sample with the same overall findings as shown below.

Group by Time Differences in Subjective Pain

Wilcoxon signed rank tests were conducted separately by

Pain Persistence Following Ostracism | Zwolinski

Zwolinski | Pain Persistence Following Ostracism

Cyberball game group to examine reactivity in reported physical pain based on change scores for Baseline to Immediate Post­Game (T1­T2), Immediate Post­Game to 20­Minute Delay (T2­T3) and from Baseline to 20­Minute Delay (T1­T3). Results indicated that ostracism caused more of change in physical pain from T1­T2 (Z = –2.54, p = .01, n = 40, r = –.40) and from T2­T3 (Z = –2.26, p = .02, n = 40, r = –.36) but not from T1­T3 (Z = –.90, p = .36, n = 40, r = –.14). Although the inclusion condition did not cause a change from T1­T2 (Z = –1.29, p = .19, n = 44, r = –.19) or T2­T3 (Z = –.36, p = .71, n = 45, r = –.05), there was a change from T1­T3 (Z = –2.18, p = .03, n = 44, r = –.33). Raw pain scores are shown in Figure 2.

Discussion

As expected, Experiment 2 showed that directly experienced ostracism caused an immediate increase in subjective physical pain but that this reactivity reverted to baseline levels within 20 minutes. This is one of the first published experiments to examine the short­term effect of ostracism on physical pain at multiple assessment points. Just as the immediate emotional distress follows ostracism for most individuals (Williams & Nida, 2011), the distressing effects of ostracism extend to immediate physical pain as well. Similarly, just as Zwolinski (2012, 2014) showed that emotional pain that follows ostracism rebounds at 20­minute delay, this study also shows that physical pain recovery occurs within 20 minutes of ostracism. This temporal tendency with directly experienced pain reactivity to ostracism in Experiment 2 was also observed with imagined pain reactivity in Experiment 1, namely that ostracism, but not inclusion, caused an expected immediate physical pain reflex that reverted to

baseline within minutes. Interestingly, inclusion caused more of a decline in physical pain from baseline to delay but not for any other time comparisons. Could this be explained by the fact that these included participants were experiencing some form of “support” that buffers against any physical pain response as shown in other research (Brown et al., 2003; Master et al., 2009)? One future option is to include a neutral control condition to replace Cyberball inclusion, such as Cybertree (Dvir et al., 2019). In this nonsocial control condition, participants have a similar experience to Cyberball (e.g., seeing similar visual game stimuli, mentally visualizing the game, and using mouse clicks). However, Cybertree does not impact negative mood or needs satisfaction outcomes (Dvir et al., 2019), which could unintentionally impact pain outcomes and skew expected results by providing the social “support” of inclusion.

Overall Discussion

The current paper presented two experiments in order to evaluate the short­term recovery of physical pain following social threats. Experiment 1 addressed ostracized participants’ expectations about pain before, during, and immediately after imagining the social threat. Experiment 2 examined whether any directly experienced increase in pain outcomes to ostracism persisted beyond 20 minutes. The overall summary of these experiments is that social threat physically hurts, regardless of whether the social threat is imagined (Experiment 1) or directly experienced (Experiment 2) but participants recover from this pain quickly (Experiments 1–2). Effects sizes were moderately strong for all effects noted above (Cohen, 1988). These experiments complement and expand on published psychobiological and neurological research on the physical pain outcomes to social threat especially in regards to published data on pain recovery following social threats.

These studies provide support for the physical painsocial pain overlap system, indicating that social threats cause short ­ term physical pain reactivity. Relatedly, these studies also support the early ostracism detection theory that proposes that it is adaptive for individuals to be sensitive to even slightly negative effects of ostracism, such as pain, further supporting the Temporal Needs Model that most pain occurs within the early (reflexive and/or reflective) stages.

Just as acute pain may be seen as an adaptive alarm that alerts the target to the cause of the pain and motivate the target to minimize tissue damage (Lumley et al., 2011), by detecting the slightest cues for ostracism immediately following the social threat, individuals can quickly respond to the immediate threat and increase survival by seeking re ­ inclusion

FIGURE 2

Mean Expected Pain Rating by Cyberball Condition at Baseline, During Cyberball, and Immediately After Cyberball.

(Wesselmann et al., 2012). By attending to the unique social nature of ostracism, individuals are motivated to address the social threat quickly to mitigate any further threat by prioritizing social connection (Ferris et al., 2019). When inclusion immediately follows ostracism, individuals have been shown to display more prosocial behavior and experience higher fundamental needs such as belonging (Leiro & Zwolinski, 2014; Tang & Richardson, 2013). Thus, a quick response by the target, regardless of whether the threat is social or physical, is adaptive at preventing more chronic physical pain from unattended or poorly addressed acute pain. Given the observed reactivity on the neuroendocrine, cardiovascular, and/or immunological systems following acute social stressors (Dickerson, 2011) as well as the link between chronic pain with increased morbidity (Lumley et al., 2011), it is critical to address pain quickly, especially as the impact of chronic pain is mitigated when individuals experience more social engagement and inclusion (Karayannis et al., 2019).

Although Experiments 1 and 2 found that rejection physically hurts, whether it is directly experienced or imagined, this observed pain does not last long. Experiment 2 found that physical pain tends to increase postostracism and then revert to baseline within 20 minutes of the ostracism episode, a temporal recovery trend that has been previously observed work examining the natural history of emotional pain following ostracism (Zwolinski, 2012, 2014). Even Experiment 1 confirmed this tendency for pain to be short­lived and reverts to baseline for rejected participants within minutes of ostracism; still, this is remarkable given that participants tend to underestimate the amount of emotional pain that they expect to experience after social threats of rejection and ostracism (DeWall et al., 2011). Neuroimaging research on several limbic and pain circuits has shown that feelings of rejection continue even after ostracism ends when participants were quickly “re­accepted” in Cyberball (Landa et al., 2020). This similarity between emotional distress and physical pain outcomes to social threat in terms of temporal trajectory reinforce the neurobiological research showing a physical and social pain overlap.

Together, these two experiments include a number of methodological strengths. These include a subjective measure of pain not yet widely used, yet still corroborating other measures of more commonly used pain sensitivity (e.g., pain tolerance and threshold), following social threats. Also this study examined pain at multiple assessment time points (e.g., baseline, during the stressor, poststressor, and 20 minute delay) with multiple methods for feedback (e.g., written and online) regarding the experience of ostracism. These studies also examined different views for evaluating social threat (e.g., imagined

and actual). Using these multiple methods and ways of evaluating the painful effects of social threat, the experiments all tended to confirm that social threat “hurts,” even if it is short­lived, and it is even shorter when social support is provided during the social threat as shown in each respective experiment’s Discussion section.

Limitations and Future Directions

This study also includes a number of limitations. For example, the study samples were specific to undergraduate students from one university, limiting generalizability. The sample size in both studies was also small, further limiting the external validity of these findings and possibly increasing the likelihood of imprecise estimate of the effects shown here. Although another potential limitation was that gender differences were not examined, metaanalytic work has found that gender distribution does not affect outcomes to ostracism (Hartgerink et al., 2015). Next, the pain ratings for Experiments 1–2 were subjective in nature. Although this was considered a strength as noted above, a more comprehensive assessment of pain that includes sensory, affective, and evaluative ratings would provide a more nuanced view of these outcomes. It is possible that asking participants to rate their pain is a bias and they may, in fact, be experiencing other behavioral outcomes not studied here. Another limitation is that participants were relatively healthy and without physical pain history. Although these experiments and literature review presented in this article were focused on nonclinical, healthy populations because experimentally induced pain allows for more controlled manipulation of the social variables of interest (Krahé et al., 2013) in relation to clinical or chronic pain, this decision may have dampened the pain outcomes observed.

Despite these limitations, these two studies makes unique contributions to the existing literature on the effects of social stress on pain. For example, these findings compliment Landa et al. (2020) who showed that both actual and perceived rejection play a role in physical pain modulation immediately following ostracism using fMRI. Experiment 1 is one of the first published studies that examined imagined pain recovery following imagined ostracism. Experiment 2 is one of the first published studies to examine the persistence of physical pain ratings up to 20 minutes following actual ostracism.

Future research questions can validate and expand upon these studies. For example, what individual factors or mechanisms contribute to a differential impact of two types of pain? Does the benefit of social support on physical pain maintain in the short­term or is there is a delay in physical pain reactivity? Do certain personal factors (e.g., cognitive tendencies, prior history of pain) and social context factors (e.g.,

Zwolinski | Pain Persistence Following Ostracism

friend vs. stranger support, severity of social threat) differentially influence immediate and short ­ term pain outcomes? What factors influence whether participants perceive social threats to be painful when imagined vs. directly experienced? At what point does short ­ term pain become more long ­ term or chronic pain and is this exacerbated by social threats? Ultimately, by understanding contextual factors that influence recovery from social threat, researchers can help identify some of the physical pain consequences to social threat in the short­term and minimize any risk of more chronic pain.

Conclusion

In summary, this article reports on two experiments to study short­term pain recovery following social threat. The results indicated several important findings. First, for Experiment 1, imagined ostracism hurt more than inclusion, but the imagined pain tended to be very shortlived. For Experiment 2, directly experienced ostracism caused subjective pain immediately after ostracism, which reverted to baseline within minutes of the social threat. Overall, data from these two experiments provide support of the growing physical pain­social pain overlap literature that social pain does indeed “hurt” but that the pain resolves within 20 minutes.

References

Alghadir, A. H., Anwer, S., Iqbal, A., & Iqbal, Z. A. (2018). Test-retest reliability, validity, and minimum detectable change of visual analog, numerical rating, and verbal rating scales for measurement of osteoarthritic knee pain. Journal of Pain Research, 11, 851–856.

https://doi.org/10.2147/JPR.S158847

Bahreini, M., Jalili, M., & Moradi-Lakeh, M. (2015). A comparison of three selfreport pain scales in adults with acute pain. The Journal of Emergency Medicine, 48(1), 10–18. https://doi.org/10.1016/j.jemermed.2014.07.039

Bernstein, M. J., & Claypool, H. M. (2012a). Social exclusion and pain sensitivity: Why exclusion sometimes hurts and sometimes numbs. Personality and Social Psychology Bulletin, 38(2), 185–196.

https://doi.org/10.1177/0146167211422449

Brown, J. L., Sheffield, D., Leary, M. R., & Robinson, M. E. (2003). Social support and experimental pain. Psychosomatic Medicine, 65, 276–283. https://doi.org/10.1097/01.psy.0000030388.62434.46

Chen, Z., Poon, K.-T., DeWall, C. N., & Jiang, T. (2020). Life lacks meaning without acceptance: Ostracism triggers suicidal thoughts. Journal of Personality and Social Psychology, 119(6), 1423–1443. https://doi.org/10.1037/pspi0000238

Chen, Z., & Williams, K. D. (2012). Imagined future social pain hurts more now than imagined future physical pain. European Journal of Social Psychology, 42(3), 314–317. https://doi.org/10.1002/ejsp.1861

Chen, Z., Williams, K. D., Fitness, J., & Newton, N. C. (2008). When hurt will not heal: Exploring the capacity to relive social and physical pain. Psychological Science, 19(8), 789–795.

https://doi.org/10.1111/j.1467-9280.2008.02158.x

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd edition). Erlbaum.

DeWall, C. N., & Baumeister, R. F. (2006). Alone but feeling no pain: Effects of social exclusion on physical pain tolerance and pain threshold, affective forecasting, and interpersonal empathy. Journal of Personality and Social Psychology, 91(1), 1–15. https://doi.org/10.1037/0022-3514.91.1.1

Dickerson, S. S. (2011). Physiological responses to experiences of social pain. In G. MacDonald & L. Jenson-Campbell (Eds.), Social pain: Neuropsychological

and health implications of loss and exclusion (pp. 79–94). American Psychological Association. https://doi.org/10.1037/12351-000

Donate, A. P. G., Marques, L. M., Lapenta, O. M., Asthana, M. K., Amodio, D., & Boggio, P. S. (2017). Ostracism via virtual chat room—Effects on basic needs, anger and pain. PloS One, 12(9), e0184215.

https://doi.org/10.1371/journal.pone.0184215

Dvir, M., Kelly, J. R., & Williams, K. D. (2019). Is inclusion a valid control for ostracism? Journal of Social Psychology, 159(1), 106–111. https://doi.org/10.1080/00224545.2018.1460301

Eisenberger, N. I. (2012). The pain of social disconnection: Examining the shared neural underpinnings of physical and social pain. Nature Reviews Neuroscience, 13, 421–434.

Eisenberger, N. I. (2014). Social pain and the brain: Controversies, questions, and where to go from here. Annual Review of Psychology, 66, 601–629. https://doi.org/10.1146/annurev-psych-010213-115146

Eisenberger, N. I., & Lieberman, M. D. (2005). Why it hurts to be left out: The neurocognitive overlap between physical and social pain. In K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 109–127). Cambridge University Press. Ferris, L. J., Jetten, J., Hornsey, M. J., & Bastian, B. (2019). Feeling hurt: Revisiting the relationship between social and physical pain. Review of General Psychology, 23(3), 320–335. https://doi.org/10.1177/1089268019857936

Hartgerink, C. H. J., van Beest, I., Wicherts, J. M., & Williams, K. D. (2015). The ordinal effects of ostracism: A meta-analysis of 120 Cyberball studies. PloSONE, 10, e0127002. https://doi.org/10.1371/journal.pone.0127002

Jensen, M. P., Karoly, P., & Braver, S. (1986). The measurement of clinical pain intensity: A comparison of six methods. Pain, 27(1), 117–126. https://doi.org/10.1016/0304-3959(86)90228-9

Karayannis, N. V., Baumann, I., Sturgeon, J. A., Melloh, M., & Mackey, S. C. (2019). The impact of social isolation on pain interference: A longitudinal study. Annals of Behavioral Medicine, 53(1), 65–74. https://doi.org/10.1093/abm/kay017

Krahé, C., Springer, A., Weinman, J. A., & Fotopoulou, A. (2013). The social modulation of pain: Others as predictive signals of salience—A systematic review. Frontiers in Human Neuroscience, 7, 386. https://doi.org/10.3389%2Ffnhum.2013.00386

Landa, A., Fallon, B. A., Wang, Z., Duan, Y., Liu, F., Wager, T. D., Ochsner, K., & Peterson, B. S. (2020). When it hurts even more: The neural dynamics of pain and interpersonal emotions. Journal of Psychosomatic Research, 128, 109881. https://doi.org/10.1016/j.jpsychores.2019.109881

Leiro, J., & Zwolinski, J. (2014). How are prosocial responses to ostracism related to thwarted psychological needs. Psi Chi Journal of Psychological Research, 19(2), 76–84. https://doi.org/10.24839/2164-8204.JN19.2.76

Lumley, M. A., Cohen, J. L., Borszcz, G. S., Cano, A., Radcliffe, A. M., Porter, L. S., Schubiner, H., Keefe, F.J. (2011). Pain and emotion: A biopsychosocial review of recent research. Journal of Clinical Psychology, 67, 942–968. https://doi.org/10.1002%2Fjclp.20816

MacDonald, G. (2008). Use of pain threshold reports to satisfy social needs. Pain Research & Management, 13(4), 309–319. https://doi.org/10.1155/2008/289575

Master, S. L., Eisenberger, N. I., Taylor, S. E., Naliboff, B. D., Shirinyan, D., & Lieberman, M. D. (2009). A picture’s worth: Partner photographs reduce experimentally induced pain. Psychological Science, 20(11), 1316–1318. https://doi.org/10.1111/j.1467-9280.2009.02444.x

McCaffery, M., & Beebe, A. (1993). Pain: Clinical manual for nursing practice V.V. Mosby Company.

Paice, J. A., & Cohen, F. L. (1997). Validity of a verbally administered numeric rating scale to measure cancer pain intensity. Cancer Nursing, 20(2), 88–93. https://doi.org/10.1097/00002820-199704000-00002

Nelson, E. E., & Panksepp, J. (1998). Brain substrates of infant–mother attachment: Contributions of opioids, oxytocin, and norepinephrine. Neuroscience and Biobehavioral Reviews, 22(3), 437–452. https://doi.org/10.1016/s0149-7634(97)00052-3

Nordgren, L. F., Banas, K., & MacDonald, G. (2011). Empathy gaps for social pain: Why people underestimate the pain of social suffering. Journal of Personality and Social Psychology, 100(1), 120–128. https://doi.org/10.1037/a0020938

Riva, P., Wirth, J. H., & Williams, K. D. (2011). The consequences of pain: The social and physical pain overlap on psychological responses. European Journal of Social Psychology, 41(6), 681–687.

https://psycnet.apa.org/doi/10.1002/ejsp.837

Rosenthal, R. (1991). Meta-analytic procedures for social research (Rev. ed.). Sage.

Tang, H. H., & Richardson, R. (2013). Reversing the negative psychological sequelae of exclusion: Inclusion is ameliorative but not protective against the aversive consequences of exclusion. Emotion, 13(1), 139–50. https://doi.org/10.1037/a0029521

Wesselmann, E. D., Nairne, J. S., & Williams, K. D. (2012). An evolutionary social psychological approach to studying the effects of ostracism. Journal of Social, Evolutionary, and Cultural Psychology, 6(3), 309–328. https://doi.org/10.1037/h0099249

Williams, K. D. (2009). Ostracism: A temporal need-threat model. In M. Zanna (Ed.), Advances in experimental social psychology, vol. 41 (pp. 279–314). Academic Press.

Williams, K. D., Cheung, C. K. T., & Choi, W. (2000). Cyberostracism: Effects of being ignored over the internet. Journal of Personality and Social Psychology, 79(5), 748–762. https://doi.org/10.1037//0022-3514.79.5.748

Williams, K. D., & Nida, S. A. (2011). Ostracism: Consequences and coping. Current Directions in Psychological Science, 20(2), 7–-75. https://doi.org/10.1177/0963721411402480

Williams, K. D., & Nida, S. A. (2022). Ostracism and social exclusion: Implications

Pain Persistence Following Ostracism | Zwolinski

for separation, social isolation, and loss. Current Opinion in Psychology, 47, 101353. Advance online publication. https://doi.org/10.1016/j.copsyc.2022.101353

Zwolinski, J. (2012). Psychological and neuroendocrine reactivity to ostracism. Aggressive Behavior, 38(2), 108–125. https://doi.org/10.1002/ab.21411

Zwolinski, J. (2014). Does inclusion after ostracism influence the persistence of affective distress? Group Dynamics: Theory, Research, and Practice, 18(4), 282–301. https://psycnet.apa.org/doi/10.1037/gdn0000012

Author Note. Jennifer Zwolinski https://orcid.org/0000­0002­8239­0190

The author has not known conflict of interest to disclose. This study was supported by University of San Diego Faculty Research Grants. Special thanks to the research students who helped with data collection. Correspondence concerning this article should be addressed to Jennifer Zwolinski, University of San Diego, Department of Psychological Sciences, 5998 Alcala Park, San Diego, CA 92110. Email: jzwolinski@sandiego.edu

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

The Role of Parental Pressure and Warmth in the Relationship Between Parental Involvement, Parental Expectations, and Child Academic Success

Madison T. Weir, Janet P. Trammell*, and Jennifer Harriger* Social Science Division, Seaver College, Pepperdine University

ABSTRACT. The purpose of this study was to investigate both parent and child perceptions of parental involvement, parental expectations, parental pressure, and parental warmth, and the relationship of these variables with GPA. It was hypothesized that each of the independent variables would be positively correlated with GPA, excluding parental pressure, which would be negatively correlated with GPA. It was also hypothesized that the combination of high parental expectations and low parental involvement would have a positive relationship with parental pressure and a negative relationship with parental warmth. Parents (n = 158) of children in 6th–8th grade and 70 6th–8th grade children completed an online survey measuring these factors. Results from the parent survey demonstrated a positive correlation between parental expectations and child GPA (p = .047). Results from the child survey showed a negative correlation between parental pressure and GPA (p = .003), and that parental involvement can predict scores for the child’s sense of parental warmth. These findings provide a better understanding of various factors related to academic success, and they can lead to effective interventions for both parents and the educational system to increase academic achievement in youth.

Keywords: middle school, GPA, academic achievement, parental involvement, parental expectations, parental pressure, parental warmth

According to the U.S. Department of Education, in 2017–18 the adjusted cohort graduation rate for public schools across the nation was 85%, meaning 15% of United States public high school students did not graduate (National Center for Education Statistics, 2020). Although the United States has the highest GDP in the world (GDP, 2019), its students place 38th and 24th out of 71 countries in math and science, respectively (Organisation for Economic Co­operation and Development, 2016). Low educational attainment, specifically not graduating from high school, has been repeatedly linked with poverty (U.S. Census Bureau, 2014), health concerns (Gutiérrez, 2015), and increased risk of incarceration (Hjalmarsson, 2008). Therefore, it is important to investigate the factors related to better educational outcomes for students in the United States.

Parental involvement in a child’s academics and parental expectations have both been positively correlated with academic success (Almroth et al., 2020).

Parental involvement is typically defined as some combination of advising, monitoring, help, and participation of parents in their child’s academic life (Bogenschneider, 1997; Fan & Williams, 2009). Parental expectations refer to the beliefs of parents about how well their child will perform academically in the future or how much education they will attain (Pingault et al., 2015). Parental warmth and parental pressure are two additional factors involved in a child’s educational experience. Parental warmth refers to supportive and encouraging behaviors that are related to a child’s feelings of acceptance and love (Quach et al., 2013). Parental pressure is a type of psychological control that makes the child feel stressed, such as excessive demand to perform well (Quach et al., 2013). Parental warmth, from the child’s perspective, has been shown to be negatively associated with mental health problems of adolescents, while parental pressure has been positively associated with anxiety and depression in this age group (Quach et al., 2013).

Parental Involvement

Parental involvement has been both directly and indirectly positively correlated to academic success in various age groups and cultures. In a group of Mexican American elementary school students, parent monitoring of child grades was positively associated with math outcomes (Gilbert et al., 2017). Another direct positive relationship between child reports of parental involvement, such as parents participating in school programs, sporting events, advising the child on course selection, helping with homework, and monitoring the grades of the child, and the child’s grades was found in American high school students (Bogenschneider, 1997). When home­based involvement and school­based involvement were examined separately in high schoolers, the positive correlation between parental involvement and academic success was stronger with home­based involvement (James et al., 2019). Similarly, parental involvement has been associated with fewer behavioral problems, and fewer behavioral problems have been associated with greater academic achievement, suggesting an indirect positive relationship between parental involvement and academic achievement (Dotterer & Wehrspann, 2015; Hill et al., 2004). Another indirect positive relationship was found between child reports of parental involvement and student academic achievement in a group of 374 American and 451 Chinese 7th and 8th grade students (Cheung & Pomerantz, 2012). Parental involvement was positively associated with the child’s academic motivation that came from wanting to satisfy their parents, which predicted the child’s school engagement, which in turn predicted their academic achievement.

Although most of the research has indicated a positive relationship between parental academic involvement and academic success, some studies have reported conflicting results. For example, among lower educated parents, parental involvement was positively correlated with the child’s academic aspirations but not their achievement (Hill et al., 2004). Additionally, a survey given to 32 parents of middle schoolers in the United States found that there was not a significant relationship between parental involvement, such as help with schoolwork and monitoring of grades, and academic achievement (Lam & Ducreux, 2013).

Parental Expectations

There is also a consensus that parental expectations are related to child academic success, although the analysis of various related studies suggests the link may be indirect, and it may be weaker than the link between parental involvement and academic success (Gordon & Cui, 2012). Such indirect relationships between parental expectations and academic success were found when

parental expectations were positively correlated with self­efficacy, engagement, and intrinsic motivation (Fan & Williams, 2009), and engagement was positively correlated with academic achievement (Cheung & Pomerantz, 2012). Another indirect relationship may exist between parental expectations and academic success because higher parental expectations were related to a child raising their expectations from 7th to 9th grade (Almroth et al., 2020), and adolescent expectations were positively correlated with educational attainment (Weinberg et al., 2019). This supports the expectancy value theory, which claims that one’s self concept of their ability and the subjective value they place on specific tasks are positively correlated with their academic achievement (Lauermann et al., 2017). However, some researchers have casted doubt that there is a relationship between parental expectations and academic achievement. For example, in a longitudinal study of 1,279 children who were assessed in kindergarten, at 12 years old, and from 22–23 years old, the relationship between maternal expectations and educational attainment was significant for boys but not girls (Pingault et al., 2015).

Parental Pressure

Parental pressure has typically been examined in relation to mental health problems, such as anxiety in adolescents, and a positive correlation has been found (Quach et al., 2013). This relationship may be more likely to exist if a child perceives high parental pressure, but does not have high parental involvement. Greenaway et al. (2015) demonstrated a correlation between aspirations exceeding expectations and depression, which may indicate that parents who have high expectations for a child without giving them the tools they need to succeed may contribute to the child having mental health issues, which in turn could affect their school performance (DeSimone, 2010). Similarly, Deb et al. (2015) found that average level performing Indian students who had no tutors, or at most 1 to 2 tutors, were more likely to feel pressure from their parents. Other factors of parental involvement, such as the value the parent transfers to the child about their academics, could affect mental health, because parent value of education was positively correlated with child value of education, which was positively correlated with performance related worry (Lauermann et al., 2017). This is another indication that high expectations with limited parental involvement may have negative mental health effects on children. Additionally, in research utilizing the Academic Pressure subscale from the Inventory of Parental Influence (Campbell, 1994), a direct positive relationship between parental pressure and child anxiety was found in 997 Chinese students ages 16 to 19 (Quach et al., 2013). A significant positive

Factors of Academic Success | Weir, Trammell, and Harriger

relationship between parental pressure and test anxiety was also found in a group of 337 Chinese high school students (Chen, 2012). A direct negative relationship has been found between parental pressure and academic achievement in adolescents (Bean et al., 2003; Levpušček & Zupančič, 2009). However, no relationship was found between parental pressure and academic achievement among a group of American 6th–8th graders (Lam & Ducreux, 2013). This may be because the research measured the parent’s perception of the pressure they placed on their child, but not the child’s perception of the pressure from their parents. Given these conflicting findings, additional research is needed to examine the link between a child’s perception of parental pressure and academic achievement.

Parental Warmth

Despite findings that suggest parental pressure may have negative effects on mental health, research has found a negative relationship between parental warmth and child anxiety and depression, measured by the Warmth/ Affection Subscale of The Parental Acceptance and Rejection Questionnaire (Rohner, 2004; Quach et al., 2013). Because parental support, which is an aspect of warmth, is positively correlated with mastery goals, there may be a relationship between parental warmth and a child’s academic achievement (Régner et al., 2009). When examined in relation to GPA, high levels of warmth alone, as well as the combination of high levels of warmth and high expectations, were positively correlated with academic achievement (Pinquart, 2015). This combination of high levels of parental warmth and high expectations is commonly referred to as representative of an authoritative parenting style (Masud et al., 2016). However, no relationship was found between the authoritative parenting style and academic performance in adolescents (Masud et al., 2016). Additionally, Lam and Ducreux (2013) found no relationship between the support factor of parental warmth and academic achievement. As previously mentioned, this study of American middle schoolers measured parental support from only the parent’s perspective. The child’s perspective of how supported they feel is an important factor to consider in future work

Current Study

Previous research has typically focused on the relationship between parental involvement and academic achievement and parental expectations and academic achievement separately. In terms of parental pressure and parental warmth, past research has commonly examined how these relate to mental health issues, such as depression and anxiety. Additionally, previous

research has focused on high school aged students. However, the current study sought to examine these factors in middle school students, given that their actions and the expectations of their parents may be more influential at a younger age, because the child’s academic behavioral patterns and self ­ concept may not yet be fully developed (Erikson, 1950). Of the few studies that examined middle schoolers, some only surveyed the students (Cheung & Pomerantz, 2012; Dotterer & Wehrspann, 2015) and others only surveyed the parents (Lam & Ducreux, 2013). In both cases, important perspectives were missed. Additionally, very few researchers have examined how parental involvement and parental expectations work together to influence academic success, and none have considered, in one cohesive study, how the additional factors of parental pressure and warmth affect this relationship.

The purpose of the current study was to investigate both parental and child perceptions of parental involvement, parental expectations, parental pressure, and parental warmth, and the relationship of these variables with GPA. Further, this relationship was examined in middle schoolers, as parents may be less likely to base their expectations off their child’s established academic record compared to a high school student who is likely to display more consistent study patterns and grades. Additionally, this study surveyed both parents and children to capture both perspectives. Learning more about the factors related to academic success in middle school children could lead to better interventions, stronger academic outcomes, and increased quality of life for students. Furthermore, it is important that the parents are advised of these factors while their child is still developing their academic behavioral patterns and expectations.

Based on the expectancy ­ value theory, strong parental academic involvement was expected to increase the value that students place on their studies (Lauermann et al. (2017), and high parental expectations was expected to lead to a higher child self ­ concept (Greenaway et al., 2015), which would in turn lead to academic achievement. It was hypothesized that parental involvement, parental expectations, and parental warmth would each be positively correlated with GPA, and parental pressure would be negatively correlated with GPA. Additionally, it was hypothesized that child scores would be positively correlated with parent scores on each of the four individual variables. It was also hypothesized that the combination of high parental expectations and low parental involvement would have a positive relationship with parental pressure, and a negative relationship with parental warmth.

Participants

Method

Participants consisted of 70 middle school children in 6th through 8th grade, and 158 parents of children in 6th–8th grade. Pepperdine University institutional review board approval (21­04­1580) was given before the study began. The researchers analyzed descriptive statistics for child demographic variables including grade level, sex, race and ethnicity, and parent demographic variables including grade level of their child, sex, race, ethnicity, educational level, and income level, as found in Table 1. Additionally, parents and children were asked how online learning during the previous school year affected the child’s grades. Participants were excluded if they answered “no” to a question that asked if they paid attention to the questions and answered honestly. Additionally, if any scale for one of the independent variables was incomplete, the participant was excluded from the study.

Measures

Parental Involvement

Parental involvement was measured using the Parental Involvement Scale, a set of 5 questions that have been used by many researchers (Crosnoe, 2001; Dotterer & Wehrspann, 2016; Steinberg et al., 1992). This scale has a Cronbach’s alpha of .74 (Dotterer & Wehrspann, 2016). In this study, the scale had a Cronbach’s alpha of .61 for the parent survey and .42 for the child survey. Items in this scale measure the level of the parent’s involvement in their child’s academics in areas such as homework help, attending school events and activities, and monitoring the child’s grades. Parental involvement in academics is measured from the perspective of the parent, with questions such as “How often do either you or your child’s other parent help with homework when asked?” and the perspective of the child, where “do either you or your child’s other parent” is substituted with “does one or more of your parents.” The question asking how often the child’s parent watches him or her in sports or other extracurricular activities was inadvertently left out on the child survey. Finally, parental involvement excluding in­person school events and extracurricular activities (Parental Involvement EIP) was examined from both the parent and child perspective to assess involvement from an angle that may be more appropriate given that most students were working remotely during the time of this study.

Parental Expectations

Parental expectations were measured by the highest level of education the parent expects the child to attain, and the highest level of education the child thinks their parents

Weir, Trammell, and Harriger | Factors of Academic Success

expect them to attain. This question was modeled from other forms of measuring parental academic expectations that have been used in the past, such as asking parents whether they have no university expectations or university expectations of their child (Almroth et al., 2020). In this study, the possible expectations were expanded to include an educational background that is less than a high school graduate, a high school graduate, a college graduate, or a postgraduate degree.

TABLE 1

Descriptive Statistics for Child and Parent Reports

Factors of Academic Success | Weir, Trammell, and Harriger

Parental Pressure

Parental pressure was measured by a slightly altered version of the Academic Pressure subscale from the Inventory of Parental Influence (Campbell, 1994). This subscale contains 9 items that include questions about parental pressure from the child’s perspective. It has a Cronbach’s alpha of .76, which suggests moderate reliability. In this study, the subscale had a Cronbach’s alpha of .94 for both the parent and child surveys. A statement indicating parental pressure was presented, such as “My parents are never satisfied with my grades,” and the child responded to this statement on a 5­point scale that ranged from 1 (strongly disagree) to 5 (strongly agree). This scale was also altered by the researchers to include the parent perspective of academic pressure they feel they are putting on their child. For example, the question above was altered to “Either I or my child’s other parent are never satisfied with our child’s grades.” The parent and child scores were examined separately. Higher scores on both the parent and child survey indicate higher levels of parental pressure. The question “I or my child’s other parent do not believe our child when they say they have no homework” was inadvertently not included in the parent survey.

Parental Warmth

Parental warmth was measured by a slightly altered version of the Warmth/Affection subscale of the Parental Acceptance and Rejection Questionnaire (Rohner, 2004). This subscale includes 19 items and was intended to measure the child’s perspective of parental warmth and affection. Although this questionnaire typically asks about mother and father warmth and affection separately, in this study the terms “mother” and “father” were replaced with “one or more parent.” An example question is “One or more parent makes me feel wanted or needed.” This scale was also altered by the researchers to include the parent perspective of warmth felt by the child. The question above was changed to “Either me or my child’s other parent makes our child feel wanted or needed.” The parent and child scores were examined separately. Higher scores indicate higher levels of warmth. The original mother specific Warmth/Affection subscale has a Cronbach’s alpha of .90, and the original father specific subscale has a Cronbach’s alpha of .95 (Rohner, 2004). In this study, the subscale had a Cronbach’s alpha of .90 for both the parent and child surveys. Additionally, the scale was significantly related (p < .001) to its respective validation scale, demonstrating convergent validity (Rohner, 2004). Discriminant validity has also been established (Rohner, 2004).

Academic Success

Academic success was measured on both the parent and child survey by asking the participant to input the child’s

GPA for their previous school year on a 4­point scale. To determine if the changes in administration of classes that occurred due to COVID­19 could have affected the child’s performance, the parent and child were both asked how they believed COVID­19 has affected the child’s grade. Additionally, in case the change in educational instruction during COVID­19 might have resulted in less variability of GPA for middle schoolers, parents and children were asked three other questions to measure academic success: what percent of assignments the child is turning in on time, how much effort the child is demonstrating in school, and how well the child is performing in school.

Procedure

Convenience sampling was used, and parent participants were recruited through MTurk, Instagram, and Facebook posts that contained the survey link. MTurk participants were qualified through a screening survey that asked participants if they had a child under 18 living at their house, then allowed them to select no or yes and specify the grade level of their child. Parents received a link to the parent survey, and they provided informed consent for both themselves and their child. At the end of the parent survey, the parent received a link to the child survey and was instructed to forward it to their child. This link was provided in the final page of the parent survey to ensure that parents had filled out the child’s informed consent before the child survey was accessible. Children gave assent at the beginning of their survey. The child survey utilized a different link than the parent survey, so the child did not see the answers of the parent, and the parent did not see the answers of the child. The child survey explained to the child that their responses would only be seen by the researchers, and that they should complete the survey without their parent present. Parent and child surveys consisted of the Parental Involvement Scale (Crosnoe, 2001; Dotterer & Wehrspann, 2016; Steinberg et al., 1992), parental expectations, the Pressure Subscale (Campbell, 1994), and the Warmth/Affection subscale (Rohner, 2004) in this order. The parent survey contained these measures from a parent perspective, whereas the child survey contained these measures from the perspective of the child. Parent participants reported their child’s GPA as an unweighted score on a 4­point scale. Each survey took between 5 and 15 minutes to complete. Parents received $2.40 for completing this survey through MTurk, and parents recruited through social media and all child participants were entered to win one of four $25 amazon gift cards. The participants were asked to include their email if they wanted to be entered to win.

Results

We investigated the relationship between each of the four independent variables, parental involvement, parental expectations, parental pressure, and parental warmth, and GPA separately for child and parent responses. Additionally, we analyzed the relationship between parental involvement, excluding in­person activities (Involvement EIP), and GPA for child and parent responses. Parent reported GPA ( M = 3.38, SD = 0.56) and child reported GPA ( M = 3.26, SD = 0.61) were normally distributed. There were no curvilinear relationships. The correlations between GPA and parental pressure from the child perspective (r = –.35, p = .003) and between GPA and parental expectations from the parent perspective (r = .19, p = .02) were significant. The correlation between GPA and parental involvement EIP from the child perspective (r = –.24, p = .047) was significant. These findings, along with the other correlational analyses can be found in Table 2, which demonstrate the parent and child results.

A multiple regression analysis, using the enter method, did not demonstrate a statistically significant prediction of a child’s sense of parental pressure based on their perceptions of parental involvement and parental expectations, F(1, 65) = 2.25, p = .12, R2 ADJ = .04. Neither parental involvement (β = –.13, t = –1.10, p = .27) nor parental expectations (β = .23, t = 1.91, p = .06) from the child’s perspective were significant predictors. Similarly, a multiple regression analysis did not demonstrate a statistically significant prediction of a parent’s sense of parental pressure based on their perspective of level of parental involvement and parental expectations, F(2, 155) = 1.47, p = .23, R2 ADJ = .006. Neither parental involvement (β = –.11, t = –1.32, p = .19) nor parental expectations (β = –.07, t = –0.91, p = .36) from the parent perspective were significant predictors. However, a multiple regression analysis demonstrated a statistically significant prediction of child perception of parental warmth based on parental involvement and parental expectations, F(2, 65) = 8.63, p < .001, R2 ADJ = .19, such that parental involvement had a positive relationship with parental warmth. Although level of parental involvement aided in the prediction (β= .44, t = 3.99, p < .001), parental expectations (β = –.17, t = –1.54, p = .13) was not a significant predictor of GPA. Alternatively, a multiple regression analysis did not demonstrate a statistically significant prediction of the parent’s perspective of parental warmth based on level of parental involvement and parental expectations, F(2, 155) = 0.68, p = .51, R2 ADJ = –.004. Neither parental involvement (β = .09, t = 1.14, p = .26) nor parental expectations (β = –.03, t = –0.40, p = .69) were significant predictors.

Discussion

The proposed hypotheses that parental expectations from the parent perspective would be positively correlated to GPA, child perception of parental pressure would be negatively correlated to GPA, and child perception of parental warmth could be predicted by child perception of parental involvement were supported. All other hypotheses were not supported by the data. Although previous work has found positive correlations between parental involvement and academic success (Bogenschneider, 1997), this research did not find significant relationships from the child nor parent perspective. The low reliability of this scale from both the parent and child perspective could be the reason for insignificant results in these analyses. The low reliability and insignificant findings could be explained by the unique learning environment present during this study, which involved many students learning from home. As a result of virtual learning, the questions that ask about parent participation in extracurricular activities and school programs might have influenced total involvement scores and their relationship to GPA, because participation in these activities was not possible for some parents during this time. When the questions relating to in­person activities were removed from the Parental Involvement Scale, a negative correlation was found between parental involvement and GPA from the child’s perspective. However, no significant correlation was found between parental involvement and GPA

TABLE 2

Correlations Between Reported GPA and Parent and Child Measures

Note. Involvement = parental involvement; InvolvementEIP = parental involvement excluding in-person; Expectations = parental expectations; Pressure = parental pressure; Warmth = parental warmth. These categories were measured from the parent and child perspective. ** p < .01 * p < .05

TABLE 3

Regression Coefficients of Parental Involvement and Expectations on Parental Warmth

Note. Involvement ChildP = parental involvement from the child perspective; Expectations ChildP = parental expectations from the child perspective; Involvement ParentP = parental involvement from the parent perspective; Expectations ParentP = parental expectations from the parent perspective.

Weir, Trammell, and Harriger | Factors of Academic Success

from the parent perspective after removing in­person related activities.

Alternatively, a positive correlation was found between parental involvement from the child perspective and the child's sense of parental warmth. Previous research has also found a positive correlation between parental expectations and child academic success, and typically only focused on the parent perspective (Fan & Williams, 2009). This research found a similar positive correlation when looking at the parent perspective, but no correlation when measuring the child perspective. This diversion from previous research may be explained by the fact that parental expectations have typically been assessed from the parent perspective and not from the perspective of middle school children (Fan & Williams, 2009; Pingault et al., 2015). Middle school students may be less aware of the expectations of their parents than older students.

Finally, previous work has found indirect negative relationships between parental pressure and child academic success, typically through child mental health issues (DeSimone, 2010; Greenaway et al., 2015). Additionally, negative correlations have been found between psychological control and academic achievement (Bean et al., 2003) and parental pressure and math grades (Levpušček & Zupančič, 2009). The current study found a direct negative correlation between parental pressure from the child perspective and GPA. Similarly, parental warmth has typically been researched in relation to mental health issues in adolescents, so there is not much previous research to compare with the insignificant findings between parental warmth and GPA in this study (Quach et al., 2013). The one known study that examined the direct relationship between parental warmth and GPA parallels this study in that it found no correlation (Lam & Ducreux, 2013).

Limitations

Certain limitations of this study should be noted. Because the GPAs reported in this research were given during the coronavirus pandemic and during a time when many schools were operating virtually, child grades may be higher or lower than usual. Parents and children were asked how they believed online learning affected their GPA, and 46% of parents reported that their child’s grades were unaffected by online learning and the coronavirus pandemic, 40% claimed their child’s grades were lower, and 23% reported that their child’s grades were higher. Meanwhile, 40% of children reported their grades were unaffected during the pandemic, 40% claimed they were lower, and 20% reported they were higher. Although these results show that many participants felt that the child’s grades were affected

by the pandemic, many also felt they were unaffected. Regardless, these results provide insight into how these various factors of academic success related to GPA during the unique circumstances of the coronavirus pandemic.

Additionally, although participants were asked to enter the unweighted GPAs on a 4­point­scale, some schools may utilize a different scale. This might have affected the results of the study. However, all reported GPAs were within the appropriate range, and participants who denied careful and honest reporting were not included in the study.

Child participants were fewer in number than parent participants. It is possible that, with additional child participants, more findings would have been significant. Another possible limitation of this study was that two questions were inadvertently omitted: the question asking how often the child’s parent watches him or her in sports or other extracurricular activities and the question “I or my child’s other parent do not believe our child when they say they have no homework” on the parental pressure subsection. Because the extracurricular question was unlike most of the other questions in the Parental Involvement Scale as it did not relate to academics, the omission did not affect the relationship between involvement and GPA being examined in this study. Despite the omission of the homework question on the parental pressure subsection, the Cronbach's alpha of this scale was .94, suggesting that the omission did not affect results.

Finally, parents who chose to participate in this study might have had higher levels of parental involvement than the general population. However, because most parent participants were recruited through MTurk, they were likely seeking out various participation opportunities for compensation that were unrelated to their child. Therefore, it is unlikely that their choice to participate indicated higher levels of parental involvement. Similarly, children who chose to complete the survey might have been more conscientious than the general population. This could have resulted in student participants with higher levels of achievement.

Future Direction

Future research should reevaluate parental pressure and warmth and their relationship to GPA, because this was the first study to examine their direct link to GPA. This study uniquely included both parent and child perspectives of variables related to academic success, but future research could expand on this by linking parent and child responses. Although some studies have considered teacher expectations in relation to academic outcomes, this should be done with middle school

students (Lafavor et al., 2019; Weinberg et al., 2019). Expectations and academic achievement of older siblings should also be measured in relation to child expectation and achievement.

The results of this study can contribute to improving academic outcomes for students at an important time in their educational journey. Teaching parents about the negative relationship between parental pressure and child academic success is one way this research can help create a better learning environment for children. This is not only the first study to find a direct correlation between parental pressure and GPA in middle school students, it is also one of the few studies that has examined the relationship of parental expectations and child GPA in this age group. Additionally, although most studies only examined these factors from the perspective of the parent or the child, this research included both perspectives for every variable. Finally, this was the first study to examine how parental involvement and parental expectations work together to predict parental pressure and parental warmth. The findings in this unique study can lead to better interventions for students that will increase academic success and lower the dropout rate in America (Allensworth et al., 2014).

References

Allensworth, E. M., Gwynne, J. A., Moore, P., & Marisa, D. la T. (2014). Looking forward to high school and college: Middle grade indicators of readiness in Chicago Public Schools. Distributed by ERIC Clearinghouse.

Almroth, M., László, K. D., Kosidou, K., & Galanti, M. R. (2020). Individual and familial factors predict formation and improvement of adolescents’ academic expectations: A longitudinal study in Sweden. PLOS ONE, 15(2). https://doi.org/10.1371/journal.pone.0229505

Bean, R. A., Bush, K. R., McKenry, P. C., & Wilson, S. M. (2003). The impact of parental support, behavioral control, and psychological control on the academic achievement and self-esteem of African American and European American adolescents. Journal of Adolescent Research, 18(5), 523–541. https://doi.org/10.1177/0743558403255070

Bogenschneider, K. (1997). Parental involvement in adolescent schooling: A proximal process with transcontextual validity. Journal of Marriage and the Family, 59(3), 718–733. https://doi.org/10.2307/353956

Bureau, U. S. Census. (2014). Census.gov https://www.census.gov/

Campbell, J. R. (1994). Developing cross-cultural/cross-national instruments: Using cross-national methods and procedures. International Journal of Educational Research, 21(7), 675–684. https://doi.org/10.1016/0883-0355(94)90040-x

Chen, H. (2012). Impact of parent’s socioeconomic status on perceived parental pressure and test anxiety among Chinese high school students. International Journal of Psychological Studies, 4(2), 235–245. https://doi.org/10.5539/ijps.v4n2p235

Cheung, C. S.-S., & Pomerantz, E. M. (2012). Why does parents' involvement enhance children's achievement? The role of parent-oriented motivation. Journal of Educational Psychology, 104(3), 820–832. https://doi.org/10.1037/a0027183

Crosnoe, R. (2001). Academic orientation and parental involvement in education during high school. Sociology of Education, 74(3), 210–230. https://doi.org/10.2307/2673275

Deb, S., Strodl, E., & Sun, H. (2015). Academic stress, parental pressure, anxiety and mental health among Indian high school students. International Journal of Psychology and Behavioral Science, 5(1), 26–34.

DeSimone, J. S. (2010). Sadness, suicidality and grades (Paper No. 16239). NBER Working Paper Series. https://doi.org/10.3386/w16239

Dotterer, A. M., & Wehrspann, E. (2015). Parent involvement and academic outcomes among urban adolescents: Examining the role of school engagement. Educational Psychology, 36(4), 812–830. https://doi.org/10.1080/01443410.2015.1099617

Erikson, E. H. (1950). Childhood and society. W W Norton & Co.

Fan, W., & Williams, C. M. (2009). The effects of parental involvement on students’ academic self‐efficacy, engagement and intrinsic motivation. Educational Psychology, 30(1), 53–74. https://doi.org/10.1080/01443410903353302

Gilbert, L. R., Spears Brown, C., & Mistry, R. S. (2017). Latino immigrant parents’ financial stress, depression, and academic involvement predicting child academic success. Psychology in the Schools, 54(9), 1202–1215. https://doi.org/10.1002/pits.22067

Greenaway, K. H., Frye, M., & Cruwys, T. (2015). When aspirations exceed expectations: quixotic hope increases depression among students. PLOS ONE, 10(9). https://doi.org/10.1371/journal.pone.0135477

Gordon, M. S., & Cui, M. (2012). The effect of school-specific parenting processes on academic achievement in adolescence and young adulthood. Family Relations, 61(5), 728–741. https://doi.org/10.1111/j.1741-3729.2012.00733.x

Gutiérrez, O. M. (2015). Contextual poverty, nutrition, and chronic kidney disease. Advances in Chronic Kidney Disease, 22(1), 31–38. https://doi.org/10.1111/j.1741-3729.2012.00733.x

Hill, N. E., Castellino, D. R., Lansford, J. E., Nowlin, P., Dodge, K. A., Bates, J. E., & Pettit, G. S. (2004). Parent academic involvement as related to school behavior, achievement, and aspirations: Demographic variations across adolescence. Child Development, 75(5), 1491–1509. https://doi.org/10.1111/j.1467-8624.2004.00753.x

Hjalmarsson, R. (2008). Criminal justice involvement and high school completion. Journal of Urban Economics, 63(2), 613–630. https://doi.org/10.1016/j.jue.2007.04.003

James, A. G., Rudy, D., & Dotterer, A. (2019). Longitudinal examination of relations between school- and home-based parent involvement and GPA across ethnic groups. Journal of Child and Family Studies, 28(11), 3000–3010. https://doi.org/10.1007/s10826-019-01475-9

Lafavor, T., Langworthy, S. E., Persaud, S., & Kalstabakken, A. W. (2019). The relationship between parent and teacher perceptions and the academic success of homeless youth. Child & Youth Care Forum, 49(3), 449–468. https://doi.org/10.1007/s10566-019-09538-0

Lam, B. T., & Ducreux, E. (2013). Parental influence and academic achievement among middle school students: Parent perspective. Journal of Human Behavior in the Social Environment, 23(5), 579–590. https://doi.org/10.1080/10911359.2013.765823

Lauermann, F., Eccles, J. S., & Pekrun, R. (2017). Why do children worry about their academic achievement? An expectancy-value perspective on elementary students’ worries about their mathematics and reading performance. ZDM, 49(3), 339–354. https://doi.org/10.1007/s11858-017-0832-1

Levpušček, M. P., & Zupančič, M. (2009). Math achievement in early adolescence: The role of parental involvement, teachers' behavior, and students' motivational beliefs about math. The Journal of Early Adolescence, 29(4), 541–570. https://doi.org/10.1177/0272431608324189

Masud, H., Ahmad, M. S., Jan, F. A., & Jamil, A. (2016). Relationship between parenting styles and academic performance of adolescents: Mediating role of self-efficacy. Asia Pacific Education Review, 17(1), 121–131. https://doi.org/10.1007/s12564-015-9413-6

National Center for Education Statistics. (2020, May). The Condition of education. https://nces.ed.gov/programs/coe/indicator_coi.asp

Organisation for Economic Co-operation and Development (OECD). (2016). The development and implementation of PISA: A collaborative effort. PISA 2015 Results (Volume I), 483–489. https://doi.org/10.1787/9789264266490-16-en

Pingault, J. B., Côté, S. M., Petitclerc, A., Vitaro, F., & Tremblay, R. E. (2015). Assessing the independent contribution of maternal educational expectations to children’s educational attainment in early adulthood: A propensity score matching analysis. PLOS ONE, 10(3). https://doi.org/10.1371/journal.pone.0119638

Pinquart, M. (2015). Associations of parenting styles and dimensions with academic achievement in children and adolescents: A meta-analysis. Educational Psychology Review, 28(3), 475–493. https://doi.org/10.1007/s10648-015-9338-y

Quach, A. S., Epstein, N. B., Riley, P. J., Falconier, M. K., & Fang, X. (2013). Effects of parental warmth and academic pressure on anxiety and depression symptoms in Chinese adolescents. Journal of Child and Family Studies, 24(1),

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

Factors of Academic Success | Weir, Trammell, and Harriger

106–116. https://doi.org/10.1007/s10826-013-9818-y

Régner, I., Loose, F., & Dumas, F. (2009). Students’ perceptions of parental and teacher academic involvement: Consequences on achievement goals. European Journal of Psychology of Education, 24(2), 263–277. https://doi.org/10.1007/bf03173016

Rohner, R. P. (2004). Parental Acceptance-Rejection Questionnaire (PARQ): Test Manual.

Steinberg, L., Lamborn, S. D., Dornbusch, S. M., & Darling, N. (1992). Impact of parenting practices on adolescent achievement: Authoritative parenting, school involvement, and encouragement to succeed. Child Development, 63(5), 1266. https://doi.org/10.2307/1131532

Weinberg, D., Stevens, G. W., Finkenauer, C., Brunekreef, B., Smit, H. A., & Wijga, A. H. (2019). The pathways from parental and neighbourhood socioeconomic status to adolescent educational attainment: An examination of the role of cognitive ability, teacher assessment, and

educational expectations. PLOS ONE, 14(5). https://doi.org/10.1371/journal.pone.0216803

World Bank Group. (n.d.). GDP (current US$). The World Bank Data. https://data.worldbank.org/indicator/%20NY.GDP.MKTP.CD

Author Note. Madison T. Weir

https://orcid.org/0000­0003­1004­8414

Janet P. Trammell

https://orcid.org/0000­0002­0304­6974

Jennifer Harriger

https://orcid.org/0000­0003­0975­5912

We have no known conflict of interest to disclose. Correspondence concerning this article should be addressed to Madison Weir, 24255 Pacific Coast Hwy, Malibu, CA 90263. Email: madisontweir@gmail.com

SUMMER 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

131 COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

Minority Stress and Psychological Distress Among Asexual Transgender and Gender Nonconforming Individuals

Jared W. Boot-Haury Department of Clinical Psychology, Michigan School of Psychology

ABSTRACT. The presence of minority stress has been well­documented among members of the transgender and gender nonconforming community, as has the effect of minority stress on their psychological distress. Little attention has been given to transgender and gender nonconforming people who identify as asexual. This study examined the relationships among minority stressors and psychological distress among individuals holding the intersecting identities of transgender and gender nonconforming and asexual. Data were collected from 300 adults using various listservs and social media platforms. It was hypothesized that all minority stressors assessed would predict psychological distress. However, multiple regression results revealed that only vigilance (β = .22, p < .001) and gender expression minority stress (β = .24, p < .001) were significant positive predictors of psychological distress, F(11, 258) = 10.21, p < .001, f 2 = .43; the overall model accounted for approximately 30% ( R 2 = .30) of the total variance in psychological distress. Implications for practice and research are discussed.

Keywords: transgender, gender nonconforming, asexual, minority stress, psychological distress

Despite increasing gender and sexual minority research, relatively little research has been conducted on individuals with intersectional sexual and gender minority identities (Vincent, 2018).

Intersectionality theory deals with understanding people of intersecting marginalized identities and identifying their “unique vulnerability [to] converging systems of domination” (Crenshaw, 1995, p. 367). Those who are asexual and transgender and gender nonconforming (TGNC) hold an intersectional identity as a gender minority and a sexual minority. Asexual people experience divergent amounts of sexual attraction relative to allosexual (nonasexual) individuals or have an absence of sexual attraction (Chasin, 2011); they often face a negative social context, systemic oppression, and devaluation (Decker, 2015; MacInnis & Hodson, 2012). Similarly, transgender, gender nonconforming (TGNC) people struggle with a profoundly negative social context

Diversity badge earned for conducting research focusing on aspects of diversity. Open Data and Open Materials badges earned for transparent research practices. Data and materials are available at https://osf.io/75fys/

through stigma and discrimination connected to their gender identity (Bockting et al., 2013; James et al., 2016). These experiences are consistent with minority stress frameworks that show higher stigmatizing experiences among sexual and gender minority individuals (Meyer, 2003; Testa et al., 2015). Meyer’s (2003) minority stress theory suggested that added psychological stress could lead to psychological dysfunction. Minority stress can be distal and can occur through various external events, such as societal stigma, prejudice, and violence. Minority stress can also be proximal and occur through various internal events, such as identity concealment and hypervigilance. When external stressors are experienced, proximal stress via processes, such as internalization via rumination or isolation, can lead to mental health issues (Sarno et al., 2020). Although empirical findings supporting minority stress were initially derived from studies on gay men, the model was expanded to include lesbians and gay men

(Meyer, 1995), bisexual men and women (Meyer, 2003), and LGB people of color (Meyer, 2010). Hendricks and Testa (2012) hypothesized that the minority stress model could also be applied to clinical work with TGNC clients. Testa et al.’s (2015) research expanded Meyer’s minority stress model to account for gender minority stress and resilience. People with intersecting sexual and gender minority identities (i.e., asexual TGNC people) may have a cumulative and significantly higher rate of minority stress than cisgender sexual minority (e.g., asexual) people (Williams et al., 2020).

Minority Stress and Psychological Distress

Asexual TGNC people may be especially susceptible to stigma compared to the overall LGBTQIA community (McInroy et al., 2020). In addition to the stigma experienced by the LGBTQIA community from people outside of that community, stigmatizing experiences within the LGBTQIA community are not uncommon (Cuthbert, 2019). Often, asexual TGNC people feel reticent to disclose their combined gender and sexual minority experience to the broader LGBTQIA community because of the discrimination they may experience within that community. Essentially, minority stressors related to TGNC minority stress may be compounded when coupled with asexual minority stressors (Gamarel et al., 2014; Gupta, 2015; Rachlin, 2019). Although minority stress is known to be linked to poor mental health, there is a shortage of literature investigating the intersection of asexual TGNC adults' relationships between minority stress and mental health. Only one study could be located, which was explicitly focused on young adults and adolescents (McInroy et al., 2020). In that study, asexual TGNC participants had higher rates of proximal stressors (e.g., self­perceived stress and internalized LGBTQIA­phobia) than their allosexual peers. Other studies have also shown that internalized LGB­phobia (Herek et al., 2009), LGBT­phobia (Newcomb & Mustanski, 2010), and LGBTQIA­phobia (Puckett et al., 2019), are linked to mental health difficulties. The asexual young adult and adolescent participants in the study by McInroy et al. (2020) also had poorer mental health than their allosexual peers (e.g., higher rates of depressive, anxious, and somatic symptoms and higher suicidal ideation). Individuals with multiple stigmatized identities may experience various forms of stigma and discrimination, the effects of which cannot be understood by studying each identity in isolation (Mink et al., 2014). Because the only existing quantitative research on asexual TGNC individuals showed that proximal stressors might be higher than distal stressors relative to allosexual individuals and because of the importance of using an intersectional

approach, it was important to assess proximal and distal stressors in this study that could be especially salient in the asexual TGNC community based on the literature. Isolation, vigilance, harassment & discrimination, gender expression minority stress, and violence seem to be the most salient minority stressors based on the literature for asexual (Brotto et al., 2010; Dawson et al., 2018; MacInnis & Hodson, 2012; Vares, 2017), TGNC (Bockting et al., 2013; Grant et al., 2011; Hendricks & Testa, 2012; James et al., 2016), and asexual TGNC individuals (Cuthbert, 2019; McInroy et al., 2020).

The Current Study

Informed by the concepts of intersectionality and minority stress theory, and based on previous sexual and gender minority research (Bockting et al., 2013; Brennan et al., 2017; Hendricks & Testa, 2012; Lefevor et al., 2019; McInroy et al., 2020), each minority stressor (gender expression, proximal [vigilance and isolation], and distal [harassment & discrimination and victimization] minority stressors) was hypothesized to be positively associated with psychological distress in asexual TGNC individuals.

Methods

Study Design and Procedures

The Michigan School of Psychology Institutional Review Board reviewed and approved the study materials and procedures (Protocol #210901). A convenience sample of participants was collected during the height of the omicron surge of the COVID­19 pandemic. Due to the sensitive nature of sexual and gender minority research, previous research has highlighted the importance of confidentiality associated with recording internet­based survey responses at one point in time (Woodford et al., 2018). Therefore, online surveys (hosted by Qualtrics) were conducted. Participants were recruited from email listservs of LGBTQIA community centers and online platforms, such as Facebook, Reddit, and Asexual Visibility and Education Network web forums. Upon accessing the online survey, participants were provided with informed consent information and asked if they identified as (a) asexual, (b) TGNC, (c) lived in the United States, (d) were fluent in English, and (e) were 18 years of age or older. Participants who affirmed that they met the study criteria and consented to participate were directed to a survey introduction and were then prompted to respond to the survey items. Across the entire dataset, only 1% had missing data on the psychological distress and minority stress questionnaires. Of that 1%, the percentage of missing data was 53.12%, beyond the threshold value of 10% of missing data identified by Bennet et al. (2001). Therefore, these participants were excluded from the study.

Boot-Haury | Psychological Distress of Asexual TGNC People

Participants

A total of 300 participants met the inclusion criteria and completed the survey between October 3, 2021, and October 24, 2021. Of those who participated, 48 were transgender men (16.2%), 51 were transgender women (17.2%), and 198 were gender nonconforming individuals (66.7%). Participants ranged from 18 to 70 years old (M = 26.27, SD = 7.86). Two hundred thirty participants (77.4%) identified as White or European American, 15 (5.1%) identified as Hispanic, Latino, or Spanish origin, 14 (4.7%) identified as Black or African American, 10 (3.4%) identified as Asian, four (1.3%) identified as American Indian or Alaska Native, two (0.7%) identified as Middle Eastern or North African, and 22 (7.4%) identified as another race or ethnicity. One hundred fifty­three participants (51.5%) had incomes of less than $25,000, 40 (13.5%) had incomes of $25,000–$34,999, 37 (12.5%) had incomes of $35,000–$49,999, 33 (11.1%) had incomes of $50,000–$74,999, 11 (3.7%) had incomes of $75,000–$99,999, seven (2.4%) had incomes of $100,000–$149,999, and five (1.7%) had incomes of $150,000 or more. Six participants (2.0%) indicated an education level of some high school, 57 (19.2%) indicated an education level of high school, 18 (6.1%) indicated an education level of associate’s degree/ trade school, 103 (34.7%) indicated an education level of some college, 74 (24.9%) indicated an education level of bachelor’s degree, 34 (11.4%) indicated an education level of master’s degree, and four (1.3%) indicated an education level of a doctoral degree. Fifty­three participants (17.8%) had a feminine gender expression, 61 (20.5%) had a masculine gender expression, 92 (31.0%) had an androgynous gender expression, 36 (12.1%) had a genderqueer/gender nonconforming gender expression, 29 (9.8%) had a gender fluid gender expression, and 25 (8.4%) reported another gender expression not listed. Twenty­five participants (8.4%) reported a homoromantic orientation, six (2.0%) reported a heteroromantic orientation, 121 (40.7%) reported an aromantic orientation, 36 (12.1%) reported a biromantic orientation, 54 (18.2%) reported a panromantic orientation, and 50 (16.8%) reported a romantic orientation not listed. Most participants (68.0%) met Yule et al.’s (2015) cut­off score for the Asexual Identification Scale (AIS).

Measures

Asexual Identification Scale

Yule et al. (2015) developed the AIS to assess asexual identity. The AIS consists of 12 items that reflect one dimension (traits of asexuality). An example item is “I lack interest in sexual activity.” Each item is rated on a 5­point Likert scale. Higher scores suggest experiences more characteristic of people who identify themselves

as asexual. A total score is used with a cut­off score of 40, with scores above 40 capturing 93% of self­identified asexual participants and scores below the cut­off capturing 95% of self­identified allosexual participants. The AIS included in the present study had good reliability with the study sample (M = 42.97, SD = 6.93; α = .74).

Support for the validity of the AIS is evidenced by an independent­samples t test showing scores on the AIS­12, which differed significantly between the groups (p < .001). Convergent validity was demonstrated with respective moderate and weak correlations with Spector et al.’s (1996) Sexual Desire Inventory, which has two subscales: Dyadic and Solitary (r = –.57, –.19). Discriminant validity was evidenced by a nonsignificant correlation with Bernstein et al.’s (1994) Childhood Trauma Questionnaire, which was used because negative sexual experiences may be construed as indicative of asexuality.

The AIS was administered as a post­hoc validation of asexual identity. Because heterogeneity within the asexual community exists and competing definitions for asexuality complicate research on this community, current researchers recommend against allowing selfidentification and operationalizing the identity (Van Houdenhove et al., 2017). Prause and Graham (2007) similarly cautioned against Bogaert’s (2004) research model that included a single self­identification question about asexuality, which has questionable validity. They criticized the lack of measures of desire, attraction, and arousability and cautioned that the lack of operationalization might hinder the collection of representative samples. Therefore, Yule et al.’s (2015) Asexuality Identification Scale is recommended for researching this population and has been used in numerous studies (Brotto et al., 2015; Yule et al., 2016; Yule et al., 2017). A cut­off score greater than or equal to 40 for participants delineates who is considered asexual. Although this scale operationalizes asexuality and may improve research, the lack of allowing participants to self­identify moves in a direction opposite to that of the American Psychological Association (APA) and the field of psychology (APA, 2013; APA, 2021). Therefore, this scale was used after the survey to validate self­identification, rather than as a strict inclusion measure.

Kessler et al. (2003) Psychological Distress Scale

Kessler et al. (2003) developed the Kessler Psychological Distress Scale (K10) to assess nonspecific psychological distress among adults. An example item includes “During the last 30 days, how often did you feel tired out for no good reason?” The K10 consists of 10 items reflecting how frequently individuals have experienced symptoms of psychological distress in the past 30 days.

Each item was rated on a 5­point scale ranging from 1 (none of the time) to 5 (all of the time). Higher scores indicate experiences of people with more psychological distress and diagnosable mental illnesses. Cronbach’s alpha for the full­scale K10 was excellent and ranged from α = .92 to .93 in initial studies (Kessler et al., 2002). Researchers assessing psychological distress in American and Australian transgender individuals reported values of α = .93 and .94, respectively (Bariola et al., 2015; Tan et al., 2020). The K10 included in the present study had acceptable reliability with the study sample (M = 27.89, SD = 6.93; α = .75).

Support for the validity of the K10 has been found with adequate prediction of affective disorders from the fourth edition of the Diagnostic and Statistical Manual using Sheehan et al.’s (1998) Mini­International Neuropsychiatric Interview (MINI; Hides et al., 2007).

Daily Heterosexist Experiences Questionnaire

Balsam et al. (2013) developed the Daily Heterosexist Experiences Questionnaire (DHEQ) to assess minority stress among lesbian, gay, bisexual, and transgender (LGBT) adults. The DHEQ includes 50 items reflecting nine dimensions: vigilance (six items), harassment & discrimination (six items), gender expression (six items), parenting (six items), victimization (four items), family of origin (six items), vicarious trauma (six items), isolation (four items), and HIV/AIDS (six items). An example item begins with the question, “How much has this problem distressed or bothered you during the past 12 months?” and includes “difficulty finding LGBT friends.” Each item is rated on a 6­point scale ranging from 0 (did not happen/not applicable to me) to 5 (it happened, and it bothered me extremely). Higher scores indicate higher levels of minority stress. The scores on the nine subscales had internal reliabilities of

α = .86, .85, .86, .83, .87, .79, .82, .76, and .79, respectively. The internal reliability of the DHEQ full ­ scale was

α = .92. Staples et al. (2017), in a study focused exclusively

TABLE 1

Descriptive Statistics

on transgender individuals, reported internal reliability for the Harassment & Discrimination and Victimization subscales of α = .76 and .87, respectively. Because Balsam et al. (2013) stated that researchers can select only the subscales relevant to their research purposes, based on the stressors faced by the asexual, TGNC, and asexual TGNC community, only the scales mentioned were used in the present study. The DHEQ subscales used in the present study had acceptable reliability with the study sample for (a) harassment and discrimination ( M = 1.01, SD = 1.11; α = .79), (b) victimization (M = 0.22, SD = 0.74; α = .75), (c) vigilance (M = 1.66, SD = 1.07; α = .76), and (d) gender expression minority stress ( M = 1.69, SD = 1.12; α = .72). The isolation subscale used in the present study had questionable reliability with the study sample (M = 1.95, SD = 1.15; α = .65).

Support for the validity of the DHEQ is evidenced by moderate correlations with assessments of psychological distress (e.g., depression, anxiety, posttraumatic stress disorder, and perceived stress). These were measured using Andresen et al.’s (1994) 10­item Center for Epidemiological Studies Depression Scale, Kroenke et al.’s (2007) Patient Health Questionnaire­Anxiety, Weathers et al.’s (1993) PTSD Checklist Civilian Version, and Cohen et al.’s (1983) Perceived Stress Scale­Short Form (PSS­SF). Similarly, concurrent validity was supported. The DHEQ scores correlated moderately with the two general LGB discrimination queries from Mohr and Fassinger’s (2000) Outness Inventory. One of these items assessed the interference of homophobia in living a rewarding and productive life. The other item was related to how different people think their lives would be if they did not have to deal with the challenges associated with LGBT identity.

Results

Raw survey data was uploaded into IBM SPSS (Version 28.0) for Mac. The primary analyses used in the present study were multiple regression analyses. All assumptions for linear regression were met, and in addition to planned analyses, demographic variables (age, education, race, gender expression, romantic orientation, and SES) were analyzed as potential covariates. Summary statistics are presented in Tables 1 and 2.

Minority Stressors and Psychological Distress

The present study assessed whether proximal, distal, and gender expression minority stress predict psychological distress. Multiple regression analysis was conducted to test whether minority stressors predicted psychological distress (see Table 3). Specifically, vigilance was a significant positive predictor of psychological distress

(β = 0.22, p < .001). Gender expression minority stress was also a significant positive predictor of psychological distress (β = 0.24, p < .001). As illustrated in Table 3, none of the other minority stressors significantly predicted psychological distress.

Discussion

The author of the present study predicted that all types of minority stressors assessed would positively predict psychological distress. This was partially supported; the results showed that participants experienced more psychological distress when they faced more vigilance and gender expression minority stress. The results of this study support previous research on vigilance and gender expression minority stress’s relationship to psychological distress in TGNC and asexual individuals (Bockting et al., 2013; Brennan et al., 2017; Hendricks & Testa, 2012; Lefevor et al., 2019; McInroy et al., 2020). A possible explanation for this is that vigilance and gender expression minority stress could be the most salient minority stressors related to psychological distress in the asexual TGNC community. This possibility is congruent with minority stress and gender minority stress theory, which suggests that internalized or proximal stressors more directly predict psychological distress than distal stressors because proximal stressors are internalized due to chronic external minority stress (Meyer, 2003; Meyer et al., 2017; Testa et al., 2015).

The salience of vigilance and gender expression minority stress in the current study could be related to previous research on vicarious discrimination/trauma. Research has indicated that vicarious trauma in gender

and sexual minority communities affects vigilance even when victimization and harassment are not personally experienced because individuals exhibit heightened awareness and a sense of vulnerability associated with having identities aligned with individuals who are victims of hate crimes (Bell & Perry, 2015; Noelle, 2002). Similarly, Gonzalez et al. (2018) found an increase in vigilance among gender and sexual minority people after the 2016 presidential election and subsequent targeting of gender and sexual minority communities, particularly among TGNC individuals. Thus, asexual TGNC individuals may face particularly challenging struggles with vigilance in the post­2016 sociopolitical climate. Research has indicated that vigilance may include concerns related to an increase in politicians attempting to restrict healthcare access and public accommodations based on gender identity and eliminate existing civil protections related to sexual and gender identity (Bockting et al., 2020; Frederick et al., 2022). Relatedly, the media has become increasingly antagonistic and has propagated transphobic messaging about TGNC individuals, as illustrated by research on anxiety related to ballot referenda on gender­based civil rights protections (Horne et al., 2022; Hughto et al., 2021). The current sociopolitical climate could have caused significant gender expression minority stress in the present study. When experiencing greater antagonism in the sociopolitical sphere, TGNC individuals may respond with more vigilance as a protective response to potential discrimination or harm (Nadal et al., 2014). This is consistent with research that has shown that, in some individuals, vigilance is used as a coping strategy

Among Variables

Pearson Bivariate Correlations

SUMMER 2023

(Alessi & Martin, 2017; Straussner & Calnan, 2014).

A predictive relationship between isolation and psychological distress was not supported. Prior research has indicated that isolation is heightened among asexual (Brotto et al., 2010; Dawson et al., 2018) and TGNC (Bockting et al., 2013; Hendricks & Testa, 2012) individuals. Individuals in the current study were recruited from affirming online spaces for asexual­ or TGNC­identified individuals and could have felt a sense of community with other asexual or TGNC individuals, which has been shown to alleviate feelings of isolation (Brotto & Yule, 2009). Research has supported the idea that LGBTQIA individuals have primarily socialized online during the COVID ­ 19 pandemic because of physical distancing (Scroggs et al., 2020). Previous research has indicated that finding community in affirming LGBTQIA spaces among gender and sexual minorities contributes to a lack of isolation, even if the identified community is solely online (Brotto & Yule, 2009; Brotto et al., 2010; Fredriksen ­ Goldsen et al., 2013; Gonzalez et al., 2012; Gupta, 2018; Riggle et al., 2008, 2011; Rostosky et al., 2010; Trujillo et al., 2016). Predictive relationships between harassment and discrimination, victimization, and psychological distress were also not supported. Harassment and

of Associations Between Minority Stressors and

discrimination and victimization might not have been significant because the sample was recruited in affirming online social media spaces (e.g., Reddit and Facebook groups specific to LGBTQIA communities). Because the DHEQ only asks about experiences over the past 12 months, participants who have been social distancing during the pandemic might not have had elevated distal or external minority stressors. Despite not currently experiencing significant distal stressors, participants might still have had heightened proximal stress from previous distal stressor experiences from before the pandemic and still struggle with isolation, internalized self ­ stigma, and anticipation of future negative distal stressor experiences outside of their affirming environments. Research has indicated that the COVID­19 pandemic and social distancing may contribute to increased online socialization and, for some LGBTQIA individuals, safety from harassment and discrimination and victimization (Fish et al., 2020; Scroggs et al., 2020). This is similar to earlier research indicating that online communication may be a source of resilience because resources, information, positive space for finding role models, navigating identity, and self­expression are more available to TGNC individuals in online communities (McInroy & Craig, 2015; Raun, 2015). Although general online socialization poses risks for harassment and discrimination and victimization minority stress, the benefits of online socialization in affirming spaces offering safety and support might have resulted in comparably fewer opportunities for recent harassment & discrimination and victimization (Craig et al., 2015; McInroy & Craig, 2018; McInroy, 2019). The results of the present study suggest that future research may need to assess how much socialization occurs online and what the quality of that socialization is like, especially as the COVID­19 pandemic subsides.

Limitations

Note N = 258. Race was dummy coded such that 0 indicates White and 1 indicates other races. There were not enough people in each subcategory to code separately. Romantic orientation is dummy coded such that 0 indicates aromantic and 1 indicates other romantic orientations. There were not enough people in each subcategory to code separately.

* p < .05. ** p < .01. *** p < .001.

This study has several limitations that could inform future research on intersectional asexual TGNC communities. First, there were unequal groups based on gender identity, with gender nonconforming individuals accounting for over half of the sample. Transgender men and transgender women each accounted for less than 25% of the sample, limiting the generalizability of the present study. Another limitation related to generalizability was the racial and ethnic homogeneity of the sample. Although efforts were made to recruit diverse participants, the sample was overwhelmingly White, accounting for approximately 75% of the sample, mirroring previous research, as indicated by Guz et al.’s (2022) scoping review of asexual community research. Racial and ethnic homogeneity mean that the findings of this study may not

represent the overall asexual TGNC population. Another limitation of this study was the lack of measures validated for use with asexual TGNC individuals. In addition to the lack of scale validation with members of the asexual community, self ­ report measures are susceptible to construct validity problems and response bias, which could have affected the results of this study (Lance & Vandenberg, 2009). Also, there is the possibility that COVID­19 could act as a confounding variable affecting the study results because of history effects. Although collecting data during this unique time could add to the literature base, it is a limitation of this study because no questionnaire was used to assess and control the impact of COVID­19­related distress. When the data for this study were collected, many places were still shut down, and people were physically distancing from one another and socializing more online than in person. The pandemic and significant social discord affecting LGBTQIA individuals could also have impacted recruitment and affected the characteristics of participants who opted to participate in the study. External variables, such as COVID­19 and antagonistic political and media messaging, are why researchers like Bleckmann et al. (2022) have called for longitudinal research on sexual and gender minorities to account for potential history effects. Regardless of the pandemic, future research on this intersectional community should assess whether socialization occurs predominantly online or in person, and the frequency of each type of socialization.

Directions for Future Research

The present study elucidates more understanding of the effects of minority stress on psychological distress among asexual TGNC individuals. However, the use of other research methods to study this community may provide a more nuanced understanding. The sample for the current study was a cross­sectional convenience sample collected online, which provides limited information regarding the causality of the observed relationships. A repeated­measures longitudinal approach could help mitigate limitations associated with cross ­ sectional research, such as the history effects associated with the COVID­19 pandemic.

Future research may also benefit from better measurement of asexuality. Better measurement of asexuality could include collecting larger sample sizes and breaking down the analysis by subgroups within the asexual spectrum because of differential experiences within the asexual community. For instance, demisexual individuals may be more likely to be perceived as allosexual (Clark et al., 2022; Kelleher et al., 2023). In addition, for the purpose of this study, those who self­identified as asexual were included regardless of

their AIS score. The AIS has historically been used as a measure with a cut­off score to identify those who are asexual. However, only 68% of the sample in this study met the cut­off score used by Yule et al. (2015) to identify asexual individuals. Previous research has indicated that more than 90% of self­identified asexual individuals meet the AIS cut­off score (Yule et al., 2015; Zheng & Su, 2018). This indicates that the scale may not be as useful as a measure of the asexual TGNC population as it is for the general asexual community. Post­hoc Kruskal­Wallis tests indicated that issues with the AIS scale might not have affected the results of the present study. Psychological distress, χ2(1) = 0.40, p = .53, η2 = –.009; harassment & discrimination, χ2(1) = 2.11, p = .15, η 2 = –.003; victimization, χ 2 (1) = 0.66, p = .42, η 2 = –.008; isolation, χ 2(1) = 0.30, p = .58, η2 = –.009; vigilance, χ2(1) = 0.29, p = .59, η2 = –.009; and gender expression minority stress, χ2(1) = 0.56, p = .45, η2 = –.008, did not significantly differ based on whether participants met the AIS cut­off score. However, refinement of this scale for the asexual TGNC community remains a possible direction for future research.

Additionally, future research on asexual TGNC minority stress and psychological distress may benefit from incorporating qualitative data collected through interviews. Prior qualitative research on the asexual community has yielded information about the community’s unique experiences (Prause & Graham, 2007; Scherrer, 2008), which has informed subsequent quantitative research. Narrative exploration of how an asexual TGNC person’s minority stress fluctuates over time and how that influences psychological distress may provide additional information and themes that explain the experiences of psychological distress and how to mitigate that distress. This could also reveal other factors that have the potential to moderate the relationship between minority stress and psychological distress.

Despite the limitations of the present study and the factors that future researchers may need to consider, the findings of this study suggest that experiences of vigilance and gender expression minority stress negatively impact the mental health of asexual TGNC individuals. The elevated psychological distress consistent with a moderate mental disorder (M = 27.89, SD = 6.93) identified among participants in the present study, using the K10 cutoff scores by Andrews and Slade (2001), underscores the importance of mitigating minority stress among asexual TGNC individuals.

Conclusion

The present study also helps researchers and clinicians better understand how the intersectional asexual TGNC populations self­report gender expression minority stress,

Boot-Haury | Psychological Distress of Asexual TGNC People

proximal (isolation and vigilance), and distal stressors (harassment and discrimination and victimization), as well as whether these variables predict psychological distress. The results indicated that vigilance and gender expression minority stress were significant positive predictors of psychological distress. The results of this study can help clinicians better understand that vigilance and gender expression minority stress could be particularly salient in this intersectional community. Knowing this could assist clinicians in tailoring case conceptualization and interventions for psychological distress in asexual TGNC clients. Individualized conceptualization and intervention may improve psychological functioning, treatment outcomes, and quality of life among asexual TGNC individuals. This study may aid other researchers, clinicians, policymakers, health professional training curriculum leaders, and LGBTQIA community leaders in focusing on specific minority stressors affecting psychological distress among asexual TGNC individuals.

References

Alessi, E. J., & Martin, J. I. (2017). Intersection of trauma and identity. In K. L. Eckstrand & J. Potter (Eds.), Trauma, resilience, and health promotion in LGBT patients: What every healthcare provider should know (pp. 3–14). Springer.

American Psychological Association. (2013). Guidelines for psychological practice with transgender and gender nonconforming people.

https://www.apa.org/practice/guidelines/transgender.pdf

American Psychological Association. (2021). Guidelines for psychological practice with sexual minority persons.

https://www.apa.org/about/policy/psychological-sexual-minority-persons.pdf

Andresen, E. M., Malmgren, J. A., Carter, W. B., & Patrick, D. L. (1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D. American Journal of Preventive Medicine, 10(2), 77–84.

https://doi.org/10.1016/s0749-3797(18)30622-6

Balsam, K. F., Beadnell, B., & Molina, Y. (2013). The daily heterosexist experiences questionnaire. Measurement and Evaluation in Counseling and Development, 46(1), 3–25. https://doi.org/10.1177/0748175612449743

Bariola, E., Lyons, A., Leonard, W., Pitts, M., Badcock, P., & Couch, M. (2015). Demographic and psychosocial factors associated with psychological distress and resilience among transgender individuals. American Journal of Public Health, 105(10), 2108–2116. https://doi.org/10.2105/ajph.2015.302763

Bell, J. G., & Perry, B. (2015). Outside looking in: The community impacts of anti-lesbian, gay, and bisexual hate crime. Journal of Homosexuality, 62(1), 98–120. https://doi.org/10.1080/00918369.2014.957133

Bennett, D. A. (2001). How can I deal with missing data in my study? Australian and New Zealand Journal of Public Health, 25(5), 464–469. https://doi.org/10.1111/j.1467-842x.2001.tb00294.x

Bernstein, D. P., Fink, L., Handelsman, L., Foote, J., Lovejoy, M., Wenzel, K., Sapareto, E., & Ruggiero, J. (1994). Initial reliability and validity of a new retrospective measure of child abuse and neglect. American Journal of Psychiatry, 151(8), 1132–1136. https://doi.org/10.1176/ajp.151.8.1132

Bleckmann, C., Leyendecker, B., & Busch, J. (2022). Sexual and gender minorities facing the coronavirus pandemic: A systematic review of the distinctive psychosocial and health-related impact. Journal of Homosexuality, 1–22. https://doi.org/10.1080/00918369.2022.2074335

Bockting, W., Barucco, R., LeBlanc, A., Singh, A., Mellman, W., Dolezal, C., & Ehrhardt, A. (2020). Sociopolitical change and transgender people’s perceptions of vulnerability and resilience. Sexuality Research and Social Policy, 17(1), 162–174. https://doi.org/10.1007/s13178-019-00381-5

Bockting, W. O., Miner, M. H., Swinburne Romine, R. E., Hamilton, A., & Coleman, E. (2013). Stigma, mental health, and resilience in an online sample of the US transgender population. American Journal of Public Health, 103(5), 943–951. https://doi.org/10.2105/ajph.2013.301241

Bogaert, A. F. (2004). Asexuality: Prevalence and associated factors in a

national probability sample. The Journal of Sex Research, 41(3), 279–287. https://doi.org/10.1080/00224490409552235

Borgogna, N. C., McDermott, R. C., Aita, S. L., & Kridel, M. M. (2018). Anxiety and depression across gender and sexual minorities: Implications for transgender, gender nonconforming, pansexual, demisexual, asexual, queer, and questioning individuals. Psychology of Sexual Orientation and Gender Diversity, 6(1), 54–63. https://doi.org/10.1037/sgd0000306

Brennan, S. L., Irwin, J., Drincic, A., Amoura, N. J., Randall, A., & Smith-Sallans, M. (2017). Relationship among gender-related stress, resilience factors, and mental health in a midwestern U.S. transgender and gendernonconforming population. International Journal of Transgenderism, 18(4), 433–445. https://doi.org/10.1080/15532739.2017.1365034

Brotto, L. A., Knudson, G., Inskip, J., Rhodes, K., & Erskine, Y. (2010). Asexuality: A mixed-methods approach. Archives of Sexual Behavior, 39(3), 599–618. https://doi.org/10.1007/s10508-008-9434-x

Brotto, L. A., & Yule, M. A. (2009). “Methodological issues for studying asexuality”: Reply to Hinderliter. Archives of Sexual Behavior, 38(5), 622–623. https://doi.org/10.1007/s10508-009-9514-6

Brotto, L. A., Yule, M. A., & Gorzalka, B. B. (2015). Asexuality: An extreme variant of sexual desire disorder? The Journal of Sexual Medicine, 12(3), 646–660. https://doi.org/10.1111/jsm.12806

Budge, S. L., Chin, M. Y., & Minero, L. P. (2017). Trans individuals’ facilitative coping: An analysis of internal and external processes. Journal of Counseling Psychology, 64(1), 12–25. https://doi.org/10.1037/cou0000178

Chasin, C. D. (2011). Theoretical issues in the study of asexuality. Archives of Sexual Behavior, 40(4), 713–723. https://doi.org/10.1007/s10508-011-9757-x

Chavanduka, T. M., Gamarel, K. E., Todd, K. P., & Stephenson, R. (2020). Responses to the gender minority stress and resilience scales among transgender and nonbinary youth. Journal of LGBT Youth, 18(2), 135–154. https://doi.org/10.1080/19361653.2020.1719257

Cicero, E. C., Reisner, S. L., Merwin, E. I., Humphreys, J. C., & Silva, S. G. (2020). The health status of transgender and gender nonbinary adults in the United States. PLOS ONE, 15(2), e0228765. https://doi.org/10.1371/journal.pone.0228765

Clark, A. N., Lefkowitz, E. S., & Zimmerman, C. (2022). Sexual and affectionate behaviors in asexual and Allosexual adults. The Journal of Sex Research, 60(2), 242–252. https://doi.org/10.1080/00224499.2022.2051161

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396. https://doi.org/10.2307/2136404

Craig, S. L., McInroy, L., McCready, L. T., & Alaggia, R. (2015). Media: A catalyst for resilience in lesbian, gay, bisexual, transgender, and queer youth. Journal of LGBT Youth, 12(3), 254–275. https://doi.org/10.1080/19361653.2015.1040193

Crenshaw, K. W. (1995). Mapping the margins: Intersectionality, identity politics, and violence against women of color. In K. W. Crenshaw, N. Gotanda, G. Peller, & K. Thomas (Eds.), Critical race theory: The key writings that formed the movement (pp. 357–383). The New Press.

Crissman, H. P., Stroumsa, D., Kobernik, E. K., & Berger, M. B. (2019). Gender and frequent mental distress: Comparing transgender and non-transgender individuals’ self-rated mental health. Journal of Women’s Health, 28(2), 143–151. https://doi.org/10.1089/jwh.2018.7411

Cuthbert, K. (2019). “When we talk about gender we talk about sex”: (A) sexuality and (a)gendered subjectivities. Gender and Society, 33(6), 841–864. https://doi.org/10.1177/0891243219867916

Dawson, M., Scott, S., & McDonnell, L. (2018). ‘“Asexual” isn’t who I am’: The politics of asexuality. Sociological Research Online, 23(2), 374–391. https://doi.org/10.1177/1360780418757540

Decker, J. S. (2015). The invisible orientation: An introduction to asexuality. Simon & Schuster.

Fredrick, E. G., Mann, A. K., Brooks, B. D., & Hirsch, J. K. (2022). Anticipated to enacted: Structural stigma against sexual and gender minorities following the 2016 presidential election. Sexuality Research and Social Policy, 19(1), 345–354. https://doi.org/10.1007/s13178-021-00547-0

Fredriksen-Goldsen, K. I., Cook-Daniels, L., Kim, H., Erosheva, E. A., Emlet, C. A., Hoy-Ellis, C. P., Goldsen, J., & Muraco, A. (2013). Physical and mental health of transgender older adults: An at-risk and underserved population. The Gerontologist, 54(3), 488–500. https://doi.org/10.1093/geront/gnt021

Gamarel, K. E., Reisner, S. L., Laurenceau, J., Nemoto, T., & Operario, D. (2014). Gender minority stress, mental health, and relationship quality: A dyadic investigation of transgender women and their cisgender male partners. Journal of Family Psychology, 28(4), 437–447. https://doi.org/10.1037/a0037171

Gonzalez, C. A., Bockting, W. O., Beckman, L. J., & Durán, R. E. (2012). Agentic and communal personality traits: Their associations with depression and resilience among transgender women. Sex Roles, 67(9–10), 528–543. https://doi.org/10.1007/s11199-012-0202-y

Grant, J. M., Mottet, L. A., Tannis, J., Harrison, J., Herman, J. L., & Keisling, M. (2011). Injustice at every turn: A report of the National Transgender Discrimination Survey. National Center for Transgender Equality; National Gay and Lesbian Task Force. https://www.transequality.org/sites/default/files/docs/resources/NTDS_Report.pdf

Gupta, K. (2015). What does asexuality teach us about sexual disinterest?

Recommendations for health professionals based on a qualitative study with asexually identified people. Journal of Sex & Marital Therapy, 43(1), 1–14. https://doi.org/10.1080/0092623x.2015.1113593

Guz, S., Hecht, H. K., Kattari, S. K., Gross, E. B., & Ross, E. (2022). A scoping review of empirical asexuality research in social science literature. Archives of Sexual Behavior, 51, 2135–2145. https://doi.org/10.1007/s10508-022-02307-6

Fish, J. N., McInroy, L. B., Paceley, M. S., Williams, N. D., Henderson, S., Levine, D. S., & Edsall, R. N. (2020). “I'm kinda stuck at home with unsupportive parents right now”: LGBTQ youths' experiences with COVID-19 and the importance of online support. Journal of Adolescent Health, 67(3), 450–452. https://doi.org/10.1016/j.jadohealth.2020.06.002

Gonzalez, K. A., Ramirez, J. L., & Galupo, M. P. (2018). Increase in GLBTQ minority stress following the 2016 US presidential election. Journal of GLBT Family Studies, 14(1–2), 130–151. https://doi.org/10.1080/1550428x.2017.1420849

Hendricks, M. L., & Testa, R. J. (2012). A conceptual framework for clinical work with transgender and gender nonconforming clients: An adaptation of the minority stress model. Professional Psychology: Research and Practice, 43(5), 460–467. https://doi.org/10.1037/a0029597

Herek, G. M. (2009). Sexual stigma and sexual prejudice in the United States: A conceptual framework. In D. A. Hope (Ed.), Contemporary perspectives on lesbian, gay, and bisexual identities (pp. 65–111).

https://doi.org/10.1007/978-0-387-09556-1_4

Hides, L., Lubman, D. I., Devlin, H., Cotton, S., Aitken, C., Gibbie, T., & Hellard, M. (2007). Reliability and validity of the Kessler 10 and Patient Health Questionnaire among injecting drug users. Australian & New Zealand Journal of Psychiatry, 41(2), 166–168.

https://doi.org/10.1080/00048670601109949

Horne, S. G., McGinley, M., Yel, N., & Maroney, M. R. (2022). The stench of bathroom bills and anti-transgender legislation: Anxiety and depression among transgender, nonbinary, and cisgender LGBQ people during a state referendum. Journal of Counseling Psychology, 69(1), 1–13. https://doi.org/10.1037/cou0000558

Hughto, J. M., Pletta, D., Gordon, L., Cahill, S., Mimiaga, M. J., & Reisner, S. L. (2021). Negative transgender-related media messages are associated with adverse mental health outcomes in a Multistate study of transgender adults. LGBT Health, 8(1), 32–41. https://doi.org/10.1089/lgbt.2020.0279

James, S. E., Herman, J. L., Rankin, S., Keisling, M., Mottet, L., & Anafi, M. (2016). The report of the 2015 U.S. Transgender Survey. National Center for Transgender Equality. https://www.transequality.org/sites/default/files/ docs/USTS-Full-Report-FINAL.PDF

Kelleher, S., Murphy, M., & Su, X. (2023). Asexual identity development and internalisation: A scoping review of quantitative and qualitative evidence. Psychology & Sexuality, 14(1), 45–72. https://doi.org/10.1080/19419899.2022.2057867

Kessler, R. C., Andrews, G., Colpe, L. J., Hiripi, E., Mroczek, D. K., Normand, S.-L. T., Walters, E. E., & Zaslavsky, A. M. (2002). Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychological Medicine, 32(6), 959–976. https://doi.org/10.1017/s0033291702006074

Kessler, R. C., Barker, P. R., Colpe, L. J., Epstein, J. F., Gfroerer, J. C., Hiripi, E., Howes, M. J., Normand, S.-L. T., Manderscheid, R. W., Walters, E. E., & Zaslavsky, A. M. (2003). Screening for serious mental illness in the general population. Archives of General Psychiatry, 60(2), 184–189. https://doi.org/10.1001/archpsyc.60.2.184

Kroenke, K., Spitzer, R. L., Williams, J. B., Monahan, P. O., & Löwe, B. (2007). Anxiety disorders in primary care: Prevalence, impairment, comorbidity, and detection. Annals of Internal Medicine, 146(5), 317–325. https://doi.org/10.7326/0003-4819-146-5-200703060-00004

Lance, C. E., & Vandenberg, R. J. (2009). Introduction. In C. E. Lance & R. J. Vandenberg (Eds.), Statistical and methodological myths and urban legends: Doctrine, verity and fable in organizational and social sciences (pp. 1–4). Routledge.

Lefevor, G. T., Boyd-Rogers, C. C., Sprague, B. M., & Janis, R. A. (2019). Health disparities between genderqueer, transgender, and cisgender individuals: An extension of minority stress theory. Journal of Counseling Psychology, 66(4), 385–395. https://doi.org/10.1037/cou0000339

MacInnis, C. C., & Hodson, G. (2012). Intergroup bias toward “Group X”: Evidence of prejudice, dehumanization, avoidance, and discrimination against asexuals. Group Processes & Intergroup Relations, 15(6), 725–743. https://doi.org/10.1177/1368430212442419

Mara, C. A., & Peugh, J. L. (2020). Validity of data collected from randomized behavioral clinical trials during the COVID-19 pandemic. Journal of Pediatric Psychology, 45(9), 971–976. https://doi.org/10.1093/jpepsy/jsaa078

McInroy, L. B., Beaujolais, B., Leung, V. W., Craig, S. L., Eaton, A. D., & Austin, A. (2020). Comparing asexual and non-asexual sexual minority adolescents and young adults: Stressors, suicidality and mental and behavioural health risk outcomes. Psychology & Sexuality, 13(2), 387–403. https://doi.org/10.1080/19419899.2020.1806103

McInroy, L. B., & Craig, S. L. (2015). Transgender representation in offline and online media: LGBTQ youth perspectives. Journal of Human Behavior in the Social Environment, 25(6), 606–617. https://doi.org/10.1080/10911359.2014.995392

McInroy, L. B., & Craig, S. L. (2018). Online fandom, identity milestones, and selfidentification of sexual/gender minority youth. Journal of LGBT Youth, 15(3), 179–196. https://doi.org/10.1080/19361653.2018.1459220

Meyer, I. H. (1995). Minority stress and mental health in gay men. Journal of Health and Social Behavior, 36(1), 38–56. https://doi.org/10.2307/2137286

Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin, 129(5), 674–697. https://doi.org/10.1037/0033-2909.129.5.674

Meyer, I. H. (2010). Identity, stress, and resilience in lesbians, gay men, and bisexuals of color. The Counseling Psychologist, 38(3), 442–454. https://doi.org/10.1177/0011000009351601

Meyer, I. H., Brown, T. T., Herman, J. L., Reisner, S. L., & Bockting, W. O. (2017). Demographic characteristics and health status of transgender adults in select US regions: Behavioral Risk Factor Surveillance System, 2014. American Journal of Public Health, 107(4), 582–589. https://doi.org/10.2105/AJPH.2016.303648

Millet, N., Longworth, J., & Arcelus, J. (2017) Prevalence of anxiety symptoms and disorders in the transgender population: A systematic review of the literature. International Journal of Transgenderism, 18(1), 27–38. https://doi.org/10.1080/15532739.2016.1258353

Mink, M. D., Lindley, L. L., & Weinstein, A. A. (2014). Stress, stigma, and sexual minority status: The intersectional ecology model of LGBTQ health. Journal of Gay & Lesbian Social Services, 26(4), 502–521. https://doi.org/10.1080/10538720.2014.953660

Mizock, L., & Mueser, K. T. (2014). Employment, mental health, internalized stigma, and coping with transphobia among transgender individuals. Psychology of Sexual Orientation and Gender Diversity, 1(2), 146–158. https://doi.org/10.1037/sgd0000029

Mohr, J., & Fassinger, R. (2000). Measuring dimensions of lesbian and gay male experience. Measurement and Evaluation in Counseling and Development, 33(2), 66–90. https://doi.org/10.1080/07481756.2000.12068999

Nadal, K. L., Davidoff, K. C., Davis, L. S., & Wong, Y. (2014). Emotional, behavioral, and cognitive reactions to microaggressions: Transgender perspectives. Psychology of Sexual Orientation and Gender Diversity, 1(1), 72–81. https://doi.org/10.1037/sgd0000011

Newcomb, M. E., Hill, R., Buehler, K., Ryan, D. T., Whitton, S. W., & Mustanski, B. (2020). High burden of mental health problems, substance use, violence, and gender related psychosocial factors in transgender, nonbinary, and gender diverse youth and young adults. Archives of Sexual Behavior, 49(2), 645–659. https://doi.org/10.1007/s10508-019-01533-9

Newcomb, M. E., & Mustanski, B. (2010). Internalized homophobia and internalizing mental health problems: A meta-analytic review. Clinical Psychology Review, 30(8), 1019–1029. https://doi.org/10.1016/j.cpr.2010.07.003

Noelle, M. (2002). The ripple effect of the Matthew Shepard murder. American Behavioral Scientist, 46(1), 27–50. https://doi.org/10.1177/0002764202046001004

Prause, N., & Graham, C. A. (2007). Asexuality: Classification and characterization. Archives of Sexual Behavior, 36(3), 341–356. https://doi.org/10.1007/s10508-006-9142-3

Puckett, J. A., Maroney, M. R., Wadsworth, L. P., Mustanski, B., & Newcomb, M. E. (2020). Coping with discrimination: The insidious effects of gender

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

SUMMER 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

Boot-Haury | Psychological Distress of Asexual TGNC People

minority stigma on depression and anxiety in transgender individuals. Journal of Clinical Psychology, 76(1), 176–194.

https://doi.org/10.1002/jclp.22865

Puckett, J. A., Matsuno, E., Dyar, C., Mustanski, B., & Newcomb, M. E. (2019). Mental health and resilience in transgender individuals: What type of support makes a difference? Journal of Family Psychology, 33(8), 954–964.

https://doi.org/10.1037/fam0000561

Rachlin, K. (2019). The intersection of gender diversity and asexuality: Psychotherapy with TGNC individuals who identify as ace. In J. C. Niemira, G. J. Jacobson, & K. J. Violeta (Eds.), Sex, sexuality, and trans identities: Clinical guidance for psychotherapists and counselors (pp. 133–154). Jessica Kingsley Publishers.

Raun, T. (2015). Video blogging as a vehicle of transformation: Exploring the intersection between trans identity and information technology. International Journal of Cultural Studies, 18(3), 365–378. https://doi.org/10.1177/1367877913513696

Riggle, E. D., Rostosky, S. S., McCants, L. E., & Pascale-Hague, D. (2011). The positive aspects of a transgender self-identification. Psychology and Sexuality, 2(2), 147–158. https://doi.org/10.1080/19419899.2010.534490

Riggle, E. D., Whitman, J. S., Olson, A., Rostosky, S. S., & Strong, S. (2008). The positive aspects of being a lesbian or gay man. Professional Psychology: Research and Practice, 39(2), 210–217. https://doi.apa.org/doi/10.1037/0735-7028.39.2.210

Rostosky, S. S., Riggle, E. D., Pascale-Hague, D., & McCants, L. E. (2010). The positive aspects of a bisexual self-identification. Psychology and Sexuality, 1(2), 131–144. https://doi.org/10.1080/19419899.2010.484595

Sarno, E. L., Newcomb, M. E., & Mustanski, B. (2020). Rumination longitudinally mediates the association of minority stress and depression in sexual and gender minority individuals. Journal of Abnormal Psychology, 129(4), 355–363. https://doi.org/10.1037/abn0000508

Scandurra, C., Amodeo, A. L., Valerio, P., Bochicchio, V., & Frost, D. M. (2017). Minority stress, resilience, and mental health: A study of Italian transgender people. Journal of Social Issues, 73(3), 563–585. https://doi.org/10.1111/josi.12232

Scherrer, K. S. (2008). Coming to an asexual identity: Negotiating identity, negotiating desire. Sexualities, 11(5), 621–641. https://doi.org/10.1177/1363460708094269

Scroggs, B., Love, H. A., & Torgerson, C. (2020). COVID-19 and LGBTQ emerging adults: Risk in the face of social distancing. Emerging Adulthood, 9(5), 639–644. https://doi.org/10.1177/2167696820968699

Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., & Dunbar, G. C. (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59(20), 22–33. https://pubmed.ncbi.nlm.nih.gov/9881538/

Spector, I. P., Carey, M. P., & Steinberg, L. (1996). The sexual desire inventory: Development, factor structure, and evidence of reliability. Journal of Sex & Marital Therapy, 22(3), 175–190. https://doi.org/10.1080/00926239608414655

Straussner, S. L., & Calnan, A. J. (2014). Trauma through the life cycle: A review of current literature. Clinical Social Work Journal, 42(4), 323–335.

https://doi.org/10.1007/s10615-014-0496-z

Tan, K. K. H., Ellis, S. J., Schmidt, J. M., Byrne, J. L., & Veale, J. F. (2020). Mental health inequities among transgender people in Aotearoa New Zealand: Findings from the Counting Ourselves Survey. International Journal of Environmental Research and Public Health, 17(8), 1–14.

https://doi.org/10.3390/ijerph17082862

Testa, R. J., Habarth, J., Peta, J., Balsam, K., & Bockting, W. (2015). Development of the Gender Minority Stress and Resilience Measure. Psychology of Sexual Orientation and Gender Diversity, 2(1), 65–77. https://doi.org/10.1037/sgd0000081

Thorne, N., Witcomb, G. L., Nieder, T., Nixon, E., Yip, A., & Arcelus, J. (2019). A comparison of mental health symptomatology and levels of social support in young treatment seeking transgender individuals who identify as binary and nonbinary. International Journal of Transgenderism 20(2–3), 241–250. https://doi.org/10.1080/15532739.2018.1452660

Trujillo, M. A., Perrin, P. B., Sutter, M., Tabaac, A., & Benotsch, E. G. (2016). The buffering role of social support on the associations among discrimination, mental health, and suicidality in a transgender sample. International Journal of Transgenderism, 18(1), 39–52. https://doi.org/10.1080/15532739.2016.1247405

Valente, P. K., Schrimshaw, E. W., Dolezal, C., LeBlanc, A. J., Singh, A. A., &

Bockting, W. O. (2020). Stigmatization, resilience, and mental health among a diverse community sample of transgender and gender nonbinary individuals in the U.S. Archives of Sexual Behavior, 49(7), 2649–2660. https://doi.org/10.1007/s10508-020-01761-4

Van Houdenhove, E., Enzlin, P., & Gijs, L. (2017). A positive approach toward asexuality: Some first steps, but still a long way to go. Archives of Sexual Behavior, 46(3), 647–651. https://doi.org/10.1007/s10508-016-0921-1

Vares, T. (2017). ‘My [asexuality] is playing hell with my dating life’: Romantic identified asexuals negotiate the dating game. Sexualities, 21(4), 520–536. https://doi.org/10.1177/1363460717716400

Veldhuis, C. B., Drabble, L., Riggle, E. D., Wootton, A. R., & Hughes, T. L. (2017). “We won’t go back into the closet now without one hell of a fight”: Effects of the 2016 presidential election on sexual minority women’s and gender minorities’ stigma-related concerns. Sexuality Research and Social Policy, 15(1), 12–24. https://doi.org/10.1007/s13178-017-0305-x

Vincent, B. W. (2018). Studying trans: Recommendations for ethical recruitment and collaboration with transgender participants in academic research. Psychology & Sexuality, 9(2), 102–116. https://doi.org/10.1080/19419899.2018.1434558

Warren, J. C., Smalley, K. B., & Barefoot, K. N. (2016). Psychological well-being among transgender and genderqueer individuals. International Journal of Transgenderism, 17(3–4), 114–123. https://doi.org/10.1080/15532739.2016.1216344

Weathers, F. W., Litz, T. B., Herman, S. D., Huska, A. J., & Keane, M. T. (1993, October, 24). The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility [Paper presentation]. International Society for Traumatic Stress Studies, San Antonio, TX.

Williams, S. L., Job, S. A., Todd, E., & Braun, K. (2020). A critical deconstructed quantitative analysis: Sexual and gender minority stress through an intersectional lens. Journal of Social Issues, 76(4), 859–879. https://doi.org/10.1111/josi.12410

Woodford, M. R., Kulick, A., Garvey, J. C., Sinco, B. R., & Hong, J. S. (2018). LGBTQ policies and resources on campus and the experiences and psychological well-being of sexual minority college students: Advancing research on structural inclusion. Psychology of Sexual Orientation and Gender Diversity, 5(4), 445–456. https://doi.org/10.1037/sgd0000289

Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2015). A validated measure of no sexual attraction: The Asexuality Identification Scale. Psychological Assessment, 27(1), 148–160. https://doi.org/10.1037/a0038196

Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2016). Sexual fantasy and masturbation among asexual individuals: An in-depth exploration. Archives of Sexual Behavior, 46(1), 311–328. https://doi.org/10.1007/s10508-016-0870-8

Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2017). Human asexuality: What do we know about a lack of sexual attraction? Current Sexual Health Reports, 9(1), 50–56. https://doi.org/10.1007/s11930-017-0100-y

Zheng, L., & Su, Y. (2018). Patterns of asexuality in China: Sexual activity, sexual and romantic attraction, and sexual desire. Archives of Sexual Behavior, 47(4), 1265–1276. https://doi.org/10.1007/s10508-018-1158-y

Author Note. Jared W. Boot­Haury https://orcid.org/0000­0002­0644­5882

Jared W. Boot­Haury is now at the Department of Clinical Psychology at the Michigan School of Psychology, Farmington Hills, MI.

Materials and data for this study can be accessed at https://osf.io/75fys/ There are no known conflicts of interest to disclose. Special thanks to Jared’s dissertation chair Dr. Danielle Balaghi and his dissertation committee members, Drs. Shepler and McInroy. Their support during his dissertation paved the way for him to author this publication.

Positionality Statement: Jared identifies as a gay, demisexual, cisgender White man. He is nondisabled and acknowledges that his perspectives are influenced by his positions within all of these dimensions of identity.

Correspondence concerning this article should be addressed to Jared W. Boot­Haury, Department of Psychology, Michigan School of Psychology, 26811 Orchard Lake Rd., Farmington Hills, MI, 48334, United States. Email: jwboot3@icloud.com

141 COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

Role of Social Gaze on Visual Search in an Eye-Tracking Paradigm

Jonathan K. Kroeger, Erin A. Conway, and Ralph G. Hale* Department of Psychological Science, University of North Georgia

ABSTRACT. Gaze cueing refers to the natural inclination to direct one’s gaze in the direction of another’s gaze. This can be used in visual search tasks to facilitate or interfere with the search. Here, we examined the interaction between 3 gaze cue conditions (congruent, incongruent, and neutral) and a search task using letter arrays. Congruent cues had a gaze cue looking at the target, incongruent cues had a gaze cue looking away from the target, and neutral cues had a gaze cue looking straight ahead. An eye tracker was used to measure search task completion times (CTs). For all conditions, participants fixated on a target stimulus in the middle of the screen to begin the trial. The target disappeared. Then the gaze cue appeared, followed by the search array. Participants then had to locate the target. We expected CTs to be shortest for congruent cues and longest for incongruent cues, with neutral cues somewhere in the middle. Trials were randomized between our 3 conditions. A 1­way ANOVA found a significant impact of gaze cueing on this search task, F(2) = 10.15, p < .001, ηp2 = .28. Pairwise comparisons found significant differences between congruent and incongruent conditions, p < .001, and between neutral and congruent conditions, p = .006. Together, these results supported our hypothesis; congruent gaze cueing facilitated visual search whereas incongruent gaze and neutral gaze failed to assist visual search. These findings provide a firmer understanding of the interactions between gaze cueing and visual search paradigms.

Keywords: visual search, gaze cue, eye tracking

Humans are inherently social creatures and, as such, tend to look to others as sources of information. These social cues can provide a rich wealth of information that one individual may not gain otherwise. One such cue is social gaze. The direction in which a person looks can focus and orient the gaze of another. This phenomenon is known as gaze cueing. The natural inclination to direct one’s gaze in the direction of another’s gaze can be used in visual search tasks to facilitate or interfere with the search (e.g., Friesen & Kingstone, 1998). People often use gaze cueing in the real­world as a form of communication (e.g., Macdonald & Tatler, 2013). As such, gaze cueing is not just some artifact of vision research in a laboratory; it has real­world applications to how people search for objects in their environment, how they socialize, and how they communicate.

A typical gaze cue paradigm, like the one used by Friesen and Kingstone (1998), uses the visual direction of another person’s gaze to locate a specific target. By adding a face stimulus to a visual search task, the

direction of the gaze can either increase or decrease reaction times depending on the congruency of the gaze. A congruent gaze is oriented toward the search target, whereas an incongruent gaze is directed away from the target. This effect has been shown to account for reaction time differences between these conditions in tactile tasks (e.g., Friesen & Kingstone, 1998). In these tactile tasks, a participant button press triggers the end of a search task trial once the target is visually located. Only more recently has gaze cueing been tested using eye tracking (Bayliss et al., 2012; Cronin & Brockmole, 2016). Reaction times for a search task are more precise using eye tracking because they do not require a button press or verbal participant response. Here, we intended to combine the gaze cue and eye tracking facilitated visual search paradigms in a visual search task.

Visual Search

Visual search is a well­studied phenomenon in which individuals are tasked with finding target objects in

of Gaze on Visual Search | Kroeger, Conway, and Hale

a variety of conditions (e.g., Castelhano & Heaven, 2010; Chan & Hayward, 2013; Duncan & Humphreys, 1989; Eckstein, 2011; Mueller and Weidemann, 2012; Neisser, 1964; Wolfe, 2015). Various cues can be utilized to achieve a successful search, such as color, motion, orientation, and size (Wolfe & Horowitz, 2017). Eye tracking technology allows for precise tracking of an individual’s eye movements as they search within an array. This technology was incorporated into the present study to provide accurate visual search task completion times (CTs) while reducing unsystematic variability from measurement error intrinsic to other methods (e.g., the time it takes for an individual to press a button to indicate they located a target in tactile tasks; ability of the researcher to confirm they in fact located the target).

For any visual search task, there are a variety of factors to consider regarding the target and distractor stimuli. Semantic similarity and visual similarity are important for visual search tasks (Menneer et al., 2014). Additionally, semantic features are typically seen quickly within an environment using top ­ down processing (Hayes & Henderson, 2019; Neider & Zelinsky, 2006). This means visual and semantic similarity needs to be accounted for when deciding what stimuli to use for an array. For the present study, the semantic content of our target and distractor stimuli was identical. Therefore, semantic similarity could not be utilized in the search task. Visual similarity, however, is trickier to reduce between target and distractors. To do this, we must first consider a theory of visual organization.

Triesman’s feature integration theory (FIT) states that objects are perceived holistically through the organization of their individual parts (Treisman & Gelade, 1980). The FIT model has two stages. When visual information is received by the retina and transduced into a neural signal, it reaches the first stage in which it is analyzed preattentively. Then, stimulus features are combined together during the second focused attention stage. Conscious perception of the visual object follows the second stage. Many stimulus features comprise a visual object, including color, line orientation, curvature, motion, and depth (Treisman & Gelade, 1980). If this system is distracted or if the cognitive load is too great, some of these features could be improperly combined prior to perception, resulting in illusory conjunctions (Treisman & Schmidt, 1982). This suggests that this functional system of object recognition has limitations and can make mistakes.

For the present study, the target and distractor stimuli that constituted our search array consisted of capital letters. These letters had a variety of stimulus features, including straight lines of various orientations, curves, and closure. To account for visual similarity,

letters were selected that were similar in terms of these features. Gestalt principles of organization and perceptual grouping are also relevant here. A Gestalt literally translates to a shape in German; however, these principles of visual perception were originally described as the way in which people’s visual system perceptually organizes and unifies parts of their visual field into more holistic combinations (e.g., Wertheimer, 1938). Perhaps most important of these Gestalt visual search features in searches of this nature are orientation and size (Wolfe & Horowitz, 2017). If all stimuli within the array are the same or similar in regard to orientation and size, then those two Gestalt principles are unable to influence the search task.

The knowledge of a target’s features can even increase performance in a visual search task (Castelhano & Heaven, 2010). If the target features are completely unique, this will result in a pop­out effect. Mueller and Weidemann (2012) tested the connection between response times and target letter search, determining the process of finding appropriate letters to avoid feature pop­out is relatively simple. So long as the features of the target are not completely unique when compared to the distractors in the letter array, the observer must still search through all items and their respective features to locate the target. If the target is completely unique (i.e., if it shares no features with any of the distractors in the array), then it is salient enough to pop out. This is a shift from goal­directed attention to stimulus­directed attention. For instance, avoiding an “O” in an array of distractors consisting of “X”s and “T”s will prevent the target stimulus from being salient enough to pop­out to the observer during the search task. An “O” shares no characteristics with the vertical and horizontal lines of a “T” or the two diagonal lines of an “X.” As such, we controlled for these visual similarities in the present study.

Previous research has suggested that participants might look to previous target locations; however, once this strategy proves unreliable, a more normative approach is likely to be applied (Manginelli & Pollmann, 2008; Peterson et al., 2001). Alternatively, participants may look at different screen locations to account for the fact that they have looked elsewhere in previous trials (Peterson et al., 2007). As the trials are presented randomly, this strategy too will prove unreliable over time and be abandoned. Travis et al. (2013) found that memory of stimulus array configurations could influence performance in later trials. This was accounted for through trial and condition randomization. Regardless of any perceived pattern, randomization ensures no order effects confound our results and no shortcut (other than the intended gaze cue) could reduce search task CTs.

In this study, three gaze cue conditions were used:

Role

congruent, incongruent, and neutral. Ramamoorthy and colleagues (2020) used images of real faces in more visually complex search tasks, but here we used a simple smiley face consistent with the original gaze cue study conducted by Friesen and Kingstone (1998). We measured the search task CTs for each condition with a target letter embedded in an array of distractor letters. We hypothesized that the congruent gaze condition would decrease search time, whereas incongruent would increase search time. The neutral gaze condition was expected to not impact the search task; as such, search CTs for the neutral condition should be between congruent and incongruent.

Method

Participants

There were 27 participants in total (20 women, 7 men). Participants were recruited from the University of North Georgia’s research participant pool and received partial course credit for their participation. This sample size and recruitment procedure was consistent with similar past studies (e.g., Bayliss et al., 2012; Cronin & Brockmole, 2016; Friesen & Kingstone, 1998). All participants were at least 18 years of age, ranging from 18 to 33, with an average age of 20. Participants were confirmed to have normal or corrected­to­normal visual acuity. Individuals with glasses were excluded from the study due to eye tracker calibration issues. Individuals with contacts had no issues with calibration and were able to participate. Information regarding race or ethnicity was not collected in this study and cannot be reported. We note this is a limitation to this study. Consistent with Roberts and colleagues (2020), cognitive research of this kind historically has not collected data related to the race or ethnicity of its participants. However, future research should aim to improve diversity and transparency in research participation.

Stimuli and Apparatus

Participants were seated in a room with low ambient light. Stimuli were presented on a Samsung U28E510 28­inch monitor (62.23 cm wide by 34.29 cm tall) with a 60 Hz refresh and a 3840 × 2160 screen resolution. Participants viewed the screen using a chin and forehead rest, seated from a distance of 124.5 cm where the screen subtended a visual angle of 28.06° × 15.68°. All stimuli were created using Adobe Photoshop. All images had a dark gray background (RGB: 60, 60, 60) with some combination of white letters (RGB: 255, 255, 255) and a gaze cue stimulus. The gaze cue stimulus was a yellow face (RGB: 255, 202, 25) that was always centrally located subtending a visual angle of 5.40° × 5.29°. The eyes subtended a visual angle of 1.43° × 1.40° with the

combined pupil and iris regions (i.e., gaze cue inducer) subtending an angle of 0.70° × 0.70°. The target letter was also centrally located. All letters (targets and distractors) subtended a visual angle of roughly 1.08° × 1.42°. Each search array was divided into nine equal quadrants, eight of which were filled with an assortment of letters (Z, X, T, and I), whereas the center only included the face. As noted previously, these letters were chosen as they have similar features, notably vertical, horizontal, and diagonal lines. As such, finding the target is difficult and contingent on locating the combination of features unique to the target letter configuration. The face directed its gaze in one of nine directions. The gaze was straight ahead for the neutral condition. For congruent and incongruent, the gaze was directed toward or away from the target stimulus, respectfully (see Figure 1). This direction could be up, down, left, right, or any of the four diagonal corners. The images for each trial were converted to videos using Windows Video Editor for the purposes of stimulus presentation. Each video started with the presentation of a target letter for 2,000 ms. This was followed by a blank screen for 600 ms, a gaze cue stimulus for 600 ms, and finally the letter array for 8,000 ms (see Figure 2 for trial sequence). Eye tracking was conducted using a Gazepoint GP3HD eye tracker utilizing Gazepoint Analysis Professional Edition software for calibration and analyses.

Procedure

This study was approved by the University of North Georgia Institutional Review Board prior to data collection. Here, we examined the interaction between three gaze cue conditions (congruent, incongruent, and

FIGURE 1

Example Stimuli

Kroeger, Conway, and Hale | Role of Gaze on Visual Search

Note. This figure depicts example stimuli from each of the three conditions: neutral gaze (top), congruent gaze (bottom left) and incongruent gaze (bottom right). In all three examples, the target is the letter “T.”

neutral) and a search task using letter arrays. Congruent cues had a gaze cue looking at the target, incongruent cues had a gaze cue looking away from the target, and neutral cues had a gaze cue looking straight ahead. An eye tracker was used to measure search task CTs. Upon arrival, participants provided informed consent and any questions were answered. Then, the participants’ vision was tested to verify they met our requirements of normal or corrected­to­normal acuity. Afterward, participants were instructed to sit in a chair with their head resting on a chin and forehead rest. Before the study could commence, a nine­point eye tracker calibration was conducted. Following calibration, participants were given instructions for the experiment and given an opportunity to ask questions. Participants completed a block of four practice trials and again were given an opportunity to ask questions. Afterward, there were 24 experimental trials. All blocks and conditions used the same trial sequence (see Figure 2). Participants fixated on a target stimulus in the middle of the screen and pressed the spacebar to initiate the trial. Note that the target stimulus here is replacing a traditional fixation cross or point. The target disappeared upon spacebar press. Then the gaze cue appeared, followed 600 ms later by the letter search array. Participants had up to eight seconds to locate the target from the time the array appeared. All trials were randomized between our three conditions. When participants visually located and fixated on the target within the area of interest (AOI) for at least 275 ms, the trial terminated and the next began. A minimum fixation duration of 275 ms was set in the GazePoint software because 275 ms is the average fixation duration of a visual search task (Rayner, 2009)

and well above the minimum fixation duration (Brand et al., 2020). A slightly longer minimum fixation duration helps to filter out shorter transient fixations that may not indicate true search completion time. Following the experimental block of trials, participants were debriefed regarding the purpose of the study and given an opportunity to ask questions.

Results

For this repeated ­ measures design, a single group of 27 individuals participated in every level of the independent variable. The gaze cue variable had three conditions: congruent, incongruent, and neutral gaze. The dependent variable was search CT. For each trial, the total duration from trial onset to AOI fixation was recorded. Because search could not begin until the stimulus array appeared, CT equaled the total recorded duration minus 3,200 ms (see trial sequence in Figure 2). As mentioned previously, only AOI fixations with a minimum fixation duration of 275 ms were recorded. The duration of the fixation was not included in the CT because the search task was complete once the AOI was located. Collapsed across condition, CT for this task took approximately three seconds per trial (M = 2.97, SD = 0.55). To compare between conditions, a one­way ANOVA was conducted. Gaze cueing significantly impacted search task CT, F(2) = 10.15, p < .001, η p 2 = .28. As seen in Figure 3, CTs increased from congruent to neutral to incongruent. To determine which the relationships between these conditions were statistically significant, post hoc pairwise comparisons found significant differences between congruent ( M = 2.64, SD = 0.60) and incongruent conditions

FIGURE 2 Trial Sequence

Note. This figure depicts the sequence of a single trial, consisting of a target, blank screen, gaze cue stimulus, and search array.

FIGURE 3

Average Search Task Completion Times

Note. Search task completion times are shown for congruent, neutral, and incongruent gaze cues.

Role of Gaze on Visual Search | Kroeger, Conway, and Hale

( M = 3.22, SD = 0.53), p < .001. Similarly, neutral (M = 3.05, SD = 0.53) and congruent conditions were also significantly different, p = .006. However, there was not a significant difference between neutral and incongruent conditions, p = .18. Therefore, the gaze cue manipulation did influence the CTs. Specifically, congruent gaze cueing reduced search CT compared to neutral or incongruent conditions.

Discussion

In the present study, social gaze cueing was used to influence search CTs in a visual search task with three conditions: congruent gaze, incongruent gaze, and neutral gaze. Target and distractor stimuli were carefully chosen as to be similar visually and semantically, two factors shown to be important in visual search (e.g., Menneer et al., 2014). Stimulus features were matched for orientation and size to avoid Gestalt characteristics impacting the search task (Wolfe & Horowitz, 2017). Curvature and other visual dissimilarity that could result in visual pop­out was also avoided (Castelhano & Heaven, 2010; Mueller & Weidemann, 2012; Triesman & Gelade, 1980). Under these controlled conditions, we were able to directly test the impact of social gaze on visual search CT. Participants were shown a target search letter and a gaze cue. Then a search array consisting of the target and distractors was displayed. Participants visually scanned the array until the target was located. Participant CTs were recorded using an eye tracker that identified when a target was successfully located. Results indicated CTs were significantly shorter for congruent gaze compared to incongruent gaze, suggesting gaze direction played a meaningful role in search time and strategy.

Our findings show that visual gaze directed at a target stimulus can impact the speed at which participants are able to locate the target within a search task. CTs were significantly shorter for congruent gaze than for neutral or incongruent gaze. However, the impact of gaze cueing was asymmetrical when comparing congruent to incongruent. Although congruent gaze had a faciliatory effect on CT, incongruent gaze was not symmetrically as inhibitory. Congruent and incongruent gaze were significantly different from each other, but incongruent gaze was not significantly different than neutral gaze (unlike congruent and neutral gaze). This suggests, at least in this paradigm, that congruent gaze has more potential as a facilitatory effect than incongruent gaze has as an inhibitory one. One potential explanation for this relates to participant search strategy. Participants were not instructed whether or not to use the gaze stimulus when completing the search task. Despite this, it is clear participants were using the gaze stimulus enough

to have a faciliatory effect in the congruent condition. However, in the incongruent condition, participants might have disengaged rapidly from the gaze direction if the target was not located immediately. In other words, if the gaze direction pointed to the target, that would facilitate a faster CT. Alternatively, if the gaze direction pointed only to distractors, participants could continue a search of the whole array in a manner similar to the neutral gaze. This strategy would result in faster CTs for the congruent condition and slower CTs for neutral and incongruent gaze. As such, this might account for the similar CTs for the neutral and incongruent gaze conditions. However, we did not test this particular explanation of the results in our study, which leaves open the possibility for alternative explanations that we did not consider.

Importantly, this pattern of results (congruent CT < neutral; incongruent = neutral) is consistent with results from Friesen and Kingstone (1998). Therefore, the gaze cue worked similarly between these two studies despite the present study being a visual search task and the other being a reflexive orienting task. A similar pattern of findings was found by Driver and colleagues (2010), even when participants were instructed that the noncued location would be the more likely target location. This shows how powerful social gaze is as a search cue.

Given all of this, we must be careful in generalizing the findings from our present study to real­world visual search scenarios. The real­world contains vast semantic and visual dissimilarity between items in a visual scene. Nevertheless, our findings do support previous research suggesting that gaze cueing is a powerful cue for visual search (e.g., Friesen & Kingstone, 1998) and visual attention (e.g., Stephenson et al., 2021). Previous studies have investigated the gaze cue paradigm as a method of manipulating search and reaction times. Other studies have explored the role of features in a letter array search paradigm (i.e., FIT). Our pattern of results suggests that these previous studies are actually replicable and therefore are reliable. In addition, this study is novel as it was the first of its kind to combine a gaze cue paradigm, including congruent, incongruent, and neutral cues, with a letter array visual search task. Using eye tracking to measure accurate reaction time data makes this design a valid and reliable test of our hypothesis (Brand et al., 2021; Carter & Luke, 2020; Spinner et al., 2013). Our findings contribute to a greater understanding of the impact of gaze cueing on visual search. Together, these results support our expectations; congruent gaze cueing facilitated visual search whereas incongruent gaze and neutral gaze failed to assist visual search. These findings provide a firmer understanding of the interactions between gaze cueing and visual search paradigms.

Kroeger, Conway, and Hale | Role of Gaze on Visual Search

Role of Gaze on Visual Search | Kroeger, Conway, and Hale

Application

Any added knowledge in the areas of gaze cueing and visual search could have an impact on applied fields, such as the research into joint attention for children diagnosed with autism spectrum disorder. Previous research has emphasized how gaze cueing is a nonverbal communication tool (Bruinsma et al., 2004). Humans develop these nonverbal skills (e.g., gaze cueing, joint attention) very early in development, well before symbolic communication like formal spoken or written language. Children with autism show difficulties with a variety of socially learned cues (e.g., eye gaze alteration). This research is nuanced, however. Children with autism are capable of attending to social gaze covertly, indicating they understand the intention of the social cues despite a difference in their own observable gaze behavior (Gernsbacher et al., 2008). A recent study by Stephenson and colleagues (2021) has integrated some of these ideas with a cognitive shared­attention model to highlight how social gaze and joint attention may share similar mechanisms and cognitive roles. Future research could expand upon these ideas to learn more about the interactions between gaze cueing, visual search, and the myriad potential applications in these settings.

Future studies could continue to improve upon these designs in a variety of ways. To reduce unsystematic variability related to individual search differences, target distances from initial fixation could be manipulated. In the current study, we collapsed across target distances from fixation. There was not a significant difference in CTs for targets of different distance from fixation, but this could potentially impact search CT variability. This could be addressed by using a square aspect ratio for our background and stimulus array quadrants. Additionally, we could create more ecologically valid visual search arrays. As mentioned previously, real faces have been used in these paradigms with success (Ramamoorthy et al., 2020). Using real faces in conjunction with natural search scenes, both static and moving, would increase ecological validity and make generalizing these findings to the real ­ world more appropriate. Eye tracking is helpful in determining precise search completion timing. Integrating eye tracking into more ecologically valid scenes on screen, in virtual reality, or in real­world environments would provide support for our current theories regarding the effects of gaze on attention and visual search. As described previously, this is important as it pertains to the real­world nature of gaze cueing as a communication tool. To truly understand the impact gaze cueing has on real­world visual searching and social communication, research environment must strive to mirror the real­world ever more closely.

One limitation to this study relates to our sam

pling and collection of demographic information. As mentioned previously, information regarding race or ethnicity was not collected in this study and cannot be reported. Historically, cognitive research of this kind has not collected data related to the race or ethnicity of its participants (see Roberts et al., 2020). However, future research should aim to improve diversity and transparency in research participation. This is important for the generalizability of our findings in basic research and for the potential application of our research in other settings.

References

Bayliss, A. P., Murphy, E., Naughtin, C. K., Kitikos, A., Schilbach, L., & Becker, S. I. (2012). “Gaze leading”: Initiating simulated joint attention influences eye movements and choice behavior. Journal of Experimental Psychology, 142(1), 76–92. https://doi.org/10.1037/a0029286

Brand, J., Diamond, S. G., Thomas, N., & Gilbert-Diamond, D. (2021). Evaluating the data quality of the Gazepoint GP3 low-cost eye tracker when used independently by study participants. Behavior Research Methods, 53, 1502–1514. https://doi.org/10.3758/s13428-020-01504-2

Bruinsma, Y., Koegel, R. L., & Koegel, L. K. (2004). Joint attention and children with autism: A review of the literature. Mental Retardation and Developmental Disabilities Research Reviews, 10(3), 169–175. https://doi.org/10.1002/mrdd.20036

Carter, B. T., & Luke, S. G. (2020). Best practices in eye tracking research. International Journal of Psychophysiology, 155, 49–62. https://doi.org/10.1016/j.ijpsycho.2020.05.010

Castelhano, M. S., & Heaven, C. (2010). The relative contribution of scene context and target features to visual search in scenes. Attention, Perception & Psychophysics, 72(5), 1283–1297. https://doi.org/10.3758/app.72.5.1283

Chan, L. K., & Hayward, W. G. (2013). Visual search. Wiley Interdisciplinary Reviews: Cognitive Science, 4(4), 415–429. https://doi.org/10.1002/wcs.1235

Cronin, D. A., & Brockmole, J. R. (2016). Evaluating the influence of a fixated object’s spatio-temporal properties on gaze control. Attention, Perception, & Psychophysics, 78(4), 996–1003. https://doi.org/10.3758/s13414-016-1072-0

Driver IV, J., Davis, G., Ricciardelli, P., Kidd, P., Maxwell, E., & Baron-Cohen, S. (1999). Gaze perception triggers reflexive visuospatial orienting. Visual Cognition, 6(5), 509–540. https://doi.org/10.1080/135062899394920

Duncan, J., & Humphreys, G. W. (1989). Visual search and stimulus similarity. Psychological Review, 96(3), 433–458. https://doi.org/10.1037/0033-295x.96.3.433

Eckstein, M. P. (2011). Visual search: A retrospective. Journal of Vision, 11(5), 14–14. https://doi.org/10.1167/11.5.14

Hayes, T. R., & Henderson, J. M. (2019). Scene semantics involuntarily guide attention during visual search. Psychonomic Bulletin & Review, 26, 1683–1689. https://doi.org/10.3758/s13423-019-01642-5

Friesen, C. K., & Kingstone, A. (1998). The eyes have it! Reflexive orienting is triggered by nonpredictive gaze. Psychonomic Bulletin & Review, 5(3), 490–495. https://doi.org/10.3758/bf03208827

Gernsbacher, M. A., Stevenson, J. L., Khandakar, S., & Goldsmith, H. H. (2008). Why does joint attention look atypical in autism? Child Development Perspectives, 2(1), 38–45. https://doi.org/10.1111/j.1750-8606.2008.00039.x

Macdonald, R. G., & Tatler, B. W. (2013). Do as eye say: Gaze cueing and language in a real-world social interaction. Journal of Vision, 13(4), 6–6. https://doi.org/10.1167/13.4.6

Manginelli, A. A., & Pollmann, S. (2008). Misleading contextual cues: How do they affect visual search? Psychological Research, 73(2), 212–221. https://doi.org/10.1007/s00426-008-0211-1

Menneer, T., Godwin, H. J., & Hout, M. C. (2014). Visual similarity is stronger than semantic similarity in guiding visual search for numbers. Journal of Vision, 14(10), 1201. https://doi.org/10.1167/14.10.1201

Mueller, S. T., & Weidemann, C. T. (2012). Alphabetic letter identification: Effects of perceivability, similarity, and bias. Acta Psychologica, 139(1), 19–37. https://doi.org/10.1016/j.actpsy.2011.09.014

Neider, M. B., & Zelinsky, G. J. (2006). Scene context guides eye movements during visual search. Vision Research, 46(5), 614–621. https://doi.org/10.1016/j.visres.2005.08.025

Neisser, U. (1964). Visual search. Scientific American, 210(6), 94–103. https://doi.org/10.1038/scientificamerican0664-94

Peterson, M. S., Beck, M. R., & Vomela, M. (2007). Visual search is guided by prospective and retrospective memory. Perception & Psychophysics, 69(1), 123–135. https://doi.org/10.3758/bf03194459

Peterson, M. S., Kramer, A. F., Wang, R. F., Irwin, D. E., & McCarley, J. S. (2001). Visual search has memory. Psychological Science, 12(4), 287–292. https://doi.org/10.1111/1467-9280.00353

Ramamoorthy, N., Jamieson, O., Imaan, N., Plaisted-Grant, K., & Davis, G. (2020). Enhanced detection of gaze toward an object: Sociocognitive influences on visual search. Psychonomic Bulletin & Review, 28(2), 494–502. https://doi.org/10.3758/s13423-020-01841-5

Rayner, K. (2009). Eye movements and attention in reading, scene perception, and visual search. The Quarterly Journal of Experimental Psychology, 62(8), 1457–1506. https://doi.org/10.1016/s0926-907x(05)80003-0

Roberts, S. O., Bareket-Shavit, C., Dollins, F. A., Goldie, P. D., & Mortenson, E. (2020). Racial inequality in psychological research: Trends of the past and recommendations for the future. Perspectives on Psychological Science, 15(6), 1295–1309. https://doi.org/10.1177/1745691620927709

Stephenson, L. J., Edwards, S. G., & Bayliss, A. P. (2021). From gaze perception to social cognition: The shared-attention system. Perspectives on Psychological Science, 16(3), 553–576. https://doi.org/10.1177/1745691620953773

Kroeger, Conway, and Hale | Role of Gaze on Visual Search

Spinner, P., Gass, S. M., & Behney, J. (2013). Ecological validity in eye-tracking: An empirical study. Studies in Second Language Acquisition, 35(2), 389–415. https://doi.org/10.1017/s0272263112000927

Travis, S. L., Mattingley, J. B., & Dux, P. E. (2013). On the role of working memory in spatial contextual cueing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 39(1), 208–219. https://doi.org/10.1037/a0028644

Treisman, A. M., & Gelade, G. (1980). A feature-integration theory of attention. Cognitive Psychology, 12(1), 97–136. https://doi.org/10.1016/0010-0285(80)90005-5

Treisman, A., & Schmidt, H. (1982). Illusory conjunctions in the perception of objects. Cognitive Psychology, 14(1), 107–141. https://doi.org/10.1016/0010-0285(82)90006-8

Wertheimer, M. (1938). Laws of organization in perceptual forms. In W. D. Ellis (Ed.), A source book of Gestalt psychology (pp. 71–88). Kegan Paul, Trench, Trubner & Company. https://doi.org/10.1037/11496-005

Wolfe, J. M. (2015). Visual search. In J. M. Fawcett, E. F. Risko, & A. Kingstone (Eds.), The Handbook of Attention (pp. 27–56). Boston Review.

Wolfe, J. M., & Horowitz, T. S. (2017). Five factors that guide attention in visual search. Nature Human Behaviour, 1(3). https://doi.org/10.1038/s41562-017-0058

Author Note. Ralph G. Hale https://orcid.org/0000­0001­5026­8417 We have no known conflict of interest to disclose. Correspondence concerning this article should be addressed to Ralph G. Hale, University of North Georgia, 3820 Mundy Mill Rd, Oakwood, GA 30566. Email: ralph.hale@ung.edu

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

Threatened Social Needs After Exclusion in Undergraduate Students With Varying Degrees of Attention Switching Difficulties

Jessica C. Reich and Richard S. Pond, Jr.* Department of Psychology, University of North Carolina Wilmington

ABSTRACT. Individuals on the autism spectrum seem to be at higher risk for social exclusion, which can have serious psychological and physiological consequences. The current study examined how individuals with varying traits of autism are impacted by social exclusion in terms of threats to their social needs for belonging, control, self­esteem, and meaningful existence. Undergraduates (N = 185) completed a self­report measure of autistic traits (i.e., Autism Spectrum Quotient), were randomly assigned to be included or excluded in a virtual ball­tossing game (i.e., CyberBall), and threats to their social needs were assessed. Results from multiple regression analyses indicated that greater challenges in attention switching impacted how individuals experienced social exclusion. More specifically, those who reported greater challenges also reported lower overall need threat (sr2 = .03) and threats to meaningful existence (sr2 = .03) after being excluded during CyberBall. These effects did not emerge after inclusion. Further, threats to overall needs (dlow = 6.73, davg = 5.21, d high = 3.70) and meaningful existence (d low = 1.47, d avg = 1.11, dhigh = 0.75) were greater in the exclusion condition compared to the inclusion condition across all levels of attention switching. However, the strength of the association between experimental condition and need threat was the weakest among individuals highest in attention switching difficulties. A better understanding of the way need threat manifests itself illuminates how individuals with varying levels of autistic traits may respond to a common type of bullying they experience: social exclusion.

Keywords: social exclusion, social rejection, traits of autism, need threat

Throughout history, social connectedness proved essential for survival (e.g., working with people to secure food, shelter, protection, and for mating purposes). As such, the need to belong became a human universal (Baumeister & Leary, 1995). Modern research finds that belongingness threats can lead to many detrimental consequences, both psychologically and physiologically (for a review, see DeWall et al., 2011).

Diversity badge earned for conducting research focusing on aspects of diversity. Open Data and Open Materials badges earned for transparent research practices. Data and materials are available at https://osf.io/cxhy7/

One major threat to belongingness is being excluded or ostracized by others.

Social exclusion and ostracism occur throughout life; people experience it in several social settings, such as at school and in the workplace. Unfortunately, some people are disproportionately affected by these threats to belonging; studies show higher rates of peer victimization (e.g., verbal abuse, social exclusion, physical

aggression) among students with disabilities (Carter, 2009; Little, 2002). Little (2002) found that during a one­year period, up to 94% of students with varying disabilities (e.g., autism, Asperger syndrome, nonverbal learning disorders) reported experiencing some form of victimization, with 75% reporting being bullied by peers and roughly 35% reporting experiencing some form of peer shunning. In a similar study, Carter (2009) found that roughly 65% of children and adolescents with Asperger syndrome (DSM­IV) were subject to bullying and being shunned by peers.

Students on the autism spectrum are at a higher risk for social exclusion (Carter, 2009; Little, 2002; Van Roekel et al., 2010). Autism Spectrum Disorder (ASD), according to the Diagnostic and Statistical Manual of Mental Disorders (DSM­V; American Psychiatric Association, 2013), is characterized by challenges in social communication and interaction, as well as rigid and repetitive patterns of behavior. Due to challenges in socialization, autistic students tend to have fewer friends and lower social statuses in their academic environment (Cappadocia et al., 2012). Furthermore, because autistic students requiring a lower level of support are the most immersed in the general education classroom, they may be the most at risk for isolation and bullying from their peers (Zablotsky et al., 2014).

Exclusion does not solely threaten the fundamental need to belong. According to the need threat model (Williams, 2009), when individuals are excluded, four fundamental social needs are threatened: the need to belong with others, the need to maintain sensibly high self­esteem, the need to have some amount of control over one’s social environment, and the need to feel like one’s existence is meaningful and worthy. Exclusion can reduce the perception that these needs are being adequately met.

More recently, researchers have examined how need threat manifests itself differently in certain groups, such as those on the autism spectrum. Sebastian and colleagues (2009) compared responses to simulated social exclusion using the CyberBall paradigm (Williams et al., 2000) between autistic and nonautistic individuals revealing that both groups reported similar threats to self­esteem, belonging, and control. Threats to meaningful existence, however, were greater among those on the spectrum (Sebastian et al., 2009).

In another study, researchers examined the psychological and physiological effects of exclusion in adults, again using the CyberBall paradigm (Trimmer et al., 2017). Trimmer and colleagues’ (2017) results paralleled that of Sebastian and colleagues (2009) in that both autistic and nonautistic individuals reported comparable need threats to self ­esteem, belonging, and control. Their findings differed in that autistic participants did not report heightened threats to meaningful existence

compared to nonautistic participants.

Traits associated with autism exist in the general population and may also impact how individuals experience exclusion. According to Baron­Cohen and colleagues (2001), these traits include variability in social skills, attention switching, communication, imagination, and attention to detail. These traits are based on the “triad” of autistic symptoms and other aspects of cognition associated with autism. Although based on autistic symptomology, it is important to note that these traits are not solely associated with autism.

However, each of these traits refers to an aspect of autism for which individuals may require additional support. Social skills refer to the challenges associated with social interactions that some autistic individuals may face, including difficulty interpreting others’ communications and deficits related to theory of mind (i.e., forming a theory related to what others are thinking and feeling; Cashin et al., 2009). Imagination refers to difficulties in imaginative play (Ten Eycke & Müller, 2015), and previous work has shown that lower levels of pretend play are associated with lower levels of peer­oriented social competence (Uren & Stagnitti, 2009). One aspect of imaginative play is social imagination, which enables individuals to imagine what another person or persons may be feeling, thinking, or experiencing. Being able to guess reasonably accurately what another person is thinking or feeling given the present context or situation is an important part of social connectedness and the ability to relate to others.

Attention switching refers to the ability of individuals to divide their attention between multiple pieces of information at the same time. For some autistic individuals, switching attention from one stimulus can be a slow process, resulting in processing delays (Williams, 1996). A related trait is exceptional attention to detail, leading to the tendency to adopt a bottom­up processing strategy, where individuals perceive parts of an object first and then the whole (O’Connor & Kirk, 2008). Attention switching and attention to detail may alter how individuals process social stimuli. Thus, these two components may drive some autistic individuals’ difficulties in social interactions and imagination. Delays in processing and a bottom­up perceptual strategy may make it increasingly difficult to understand the feelings, thoughts, or experiences of others.

The last trait, communication, refers to difficulties such as limited use of verbal and nonverbal communication, a lack of variance in pitch and tone, and restricted facial expressions (Cashin et al., 2009). Although these challenges may increase one’s likelihood of being socially excluded by others, they seem unlikely to impact how the exclusion is perceived or processed, and subsequently experienced, like the other traits.

In sum, difficulties in social skills, attention switching, imagination, and exceptional attention to detail seem likely to make individuals less vulnerable to the effects of exclusion. Difficulties associated with socialization and imagination may make individuals more at risk for exclusion, but less vulnerable to the effects of exclusion because they may hinder the ability to fully understand the actions of others, including exclusion. However, the lowered vulnerability from social skills and imagination may be due in part because of attention switching and attention to detail, which alter the processing of social stimuli. Communication, which is related more to the delivery rather than the reception of social interaction, seems unlikely to impact the effects of social exclusion.

To our knowledge, this was the first study to examine exclusion and these specific components of autism as they relate to need threat. To add to the literature on the intersection of exclusion, need threat, and autism, we examined how need threat manifests after exclusion among individuals with varying degrees of autistic traits. Whereas previous work examining exclusion and autism has focused on individuals diagnosed with autism, the current work examines how the experience of exclusion differs across the continuum of a variety of traits associated with autism. This level of analysis is important as it reveals how individuals high in such traits, but not necessarily diagnosed with autism, experience exclusion.

Hypotheses

We hypothesized that threats to social needs after exclusion would not vary as a function of individuals’ overall traits of autism (Sebastian et al., 2009; Trimmer et al., 2017). However, we hypothesized that need threats would vary as a function of autistic­like traits. Specifically, we expected that (a) greater difficulties in social skills, imagination, and attention­switching, as well as greater attention to detail, would relate to lower threats to social needs, and (b) communication difficulties would be unrelated to social need threats following exclusion.

Method

the averaged effect size, at least 126 participants would be needed for a multiple linear regression with 11 predictors, a .05 alpha level, and .80 power. Data collection continued across three semesters to ensure adequate power.

Participants

Two hundred thirteen university students were recruited as part of their Introductory to Psychology course requirements. Data from 28 participants were discarded because they indicated some suspicion regarding their partners in the CyberBall task. The final sample used for analyses included 185 participants (60.5% women, 32.4% men, 7% unspecified). These participants ranged in age from 18 to 30 years old (Mage = 19.27, SD = 1.83). Participants reported their race as European American (84.9%), African American (3.8%), Asian American (3.8%), American Indian/Alaskan Native (0.5%), Other (5.40%), or did not report their race (1.6%). All participants were awarded course credit for participating.

Measures

Power Analysis

An a priori power analysis was conducted using G*Power software for Mac (Faul et al., 2009) to determine the sample size required to achieve a power of .80. To conduct this power analysis, we used effect size estimates from Sebastian and colleagues (2009). In their study, Sebastian and colleagues (2009) only found a significant main effect of meaningful existence, however, we do not know the reasons for their lack of significant findings (e.g., small N, whether null is actually true). Therefore, we based our power analysis on all their need threat effects, averaged across analyses (avg. R2 = .06). Using

Traits of Autism

Participants completed the Autism Spectrum Quotient to assess subclinical traits of autism (AQ; Baron­Cohen et al., 2001). The AQ is a sensitive measure of autistic traits in the general population (i.e., traits found in clinical populations are also found in nonclinical populations to a lesser degree; Ruzich et al., 2015). However, the scale was designed to be a descriptive measure of autistic traits and has no reliable diagnostic power (Ruzich et al., 2015). Research finds this measure to be reliable and valid (Hoekstra et al., 2008; Lundqvist & Linder, 2017) and a recent meta­analysis found evidence for its usefulness in measuring autistic traits in adults of normal intelligence (Ruzich et al., 2015).

The AQ assesses five domains of autism across 50 items: social skills (e.g., “I find social situations easy” [reverse scored]), attention­switching (e.g., “I frequently get so strongly absorbed in one thing that I lose sight of other things”), attention to detail (e.g., “I often notice small sounds when others do not”), communication (e.g., “I enjoy social chit­chat” [reverse scored]), and imagination (e.g., “If I try to imagine something, I find it very easy to create a picture in my mind” [reverse scored]).

Participants responded to each item on a Likert scale from 1 (definitely agree) to 4 (definitely disagree), and 24 items required reverse­scoring. In the original scale, responses to each item are subsequently scored as either a 1 (autistic-like) or 0 (nonautistic-like) depending on the question (see Baron­Cohen et al., 2001, for more details). However, for increased variability, we maintained the Likert scale scoring for each item. Composite scores were

calculated for each domain by summing scores across items within each domain, with higher scores indicating greater agreement with autistic­like characteristics. The range of possible scores for the full scale is from 50 to 200 (10 to 40 for each domain). According to Baron­Cohen and colleagues (2001), a useful cutoff score on the AQ for distinguishing individuals with autism from those without is 128 (a score of 32 in original dichotomous scoring). Twenty­seven participants (14%) scored at or above this cutoff. See Table 1 for AQ facet descriptive statistics, intercorrelations, and reliability coefficients.

Threatened Social Needs

Participants completed the 12­item Need Threat Scale (Williams, 2009) to determine how participants’ needs for self­esteem (e.g., “During the CyberBall game, I felt good about myself” [reverse scored]), belonging (e.g., “I felt like an outsider during the CyberBall game”), control (e.g., “I felt in control during the CyberBall game”[reverse scored]), and meaningful existence (e.g., “I felt as though my existence was meaningless during the CyberBall game”) were threatened after social exclusion. Research has demonstrated the reliability and validity of this scale, both cross­sectionally and longitudinally (Davidson et al., 2019). Participants indicated their experienced feelings during the game on a 9­point scale (1 = not at all, 9 = very much so). Five items were reverse scored. Scores were summed to create an overall need threat score, such that higher scores indicated higher experienced need threat during the CyberBall game (α = .92). Scores were also determined for each individual domain. The range of possible scores for the full scale is from 12 to 108 (3 to 27 for each individual need). See Table 2 for descriptive statistics, intercorrelations, and reliability coefficients.

Procedure

Institutional review board approval was received prior to data collection. After arriving to the laboratory, participants provided informed consent and then completed the AQ and demographic information. Next, participants completed the CyberBall task, a manipulation with well­established validity that reliably elicits feelings of rejection (Williams et al., 2000). Participants were told that they would be playing a virtual ball­tossing game with two online partners; in reality, they played with the computer. To increase and enhance the believability of the task, a digital photograph of the participant was taken and uploaded to the game, as well as pictures of two same­sex confederates. Participants were randomly assigned to one of two conditions: social exclusion (n = 89) or inclusion (n = 96). In the task, if participants received the ball, they had the choice of whom to throw

the ball to next. When they did not receive the ball, they still had the opportunity to watch the decisions of the other players. An algorithm determined how often the participant received the ball based on his or her condition; those in the included condition received the ball 33% of the time, and those in the excluded condition received the ball only twice at the beginning of the task. After completing the game (30 trials), participants completed the Need Threat questionnaire. Before debriefing, the experimenter asked participants what they thought of their partners during the study to determine whether they suspected that their partners during CyberBall were fake. Twenty­eight participants indicated they were aware that they were playing against a computer algorithm. Then, participants were fully debriefed, informed about the purpose of the study as well as the deception used by the experimenters, and thanked for their participation.

Results

The current study tested whether traits of autism predicted need threat following social exclusion using independent samples t tests and multiple regression. The distributional properties of the variables were assessed via descriptive statistics prior to analysis. All data and code for the manuscript can be found on the Open Science Framework (at: https://osf.io/cxhy7/)

SUMMER

We first tested whether being excluded induced significant threatened needs. Homogeneity of variance was violated, F(1,181) = 13.23, p = .004, therefore the Welch correction was applied. As expected, those excluded in CyberBall had higher overall need threats

(M = 72.24, SD = 24.47) compared to those included ( M = 46.82, SD = 18.05), t (159.33) = 7.94, p < .001, d = 1.18. When broken down by social need, those excluded had significantly higher threats to belonging, t(169.03) = 7.66, p < .001, d = 1.13, control, t(182) = 7.30, p < .001, d = 1.07, self­esteem, t(153.97) = 4.37, p < .001, d = .65, and meaningful existence, t(167.69) = 7.47, p < .001, d = 1.10, than those included. Thus, we proceeded to test whether traits of autism moderated the relationship between exclusion and need threat.

To limit the number of analyses conducted, we conducted multiple regression analyses including all AQ traits and their respective interactions with condition for each need threat outcome1. Prior to creating cross­products for the moderation analysis, experimental condition was dummy coded (1 = excluded, 0 = included). Assumptions of regression (i.e., linearity, homoscedasticity, normality) were checked visually using residual plots. Due to violations of both normality and homoscedasticity, self­esteem and meaningful existence scores were square root transformed. These transformations lessened the extent to which these

assumptions were violated but did not completely rid of their violations.

The model predicting overall need threat was significant, F(11, 157) = 7.01, p < .001, R2adj = .28, as well as the model predicting threats to meaningful existence, F(11, 158) = 5.99, p < .001, R2adj = .25. Analyses revealed a significant two ­ way interaction between attention switching and condition for overall need threat, β = –1.61, t (157) = –2.03, p = .04, sr 2 = .01. When broken down by individual need, there was a significant interaction between attention switching and condition for meaningful existence, β = –0.09, t(158) = –2.11, p = .04, sr2 = .01. All other effects related to AQ traits were nonsignificant.

To decompose the nature of the significant interactions, the associations between need threat and attention switching were examined via simple slopes tested in the inclusion and exclusion conditions (Cohen et al., 2003). When participants were excluded, the relationships between overall need threat (t = –3.33, p = .001, sr2 = .03), meaningful existence (t = –3.29, p = .001, sr2= .03), and attention switching were significantly negative such that greater difficulties in attention switching related to lower need threats (see Figure 1). In the inclusion condition, the relationships between overall need threat (sr2 < .01), meaningful existence (sr2 < .01), and attention switching were nonsignificant (ps > .80). The associations between need threat and condition were also examined via simple slopes tested in low (–1 SD), average (M), and high (+1 SD) attention switching. Overall need threat was significantly higher in the exclusion condition than inclusion condition for those with low (t = 6.45, p < .001, d = 6.73), average (t = 7.31, p < .001, d = 5.21), and high (t = 3.62, p < .001, d = 3.70) attention switching. In an examination of threats to meaningful existence specifically, need threat was higher in the exclusion condition compared to the inclusion condition among those with low (t = 6.18, p < .001, d = 1.47), average (t = 6.68, p < .001, d = 1.11) and high (t = 2.23, p = .002, d = 0.75) levels of attention switching (see Figure 2). These findings suggest that participants’ needs were threatened in the exclusion condition regardless of attention switching. However, the slopes for those with low attention switching difficulties were the steepest, followed by average, and last high. Therefore, although exclusion is harmful to overall need threat and meaningful existence, it is less so for those high in attention switching difficulties than those lower in such difficulties.

1Separate analyses were also conducted for each AQ facet predicting each need threat subscale. The pattern of results was identical, except that the interaction between attention switching and condition was also a significant predictor of belonging and control when tested individually.

FIGURE 1

Association Between Attention Switching Deficits and Meaningful Existence as a Function of Experimental Condition

Note. AS = Attention Switching. The simple slope for exclusion was significant (p < .001). The pattern depicted above was the same for the significant two-way interaction with AS and condition predicting overall need threat.

Discussion

The current study examined how need threat following social exclusion manifests as a function of autistic­like traits with the aim of replicating and extending previous work on this topic (Sebastian et al., 2009; Trimmer et al., 2017). Sebastian and colleagues (2009) found that there were no differences in need threat between control and autistic participants after being excluded, except autistic participants had heightened threats to meaningful existence. Trimmer and colleagues (2017) found that need threat following exclusion did not differ between autistic and nonautistic participants. In the current study, we found that need threat after exclusion does differ as a function of difficulties in attention switching. Those with greater reported difficulties in this area reported lower overall need threat and threats to meaningful existence after being excluded during Cyberball. These findings suggest that certain autistic­like traits, such as attention switching, may impact how individuals with and without autism experience exclusion.

Meaningful existence may be especially relevant for those with greater autistic characteristics, as they may lack social support systems that those with fewer characteristics are afforded. For example, social skills and communication difficulties make it more difficult for individuals to cultivate rich social support (e.g., close friends) that can help buffer against the negative impacts of exclusion (Jobe & White, 2007; Müller et al., 2008). In the current study, individuals with higher social skills and communication difficulties also tended to report greater difficulties in attention switching, albeit these relationships were relatively weak. These weak relations may explain why individuals with greater attention switching difficulties were less impacted by exclusion in terms of their meaningful existence. Attention switching difficulties were not strongly associated with social skill and communication difficulties in the participants of the current study, therefore, these participants may have social support systems that afford them protection against these exclusion­related threats. Thus, meaningful existence is not as impacted by one’s level of attention switching difficulties, as shown in the results.

Another possible explanation for the observed pattern of results is related to the nature of difficulties associated with attention switching, such as dividing attention between several pieces of information simultaneously or shifting attention from one stimulus to another. When attention switching is slowed, it can result in pauses or delays in processing (Williams, 1996). Delayed processing may result in experiencing meaning out of context, as situations typically change by the time full understanding of the stimuli is achieved (Williams,

1996). Delays in processing may also help explain why individuals with greater difficulties in attention switching reported less need threat. These individuals may have not fully processed the stimuli in the Cyberball task and thus, may not have completely realized that they were being excluded by the other two players. Therefore, their needs were not threatened to the same extent because they never fully perceived the exclusion taking place. Although relevant, attention switching is not exclusive to autism. For example, attention­switching difficulties are also characteristic of attention deficit hyperactivity disorder (ADHD; White & Shah, 2006). Individuals also get better at attention­switching with age as their executive control improves (Hanania & Smith, 2010). Rumination, often associated with depression, also impedes an individual’s ability to attention switch (Yee Lo et al., 2012). Thus, the findings of the current study likely have implications outside of autism. A key strength of this study is the use of a continuous measure of autistic traits. Previous work in this area has focused primarily on group comparisons, between those with and without autism (Sebastian et al., 2009; Trimmer et al., 2017). In examining traits of autism in this way, we were better able to examine how exclusion impacts individuals across the autism spectrum and those without autism but who have difficulties in these key areas. This level of analysis is especially important given that autism is underdiagnosed in certain groups,

Note. AS = Attention Switching. Both simple slopes were significant (ps < .001). The pattern depicted above was the same for the significant two-way interaction with AS and condition predicting overall need threat.

FIGURE 2

Association Between Experimental Condition and Need Threat as a Function of Attention Switching Deficits.

such as racial minorities (Wiggins et al., 2020) and girls and women (Green et al., 2019).

Limitations

Several limitations of the current study might have impacted our results. First, it is important to consider that our participants came from a young, nonclinical, undergraduate population that were mostly White women. Therefore, the results of this study may not generalize to populations with differing characteristics. It is unclear whether the same findings would emerge in an older, more racially and gender­diverse community sample. Thus, all findings should be interpreted with caution.

Some research indicates that exclusion is experienced differently by girls and women (Benenson et al., 2013). Due to low power, we were unable to examine gender differences within our sample. Therefore, the current research, unfortunately, does not provide insight into whether any of the relationships found are impacted by gender.

Additionally, the AQ may not be the best measure for determining traits of autism in undergraduate student samples. Although the AQ had an overall Cronbach’s alpha of .83 in our sample, its subscales had far less reliability, with Cronbach’s alphas ranging from .61 to .76. Using a nonclinical sample of individuals may have also restricted the range of our measure of autistic traits. Participants consistently scored low; the highest score on the AQ obtained in our sample was 161 (the maximum possible score is 200). In addition, only 27 participants (14%) scored above 128 (a score of 32 in original scoring, the AQ’s cutoff score for clinically significant levels of autistic traits; Baron­Cohen et al., 2001).

Further, it is unclear whether similar effects would emerge with other exclusion paradigms. If lowered threats to meaningful existence are solely driven by a delay in processing exclusion during Cyberball, then it is possible that this effect is unique to the Cyberball paradigm. Another frequently used exclusion paradigm is the get acquainted paradigm (e.g., Twenge et al., 2001) where participants engage with several others and reciprocally share basic information about themselves. Afterward, participants are randomly assigned to receive false feedback that no one wanted to work with them on an upcoming task.

Attention switching difficulties, and resulting delays in processing, do not seem likely to be implicated in such a paradigm. However, social and imaginative skills might be. Difficulties in social and imaginative skills make it harder for individuals to fully understand the thoughts, feelings, intentions, and experiences of others, especially as it relates to theory of mind (Broekhof et al., 2015; Ten Eycke & Müller, 2015; Uren & Stagnitti, 2009; Zhou et al., 2019). In fact, autistic individuals have been found to misperceive whether

someone is their friend (Rotheram­Fuller et al., 2010), which requires understanding their thoughts, feelings, and intentions, and identification of inclusion or exclusion.

These difficulties may result in inaccurate interpretations of feedback from the get acquainted paradigm, impacting the effects of exclusion. For example, these difficulties could lead to interpretations that include explanations other than exclusion, resulting in lowered threats to social needs. This is unlikely, however, given the blatant nature of the exclusion and the fact that autistic individuals are able to identify when they are being excluded in a more ambiguous setting such as Cyberball (Sebastian et al., 2009).

Future Research

Although significant findings emerged, future studies of this kind should include more participants with high levels of autistic traits, especially those diagnosed if possible. Additionally, researchers should consider using multiple scales to evaluate traits of autism to ensure proper reliability as well as construct validity. This will allow researchers to be more confident in their conclusions about how exclusion threatens needs of selfesteem, belonging, control, and meaningful existence in autistic individuals across the spectrum.

Laboratory examinations of the experiences of exclusion among autistic individuals seem to be largely limited to Cyberball. Future researchers should consider utilizing other paradigms, including the get acquainted paradigm, when possible, to get a fuller understanding of the effects of exclusion among individuals on the autism spectrum. Additionally, a broadening of paradigms used will allow researchers to determine whether effects are a product of Cyberball or exclusion in general.

Conclusion

The current study serves as a starting point for future research on subclinical and clinical populations of autistic people. Although there is some research pertaining to social exclusion and autism, this study offers a more nuanced examination of social exclusion’s impact on individuals with a variety of autistic traits. A better understanding of the way need threat manifests itself will provide insight into how individuals with varying levels of autistic traits may respond to one of the most common types of bullying they experience: social exclusion. Additional research is necessary to attain clarity on what kinds of coping mechanisms people with autistic traits use in response to threatened social needs.

Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/highfunctioning autism, males and females, scientists and mathematicians. Journal of Autism and Developmental Disorders, 31(1), 5–17. https://doi.org/10.1023/A:1005653411471

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. https://doi.org/10.1037/0033-2909.117.3.497

Benenson, J. F., Markovits, H., Hultgren, B., Nguyen, T., Bulock, G., & Wrangham, R. (2013). Social exclusion: More important to human females than males. PLoS ONE, 8(2), ee55851. https://doi.org/10.1371/journal.pone.0055851

Broekhof, E., Ketelaar, L., Stockmann, L., van Zijp, A., Bos, M. G. N., & Rieffe, C. (2015). The understanding of intentions, desires, and beliefs in young children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 45(7). https://doi.org/10.1007/s10803-015-2363-3

Cappadocia, C. M., Weiss, J. A., & Pepler, D. (2012). Bullying experiences among children and youth with autism spectrum disorders. Journal of Autism and Developmental Disorders, 42(2), 266–277. https://doi.org/10.1007/s10803-011-1241-x

Carter, S. (2009). Bullying of students with Asperger syndrome. Issues in Comprehensive Pediatric Nursing, 32(3), 145–154. https://doi.org/10.1080/01460860903062782

Cashin, A., Sci, D. A., & Barker, P. (2009). The triad of impairment in autism revisited. Journal of Child and Adolescent Psychiatric Nursing, 22(4), 189–193. https://doi.org/10.1111/j.1744-6171.2009.00198.x

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Lawrence Erlbaum Associates Publishers.

Davidson, C. A., Willner, C. J., van Noordt, S. J. R., Banz, B. C., Wu, J., Kenney, J. G., Johannesen, J. K., & Crowley, M. J. (2019). One-month stability of Cyberball post-exclusion ostracism distress in adolescents. Journal of Psychopathology and Behavioral Assessment, 41(3), 400–408. https://doi.org/10.1007/s10862-019-09723-4

DeWall, C. N., Deckman, T., Pond, R. S., & Bonser, I. (2011). Belongingness as a core personality trait: How social exclusion influences social functioning and personality expression. Journal of Personality, 79(6), 979–1012. https://doi.org/10.1111/j.1467-6494.2010.00695.x

Faul, F., Erdfelder, E., Buchner, A., & Lang, A-G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. https://doi.org/10.3758/BRM.41.4.1149

Green, R. M., Travers, A. M., Howe, Y., & McDougle, C. J. (2019). Women and autism spectrum disorder: Diagnosis and implications for treatment of adolescents and adults. Current Psychiatry Reports, 21(4), 22. https://doi.org/10.1007/s11920-019-1006-3

Hanania, R., & Smith, L. B. (2010). Selective attention and attention switching: Towards a unified developmental approach. Developmental Science, 13(4), 622–635. https://doi.org/10.1111/j.1467-7687.2009.00921.x

Hoekstra, R. A., Bartels, M., Cath, D. C., & Boomsma, D. I. (2008). Factor structure, reliability and criterion validity of the Autism-Spectrum Quotient (AQ): A study in Dutch population and patient groups. Journal of Autism and Developmental Disorders, 38(8), 1555–1566. https://doi.org/10.1007/s10803-008-0538-x

Jobe, L. E., & White, S. W. (2007). Loneliness, social relationships, and a broader autism phenotype in college students. Personality and Individual Differences, 42(8), 1479–1489. https://doi.org/10.1016/j.paid.2006.10.021

Little, L. (2002). Middle-class mothers’ perceptions of peer and sibling victimization among children with Asperger’s syndrome and nonverbal learning disorders. Issues in Comprehensive Pediatric Nursing, 25(1), 43–57. https://doi.org/10.1080/014608602753504847

Lundqvist, L. O., & Lindner, H. (2017). Is the Autism-Spectrum Quotient a valid measure of traits associated with the autism spectrum? A Rasch validation in adults with and without autism spectrum disorders. Journal of Autism and Developmental Disorders, 47(7), 2080–2091. https://doi.org/10.1007/s10803-017-3128-y

Müller, E., Schuler, A., & Yates, G. B. (2008). Social challenges and supports from the perspective of individuals with Asperger syndrome and other autism spectrum disabilities. Autism, 12(2), 173–190. https://doi.org/10.1177/1362361307086664

O’Connor, K., & Kirk, I. J. (2008). Brief report: Atypical social cognition and social behaviors in autism spectrum disorder: A different way of processing rather than an impairment. Journal of Autism and Developmental Disorders, 38(10), 1989–1997. https://doi.org/10.1007/s10803-008-0559-5

Rotheraam-Fuller, E., Kasari, C., Chamberlain, B., & Locke, J. (2010). Social involvement of children with autism spectrum disorders in elementary school classrooms. Journal of Child Psychology and Psychiatry, 51(11), 1227–1234. https://doi.org/10.1111/j.1469-7610.2010.02289.x

Ruzich, E., Allison, C., Smith, P., Watson, P., Auyeung, B., Ring, H., & Baron-Cohen, S. (2015). Measuring autistic traits in the general population: A systematic review of the Autism-Spectrum Quotient (AQ) in a nonclinical population sample of 6,900 typical adult males and females. Molecular Autism, 6(2), 1–12. https://doi.org/10.1186/2040-2392-6-2

Sebastian, C. L., Blakemore, S-J., & Charman, T. (2009). Reactions to ostracism in adolescents with autism spectrum conditions. Journal of Autism and Developmental Disorders, 39(8), 1122–1130. https://doi.org/10.1007/s10803-009-0725-4

Ten Eycke, K. D., & Müller, U. (2015). Brief report: New evidence for a socialspecific imagination deficit in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 45(1), 213–220. https://doi.org/10.1007/s10803-014-2206-7

Trimmer, E., McDonald, S., Kelly, M., & Rushby, J. A. (2017). The physiological and psychological effects of ostracism in adults with ASD. Journal of Autism and Developmental Disorders, 47(8), 2326–2335. https://doi.org/10.1007/s10803-017-3146-9

Twenge, J. M., Baumeister, R. F., Tice, D. M., & Stucke, T. S. (2001). If you can't join them, beat them: Effects of social exclusion on aggressive behavior. Journal of Personality and Social Psychology, 81(6), 1058–1069. https://doi.org/10.1037//0022-3514.81.6.1058

Uren, N., & Stagnitti, K. (2009). Pretend play, social competence and involvement in children aged 5–7: The concurrent validity of the ChildInitiated Pretend Play Assessment. Australian Occupational Therapy Journal, 56(1), 33–40. https://doi.org/10.1111/j.1440-1630.2008.00761.x

Van Roekel, E., Scholte, R. H. J., & Didden, R. (2010). Bullying among adolescents with autism spectrum disorders: Prevalence and perception. Journal of Autism and Developmental Disorders, 40(1), 63–73. https://doi.org/10.1007/s10803-009-0832-2

White, H. A., & Shah, P. (2006). Uninhibited imaginations: Creativity in adults with attention-deficit/hyperactivity disorder. Personality and Individual Differences, 40(6), 1121–1131. https://doi.org/10.1016/j.paid.2005.11.007

Wiggins, L. D., Durkin, M., Esler, A., Lee, L-C., Zahorodny, W., Rice, C., YearginAllsopp, M., Dowling, N. F., Hall-Lande, J., Morrier, M. J., Christensen, D., Shenouda, J., & Baio, J., (2020). Disparities in documented diagnoses of autism spectrum disorder based on demographic, individual, and service factors. Autism Research: Official Journal of the International Society for Autism Research, 13(3), 464–473. https://doi.org/10.1002/aur.2255

Williams, D. (1996). Autism: An inside out approach. Jessica Kingsley Publishers.

Williams, K. D., Cheung, C. K., & Choi, W. (2000). Cyberostracism: Effects of being ignored over the internet. Journal of Personality and Social Psychology, 79(5), 748–762. https://doi.org/10.1037/0022-3514.79.5.748

Williams, K. D. (2009). Ostracism: Effects of being ignored and excluded. In M. Zanna (Ed.), Advances in Experimental Social Psychology, 41, 279–314. Academic Press.

Yee Lo, B. C., Lau, S., Cheung, S., & Allen, N. B. (2012). The impact of rumination on internal attention switching. Cognition and Emotion, 26(2), 209–223. https://doi.org/10.1080//02699931.2011.574997

Zablotsky, B., Bradshaw, C. P., Anderson, C. M., & Law, P. (2014). Risk factors for bullying among children with autism spectrum disorders. Autism, 18(4), 419–427. https://doi.org/10.1177/1362361313477920

Zhou, P, Zhan, L., & Ma, H. (2019). Understanding others’ minds: Social inference in preschool children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 49(11), 4523–4534. https://doi.org/10.1007/s10803-019-04167-x

Author Note. Jessica C. Reich

https://orcid.org/0000­0001­7520­1999

Richard S. Pond, Jr. https://orcid.org/0000­0001­5446­7456

This study was not preregistered. Materials, R code, and data for this study can be accessed at https://osf.io/cxhy7/. We have no known conflicts of interest to disclose. Correspondence concerning this article should be addressed to Jessica C. Reich, Department of Psychology, University of North Carolina Wilmington, 601 S. College Rd., Wilmington, NC 28403. Email: jcr2353@uncw.edu

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 2/ISSN 2325-7342)

Like Mother, Like Daughter: Excessive Reassurance Seeking Across Mother–Daughter and Romantic Relationships

Jaidyn K. Charlton, Elizabeth Grassia, Alexandra D. Popowich*, and Aislin R. Mushquash* Department of Psychology, Lakehead University

ABSTRACT. Excessive reassurance seeking (ERS) is a maladaptive self­regulatory behavior whereby an individual persistently seeks reassurance that one is loveable and worthy (Joiner et al. , 1992; Joiner et al., 1999). Despite research evidencing the negative consequences of ERS on relationships, little is known about how ERS presents across relationships. The present study evaluated the associations between mothers’ and daughters’ ERS across the parent–child relationship and across their respective romantic relationships. Sixty­seven mother–daughter dyads were recruited to separately complete measures of ERS in reference to their relationships with each other and their relationships with their romantic partners. Consistent with our hypotheses, mothers’ ERS directed toward their romantic partners correlated with mothers’ ERS directed toward their daughters, r(65) = .36, p = .003, and daughters’ ERS directed toward their mothers also correlated with daughters’ ERS directed toward daughters’ romantic partners, r(65) = .25, p = .04. These preliminary findings highlight the interpersonal nature of ERS and contribute to the understanding of the consistency of this maladaptive interpersonal behavior across relationships.

Keywords: excessive reassurance seeking, romantic relationships, mother–daughter dyads, interpersonal relationships

Excessive reassurance seeking (ERS) is a maladaptive interpersonal feedback ­ seeking behavior that involves a persistent and repetitive propensity to seek affirmations of one’s worth, value, and lovability despite the continued provision of such reassurances (Hames et al., 2013; Joiner et al., 1992; Joiner et al., 1999; Kane et al., 2018). ERS can be burdensome to the intended target of these behaviors and contribute to relationship dysfunction (Starr & Davila, 2008; Van Orden & Joiner, 2006). For example, in a sample of undergraduate women and their romantic partners, women’s ERS was associated with lower ratings of relationship quality and dyadic adjustment by both partners and was predictive of partner­initiated rejection (Stewart & Harkness, 2015). ERS can also manifest specifically in relation to difficulties with depression or anxiety. For example, those with social anxiety may seek reassurance about their appearance or performance (Cougle et al., 2012), whereas those with depression may seek reassurance to combat persistent self­criticism and self­doubt (Van Orden & Joiner, 2006). When individuals seeking excessive

reassurance do not receive support or acceptance, their mental health and well­being tends to deteriorate (Abe & Nakashima, 2022). Although reassurance from others may provide brief reductions in anxiety or concerns about self­worth, evidence suggests that this strategy ultimately maintains anxiety and negative cognitive appraisals in the longer term (Kane et al., 2018).

Although much research has established ERS as a maladaptive coping behavior associated with poor romantic relationship functioning (Lord et al., 2020; Shaver et al., 2005; Starr & Davila, 2008), there is still much to learn about how ERS manifests within and across different types of relationships (e.g., Ainsworth, 1989). Familial relationships are important to consider as they are the first interpersonal relationship a person typically develops, and heritability often plays a role in the development and presentation of certain traits. For example, greater anxiety and other related disorders, like obsessive ­ compulsive disorder, are associated with greater ERS within relationships (e.g., Cougle et al., 2012), and trait anxiety itself has a notable genetic component, increasing in influence as individuals grow

older and environmental contexts start to differ (Garcia et al., 2013). Twin studies also reveal that genetics impact the development of anxious attachment (approximately 40% heritable influence; Crawford et al., 2007), which is a dimension of insecure attachment wherein individuals worry about abandonment and rejection (Ainsworth et al., 1978; Ainsworth, 1989; Crawford et al., 2007), causing them to seek frequent reassurance. Although heritability plays a role in anxious attachment, there are also environmental influences (Crawford et al., 2007), suggesting that not only do genes contribute, but familial interactions and other external factors as well. Therefore, growing research into the origins of anxiety, and consequently ERS, heavily implicates the role of family.

Behaviorally, the quality of early familial relationships can shape future interpersonally oriented behaviors as well as the development and use of adaptive and maladaptive coping behaviors (Barry et al., 2007; Evraire et al., 2014; Katz et al., 2009; Merlo & Lakey, 2007). As parents may satisfy their own unmet needs through relationships with their children (e.g., Katz et al., 2009), a romantic relationship where reassurance needs are perceived to remain unmet may lead parents to seek similar repetitive assurances from their children. As patterns of interaction and behaviors initiated in childhood contribute to the development of an internal working model (i.e., schema or mental concept) for interacting with subsequent attachment figures (i.e., romantic partners; Bowlby, 1988; Bretherton, 1990), children who seek repeated reassurance from their parents may continue this pattern into their own romantic relationships. Indeed, there is some consistency between familial and peer relationships regarding coping styles, such as support seeking (Merlo & Lakey, 2007), which, in some contexts, can be construed as a form of reassurance seeking when there is a problem or worry. Taken together, exploring familial influence in greater detail will deepen the conceptualization of ERS and the extent family (i.e., genetics, attachment, and behavioral learning) plays a role in its development and maintenance, especially when other relationships are involved (e.g., romantic).

In the present study, we focused on the mother–daughter relationship and explored how mothers’ ERS directed toward their romantic partners associated with mothers’ ERS directed toward their daughters. We also tested whether daughters’ ERS directed toward their mothers associated with daughters’ ERS directed toward their own romantic partners. This was the first study exploring ERS across multiple relationships concurrently, and it provides an important contribution to the understanding of the persistence of this maladaptive interpersonal behavior. The exploratory nature of this

study can also provide a solid foundation for future research.

Methods

Between August 2018 and April 2019, we recruited 67 mother–daughter dyads. Daughters were young adult undergraduate students at a medium­sized Canadian university with a mean age of 19.59 (SD = 3.69); and were White (86.5%), Biracial (4.5%), Indigenous (3.0%), Middle Eastern (3.0%), Indian (1.5%), and Black (1.5%). Daughters were required to be in a monogamous romantic relationship of at least 3 months, and the most were exclusively heterosexual (74.7%). Mothers were classified as any primary female caregiver identified by daughters and included biological mothers (95.5%), adoptive mothers (1.5%), guardian mothers (1.5%), and older sisters (1.5%). Mothers were required to be in a romantic relationship of any duration, of which most were married (82.1%) and exclusively heterosexual (94.0%). Mothers had a mean age of 47.90 (SD = 6.00), and were White (88.0%), Indian (3.0%), Middle Eastern (3.0%), Biracial (1.5%), Black (1.5%), Indigenous (1.5%), and Chinese (1.5%). On average, mothers reported being involved in their daughters’ lives for 19.12 years (SD = 4.39), more than half (56.7%) lived in the same household as their daughter at the time of the study, and 55.3% had daily contact with their daughter.

Following review and approval from a university research ethics board, daughters were recruited to participate in a larger study focused on attachment, coping behaviors, and relationships. Daughters completed the study measures and provided contact information for their mothers. Subsequently, research assistants contacted mothers by phone and provided information about the study. Mothers had the opportunity to ask questions, then provided their consent to be emailed further information and the study’s online questionnaires. If the research assistant was unable to reach a mother by phone, the study’s information was sent to the mother’s email (provided by the daughter). For participating in the study, daughters were compensated with either 2 bonus credits toward an eligible psychology course or $15.00 CAD. Mothers were compensated with entry into a draw for a $50.00 CAD gift certificate. Analyses were conducted using SPSS Version 25. We used simple bivariate correlations to test hypotheses.

Measures

Excessive Reassurance Seeking

The 4 ­ item Reassurance Seeking subscale of the Depressive Interpersonal Relationships Inventory (DIRI­RS; Joiner & Metalsky, 2001) was used as a measure of ERS. Daughters and mothers responded to each

Charlton, Grassia, Popowich, and Mushquash | ERS Across Relationships

item on a 7­point Likert scale ranging from 1 (strongly disagree ) to 7 ( strongly agree ) with higher averaged scores indicating a higher degree of ERS directed toward their romantic partner. Although the original DIRI­RS assesses ERS in romantic relationships, and since there are no current measures assessing ERS in familial dyads that we are aware of, we modified the measure to also assess ERS in the mother–daughter relationship. For example, the item, “Do you find yourself often asking your romantic partner how he/she truly feels about you?” was modified to “Do you find yourself often asking your mother/daughter how she truly feels about you?” The practice of slightly modifying measures to reflect other relationships, dyads, or contexts is quite common, and is often a sound method of measuring constructs if psychometrically validated measures for the target relationships may not exist (see Mushquash & Sherry, 2012, as example). The original DIRI­RS is a psychometrically sound measure of reassurance seeking, often yielding alpha reliabilities between .80–.90 (Joiner et al., 1992; Joiner & Metalsky, 2001; Katz et al., 2009). The Cronbach’s alphas for the DIRI­RS for mother–daughter dyads and their respective romantic relationships for the present study are presented in Table 1, all of which are satisfactory.

Results

Means, standard deviations, and bivariate correlations are presented in Table 1. Means are consistent with those reported by other studies with similar samples, displaying low–moderate overall ERS (e.g., Forchuk et al., 2021; Fowler & Gasiorek, 2017; Taylor et al., 2019). As predicted, correlational analyses demonstrated that (a) mothers’ ERS directed toward their romantic partners was significantly positively associated with mothers’ ERS directed toward their daughters, r(65) = .36, p = .003, and (b) daughters’ ERS directed toward their mothers was significantly positively associated with daughters’

ERS directed toward daughters’ romantic partners, r(65) = .25, p = .04. There is also evidence to suggest that mothers’ ERS toward their romantic partners was positively associated with daughters’ ERS toward their romantic partners, r(65) = 26, p = .04. In summary, this suggests that, as mothers’ ERS increases, daughters’ ERS also increases, both within their familial relationship and their romantic relationships.

Discussion

The present exploratory study investigated the associations between ERS across familial and romantic relationships. Hypotheses were supported and speak to the persistence of ERS across relationships. Although we do not know about mothers’ ERS directed to their own caregivers, we did see preliminary evidence that this maladaptive interpersonal behavior occurs within mothers’ relationships with their own partners and mothers’ relationships with their daughters.

Our findings corroborated extant literature on the consistency of coping behaviors across generations, which maintains that parental influences (through inherited predisposition and/or modeling) of maladaptive coping responses, such as disengagement and related avoidant behaviors, may influence children’s adoption of similar coping behaviors when faced with stressful situations (e.g., Cougle et al., 2012; Garcia et al., 2013; Skinner & Zimmer­Gembeck, 2007). For example, evidence has shown that children’s emotional responses to vague events can be influenced by the emotion displayed by the parent (Power, 2004; Skinner & Zimmer­Gembeck, 2007). Relatedly, the suggestions that parents make to their children subsequently influence children’s coping. In a study of youth ages six to twelve, coping strategies suggested by mothers were positively associated with youth’s reported coping strategies (Power, 2004). Thus, parental modeling and suggestion provide plausible mechanisms for the consistency of coping behaviors, such as ERS, across generations.

Findings were also consistent with attachment theories suggesting dysfunctional attachment patterns and poorer adjustment can occur when caregivers switch roles with their child (i.e., when a caregiver seeks support/reassurance from their child; Ainsworth, 1989). In the context of our study, the “switching roles” is seen when mothers consistently seek reassurance from their daughter as opposed to the daughters seeking it from their mothers. Moreover, our findings can be understood within research suggesting that romantic relationships often resemble initial caregiving relationships, especially in terms of attachment security (Bowlby, 1988; Laurent & Powers, 2007; Thompson & Meyer, 2007). Research has noted that early insecure attachment is linked to

problems in adult romantic relationships; whereas, secure attachment is associated with warmer, supportive behaviors toward one’s romantic partner (Conger et al., 2000; McCarthy & Maughan, 2010; Shaver et al., 2005). For example, those demonstrating anxious attachment in adulthood often have great fears of rejection and abandonment that clash with their yearning for close interpersonal relationships. As such, this fear of rejection and abandonment prompts ERS and desires for greater proximity to their partner (Mikulincer et al., 2002), both of which could cause relationship problems. Similar findings appeared in the present study, as the ERS behavior daughters displayed toward their mothers was strongly associated with the ERS behaviors they displayed toward their romantic partners. These findings added to the literature suggesting that attachment to caregivers and the coping behaviors individuals are exposed to earlier in life play a notable role in determining the coping behaviors utilized into adulthood. This research not only informs researchers as to the impacts of maladaptive coping behaviors, but also informs clinicians providing support or counselling to families and couples. For example, issues can arise in romantic couples when there is an emotional disconnect between partners, which can prompt anxiety, depression, “clinginess,” and consequently ERS (e.g., frequently asking questions such as “Can I count on you?”; Johnson & Greenman, 2006). To combat these issues, emotionally focused therapy is a popular intervention for romantic couples and is founded in attachment theory (Johnson & Greenman, 2006), wherein individuals can actively engage with their emotions and how they interact in their interpersonal relationships. As such, clinicians could potentially guide clients through reflection on their ERS and other applicable coping behaviors and emotions in efforts to remedy whatever difficulties, relational or otherwise, they brought forth in psychotherapy. Armed with these findings, clinicians can also be better prepared to recognize potential interpersonal “risk” factors for interpersonal dysfunction. For example, if a client’s family, notably their mother, has a history of ERS and the client is experiencing difficulty in their relationships, exploring the client’s potential manifestations of ERS may be a viable avenue to explore. Furthering the understanding of attachment and how early experiences shape an individual’s behaviors, such as bolstering or hindering ERS, can aid in fine­tuning therapies to better fit specific individuals and dyads seeking treatment, and encourage clients to seek relationships with individuals who accept and support them (e.g., Abe & Nakashima, 2022).

This was the first study exploring ERS across multiple relationships concurrently and provides an

important contribution to the understanding of the persistence of this interpersonal behavior. Nevertheless, it is not without its limitations. The present study utilized bivariate correlations as a straight­forward, exploratory procedure. ERS research may hereafter require more sophisticated and complex statistical models and analyses (e.g., serial mediation or structural equation modelling) to determine elaborative causal, developmental relationships between variables, such as potential moderators or mediators related to attachment and coping. The analyses also yielded small to moderate effect sizes, so it is reasonable to theorize that other variables may impact the manifestation of ERS in mothers and daughters, like overall trait anxiety. Relatedly, we cannot know for certain if adapting the DIRI­RS to suit mother–daughter relationships maintains all psychometric properties of the original measure.

Another limitation of the present study is its reliance on cross­sectional data, which limits generalizability to other contexts and demographics. The data was also self­report, which can pose numerous subtle limitations, such as socially desirable responding or an inability to accurately reflect retrospectively or prospectively. Therefore, assessing ERS across time would allow for an examination of these and other detailed explanatory possibilities, as well as providing a more robust design and opportunity to minimize respondent biases. Moreover, since our sample consisted primarily of White, young adult women and their mothers, it is important for future research to examine ERS across relationships within a more diverse sample as it is unknown whether results herein are generalizable, such as to father–son relationships and samples with a greater representation of ethnicity.

Finally, further research examining ERS and other maladaptive coping behaviors within multiple relationships across the lifespan is warranted. For instance, longitudinal research exploring the development of ERS within a child in the context of the familial relationship, the factors contributing to this development (e.g., caregivers’ ERS directed toward a child), and the later consequences of this development (e.g., ERS directed toward a child’s family members, friends, and eventual romantic partners) would be worthwhile.

Conclusion

Excessively seeking reassurance from others has negative consequences on relationships and well­being. The present study advanced prior research on ERS by providing preliminary evidence to suggest that this pattern of interpersonal behavior is consistent across relationships. Results demonstrated the value of assessing the stability of maladaptive interpersonal coping behaviors, such as

Charlton, Grassia, Popowich, and Mushquash | ERS Across Relationships

SUMMER 2023

ERS, as they manifest across multiple relationships. Early relationships with caregivers and the patterns observed during developmental years are important in shaping the development and continued use of maladaptive interpersonal strategies, like ERS, over time.

References

Abe, K., & Nakashima, K. (2022). Excessive-reassurance seeking and mental health: Interpersonal networks for emotion regulation. Current Psychology, 41, 4711–4721. https://doi.org/10.1007/s12144-020-00955-2

Ainsworth, M. D. S. (1989). Attachments beyond infancy. American Psychologist, 44(4), 709–716. https://doi.org/10.1037//0003-066x.44.4.709

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Lawrence Erlbaum.

Barry, R. B., Lakey, B., & Orehek, E. (2007). Links among attachment dimensions, affect, the self, and perceived support for broadly generalized attachment styles and specific bonds. Personality and Social Psychology Bulletin, 33(3), 340¬–353. https://doi.org/10.1177/0146167206296102

Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. Basic Books.

Bretherton, I. (1990). Open communication and internal working models: Their role in attachment relationships. In R. Thompson (Ed.), Socioemotional development (Nebraska Symposium, 1987). University of Nebraska Press.

Conger, R. D., Cui, M., Byrant, C. M., & Elder Jr., G. H. (2000). Competence in early adult romantic relationships: A developmental perspective on family influences. Journal of Personality and Social Psychology, 79(2), 224–237. https://doi.org/10.1037//0022-3514.79.2.224

Cougle, J. R., Fitch, K. E., Fincham, F. D., Riccardi, C. J., Keough, M. E., & Timpano, K. R. (2012). Excessive reassurance seeking and anxiety pathology: Tests of incremental associations and directionality. Journal of Anxiety Disorders, 26(1), 117–125.

https://doi.org/10.1016/j.janxdis.2011.10.001

Evraire, L. E., Ludmer, J. A., & Dozois, D. J. A. (2014). The influence of priming attachment styles on excessive reassurance seeking and negative feedback seeking in depression. Journal of Social and Clinical Psychology, 33(4), 295–318.

https://doi.org/10.1521/jscp.2014.33.4.295

Forchuk, C. A., Plouffe, R. A., & Saklofske, D. H. (2021). Do you “like” me? The roles of Facebook reassurance seeking and attachment style on depression. Psychology of Popular Media, 10(2), 223–229. https://doi.org/10.1037/ppm0000312

Fowler, C., & Gasiorek, J. (2017). Depressive symptoms, excessive reassurance seeking, and relationship maintenance. Journal of Social and Personal Relationships, 34(1), 91–113. https://doi.org/10.1177/0265407515624265

Garcia, S. E., Tully, E. C., Tarantino, N., South, S., Iacono, W. G., & McGue, M. (2013). Changes in genetic and environmental influences on trait anxiety from middle adolescence to early adulthood. Journal of Affective Disorders, 151(1), 46–53. https://doi.org/10.1016/j.jad.2013.05.046

Hames, J. L., Hagan, C. R., & Joiner, T. E. (2013). Interpersonal processes in depression. Annual Review of Clinical Psychology, 9, 355–377. https://doi.org/10.1146/annurev-clinpsy-050212-185553

Joiner, T. E., Alfano, M. S., & Metalsky, G. I. (1992). When depression breeds contempt: Reassurance seeking, self-esteem, and rejection of depressed college students by their roommates. Journal of Abnormal Psychology, 101(1), 165–173. https://doi.org/10.1037/0021-843X.101.1.165

Joiner, T. E., Metalsky, G. I., Katz, J., & Beach, S. R. H. (1999). Depression and excessive reassurance seeking. Psychological Inquiry, 10(3), 269–278. https://doi.org/10.1207/S15327965PLI1004_1

Joiner, T. E., & Metalsky, G. I. (2001). Excessive reassurance seeking: Delineating a risk factor involved in the development of depressive symptoms. Psychological Science, 12(5), 371–378. https://doi.org/10.1111/1467-9280.00369

Johnson, S. M., & Greenman, P. S. (2006). The path to a secure bond: Emotionally focused couple therapy. Journal of Clinical Psychology: In Session, 62(5), 597–609. https://doi.org/10.1002/jclp.20251

Kane, L., Bahl, N., & Ouimet, A. J. (2018). Just tell me it’s going to be OK! Fear of negative evaluation may be more important than fear of positive evaluation in predicting excessive reassurance seeking. Canadian Journal

of Behavioral Science, 50(4), 217–225. https://doi.org/10.1037/cbs0000109

Katz, J., Petracca, M., & Rabinowitz, J. (2009). A retrospective study of daughters’ emotional role reversal with parents, attachment anxiety, excessive reassurance-seeking, and depressive symptoms. American Journal of Family Therapy, 37(3), 185–195. https://doi.org/10.1080/01926180802405596

Laurent, H., & Powers, S. (2007). Emotion regulation in emerging adult couples: Temperament, attachment, and HPA response to conflict. Biological Psychology, 76(1–2), 61–71. https://doi.org/10.1016/j.biopsycho.2007.06.002

Lord, K. A., Harkness, K. L., Suvak, M. K., & Stewart, J. G. (2020). The affective dynamics of reassurance-seeking in real-time interactions. Journal of Social and Personal Relationships, 37(2), 676–694. https://doi.org/10.1177/0265407519876991

McCarthy, G., & Maughan, B. (2010). Negative childhood experiences and adult love relationships: The role of internal working models of attachment. Attachment & Human Development, 12(5), 445–461. https://doi.org/10.1080/14616734.2010.501968

Merlo, L. J., & Lakey, B. (2007). Trait and social influences in the links among adolescent attachment, depressive symptoms, and coping. Journal of Clinical Child and Adolescent Psychology, 36(2), 195–206. https://doi.org/10.1080/15374410701277846

Mikulincer, M., Gillath, O., & Shaver, P. R. (2002). Activation of the attachment system in adulthood: Threat-related primes increase the accessibility of mental representations of attachment figures. Journal of Personality and Social Psychology, 83(4), 881–895.

https://doi.org/10.1037/0022-3514.83.4.881

Mushquash, A. R., & Sherry, S. B. (2012). Understanding the socially prescribed perfectionist’s cycle of self-defeat: A 7-day, 14-occasion daily diary study. Journal of Research in Personality, 46, 700–709. https://doi.org/10.1016/j.jrp.2012.08.006

Power, T. G. (2004). Stress and coping in childhood: The parents' role. Parenting: Science and practice, 4(4), 271–317.

https://doi.org/10.1207/s15327922par0404_1

Shaver, P. R., Schachner, D. A., & Mikulincer, M. (2005). Attachment style, excessive reassurance seeking, relationship processes, and depression. Personality and Social Psychology Bulletin, 31(3), 343–359.

https://doi.org/10.1177/0146167204271709

Skinner, E. A., & Zimmer-Gembeck, M. J. (2007). The development of coping. Annual Review of Psychology, 58, 119–144.

https://doi.org/10.1146/annurev.psych.58.110405.085705

Starr, L. R., & Davila, J. (2008). Excessive reassurance seeking, depression, and interpersonal rejection: A meta-analytic review. Journal of Abnormal Psychology, 117(4), 762–775. https://doi.org/10.1037/a0013866

Stewart, J. G., & Harkness, K. L. (2015). The interpersonal toxicity of excessive reassurance- seeking: Evidence from a longitudinal study of romantic relationships. Journal of Social and Clinical Psychology, 34(5), 392–410. https://doi.org/10.1521/jscp.2015.34.5.392

Taylor, D. L., Kraines, M. A., Grant, D. M., & Wells, T. T. (2019). The role of excessive reassurance seeking: An eye tracking study of the indirect effect of social anxiety symptoms on attention bias. Psychiatry Research, 274, 220–227. https://doi.org/10.1016/j.psychres.2019.02.039

Thompson, R. A., & Meyer, S. (2007). The socialization of emotion regulation in the family. In J. Gross (Ed.), Handbook of emotion regulation (pp. 249–268). Guilford Press.

Van Orden, K. A., & Joiner, T. E. (2006). The inner and outer turmoil of excessive reassurance seeking: From self-doubts to social rejection. In E. J. Finkel & K. D. Vohs (Eds.), Self and relationships: Connecting intrapersonal and interpersonal processes (pp. 104–129). Guilford Press.

Author Note. Jaidyn K. Charlton

https://orcid.org/0000­0003­0213­6174

Aislin R. Mushquash

https://orcid.org/0000­0002­4494­1326

We have no conflicts of interest to disclose. Correspondence concerning this article should be addressed to Jaidyn K. Charlton, Department of Psychology, Lakehead University, 955 Oliver Road, Thunder Bay, Ontario, Canada, P7B 5E1. Email: jkcharlt@lakeheadu.ca

161 COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL.

Publish Your Research in Psi Chi Journal

Undergraduate, graduate, and faculty submissions are welcome year round. Only one author (either first author or coauthor) is required to be a Psi Chi member. All submissions are free. Reasons to submit include

• a unique, doctoral­level, peer­review process

• indexing in PsycINFO, EBSCO, and Crossref databases

• free access of all articles at psichi.org

• our efficient online submissions portal View Submission Guidelines and submit your research at www.psichi.org/?page=JN_Submissions

Become a Journal Reviewer

Doctoral­level faculty in psychology and related fields who are passionate about educating others on conducting and reporting quality empirical research are invited become reviewers for Psi Chi Journal. Our editorial team is uniquely dedicated to mentorship and promoting professional development of our authors—Please join us!

To become a reviewer, visit www.psichi.org/page/JN_BecomeAReviewer

Resources for Student Research

Looking for solid examples of student manuscripts and educational editorials about conducting psychological research? Download as many free articles to share in your classrooms as you would like.

Search past issues, or articles by subject area or author at www.psichi.org/journal_past

Learn About Psi Chi

Psi Chi is the International Honor Society in Psychology. Membership is primarily open to undergraduates, graduate students, transfer students, full­time and part­time faculty members, and alumni.

See membership benefits and a link to apply at www.psichi.org/page/member_benefits

Register an account: http://pcj.msubmit.net/cgi-bin/main.plex