Journal of Media Psychology 1/2018 by Hogrefe

Volume 30 / Number 1 / 2018

Journal of

Media Psychology

Journal of Media Psychology

Editor-in-Chief Christoph Klimmt Associate Editors Nick D. Bowman Jesse Fox Nicole Krämer Diana Rieger Ivar Vermeulen

Theories, Methods, and Applications

Hogrefe OpenMind Open Access Publishing? It’s Your Choice! Your Road to Open Access Authors of papers accepted for publication in any Hogrefe journal can now choose to have their paper published as an open access article as part of the Hogrefe OpenMind program. This means that anyone, anywhere in the world will – without charge – be able to read, search, link, send, and use the article for noncommercial purposes, in accordance with the internationally recognized Creative Commons licensing standards.

The Choice Is Yours 1. Open Access Publication: The final “version of record” of the article is published online with full open access. It is freely available online to anyone in electronic form. (It will also be published in the print version of the journal.) 2. Traditional Publishing Model: Your article is published in the traditional manner, available worldwide to journal subscribers online and in print and to anyone by “pay per view.” Whichever you choose, your article will be peer-reviewed, professionally produced, and published both in print and in electronic versions of the journal. Every article will be given a DOI and registered with CrossRef.

www.hogrefe.com

How Does Hogrefe’s Open Access Program Work? After submission to the journal, your article will undergo exactly the same steps, no matter which publishing option you choose: peer-review, copy-editing, typesetting, data preparation, online reference linking, printing, hosting, and archiving. In the traditional publishing model, the publication process (including all the services that ensure the scientific and formal quality of your paper) is financed via subscriptions to the journal. Open access publication, by contrast, is financed by means of a one-time article fee (€ 2,500 or US $3,000) payable by you the author, or by your research institute or funding body. Once the article has been accepted for publication, it’s your choice – open access publication or the traditional model. We have an open mind!

Journal of

Media Psychology Theories, Methods, and Applications

Volume 30/ Number 1/2018

Editor-in-Chief Editorial Assistants

Associate Editors

Editorial Board

Publisher

Production Subscriptions Advertising/Inserts ISSN Copyright Information

Publication Subscription Prices

Payment

Electronic Full Text Abstracting/Indexing Services

Journal of Media Psychology (2018), 30(1)

Christoph Klimmt, Department of Journalism and Communication Research, Hanover University of Music, Drama and Media, 30539 Hanover, Germany (E-mail christoph.klimmt@ijk.hmtm-hannover.de) Sophie Bruns, Katharina Emde-Lachmund, Katharina Knop-Hülß, and Sabine Reich, Department of Journalism and Communication Research, Hanover University of Music, Drama and Media, 30539 Hanover, Germany (E-mail jmp@ijk.hmtm-hannover.de) Nick D. Bowman, West Virginia University, Morgantown, VA, USA, E-mail Nicholas.Bowman@mail.wvu.edu Jesse Fox, The Ohio State University, Columbus, OH, USA, E-mail fox775@osu.edu Nicole Krämer, University Duisburg-Essen, Duisburg, Germany, E-mail nicole.kraemer@uni-due.de Diana Rieger, University of Mannheim, Germany, E-mail diana.rieger@uni-mannheim.de Ivar Vermeulen, Vrije Universiteit Amsterdam, The Netherlands, E-mail i.e.vermeulen@vu.nl Annie Lang (Bloomington, IN, USA) Markus Appel (Koblenz-Landau, Germany) Florian Arendt (München, Germany) Eun-Ju Lee (Seoul, South Korea) Jörg Matthes (Vienna, Austria) Omotayo Banjo (Cincinnati, OH, USA) Peter Nauroth (Marburg, Germany) Anne Bartsch (München, Germany) Anne Oeldorf-Hirsch (Mansfield, CT, USA) Gary Bente (Cologne, Germany) Jochen Peter (Amsterdam, Netherlands) Paul Bolls (Columbia, MO, USA) Johannes Breuer (Cologne, Germany) Daniel Pietschmann (Chemnitz, Germany) Robert F. Potter (Bloomington, IN, USA) Jennings Bryant (Tuscaloosa, AL, USA) Arthur A. Raney (Tallahassee, USA) Caleb Carr (Normal, IL, USA) Elizabeth Cohen (Morgantown, WV, USA) Leonard Reinecke (Mainz, Germany) Meghan Sanders (Baton Rouge, LA, USA) Enny Das (Nijmegen, The Netherlands) Frank Schwab (Würzburg, Germany) Kevin Durkin (Glasgow, UK) Stephan Schwan (Tübingen, Germany) Allison Eden (Amsterdam, The Netherlands) Michael Slater (Columbus, OH, USA) Nicole Ellison (Ann Arbor, MI, USA) Kaveri Subrahmanyam (Los Angeles, CA, USA) Malte Elson (Bochum, Germany) Ron Tamborini (East Lansing, MI, USA) David Ewoldsen (Columbus, OH, USA) Catalina Toma (Madison, WI, USA) Christopher Ferguson (DeLand, FL, USA) Sabine Trepte (Hohenheim, Germany) Jesse Fox (Columbus, OH, USA) Mina Tsay-Vogel (Boston, MA, USA) Sabine Glock (Wuppertal, Germany) Dagmar Unz (Würzburg-Schweinfurt, Germany) Melanie Green (Chapel Hill, NC, USA) Sonja Utz (Tübingen, Germany) Matthew Grizzard (Buffalo, NY, USA) Sebastián Valenzuela (Santiago de Chile, Chile) Dorothée Hefner (Hannover, Germany) Brandon van der Heide (East Lansing, MI, USA) Shirley S. Ho (Singapore) Christian von Sikorski (Vienna, Austria) Matthias Hofer (Zurich, Switzerland) Peter Vorderer (Mannheim, Germany) Juan José Igartua (Salamanca, Spain) Patrick Weber (Hohenheim, Germany) Jimmy Ivory (Blacksburg, VA, USA) René Weber (Santa Barbara, CA, USA) Jeroen Jansz (Rotterdam, The Netherlands) Stephan Winter (Duisburg-Essen, Germany) Julia Kneer (Rotterdam, The Netherlands) Mike Yao (Hong Kong, ROC) Elly Konijn (Amsterdam, The Netherlands) Maja Krakowiak (Colorado Springs, CO, USA) Hogrefe Publishing, Merkelstr. 3, 37085 Göttingen, Germany, Tel. +49 551 99950-0, Fax +49 551 99950-425, E-mail publishing@hogrefe.com, Web www.hogrefe.com North America: Hogrefe Publishing, 7 Bulfinch Place, 2nd floor, Boston, MA 02114, USA, Tel. (866) 823-4726, Fax (617) 354-6875, E-mail publishing@hogrefe.com Regina Pinks-Freybott, Hogrefe Publishing, Merkelstr. 3, 37085 Göttingen, Germany, Tel. +49 551 99950-0, Fax +49 551 99950-425, E-mail production@hogrefe.com Hogrefe Publishing, Herbert-Quandt-Str. 4, 37081 Göttingen, Germany, Tel. +49 551 99950-999, Fax +49 551 99950-998, E-mail zeitschriftenvertrieb@hogrefe.de Melanie Beck, Hogrefe Publishing, Merkelstr. 3, 37085 Göttingen, Germany, Tel. +49 551 99950-423, Fax +49 551 99950-425, E-mail marketing@hogrefe.com ISSN-L 1864-1105, ISSN-Print 1864-1105, ISSN-Online 2151-2388 Ó 2018 Hogrefe Publishing. This journal as well as the individual contributions and illustrations contained within it are protected under international copyright law. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without prior written permission from the publisher. All rights, including translation rights, reserved. Published in 4 issues per annual volume. The Journal of Media Psychology is the continuation of Zeitschrift für Medienpsychologie (ISSN 1617-6383), the last annual volume of which (Volume 19) was published in 2007. Calendar year subscriptions only. Rates for 2018: Institutions: US $372.00 / 292.00; Individuals: US $195.00 / 139.00 (all plus US $16.00 / 12.00 postage & handling). Single issue US $93.00 / 73.00 (plus postage & handling) Payment may be made by check, international money order, or credit card, to Hogrefe Publishing, Merkelstr. 3, 37085 Göttingen, Germany. US and Canadian subscriptions can also be ordered from Hogrefe Publishing, 7 Bulfinch Place, 2nd floor, Boston, MA 02114, USA. The full text of Journal of Media Psychology is available online at http://econtent.hogrefe.com and in PsycARTICLES. Abstracted/indexed in Current Contents/Social and Behavioral Sciences (CC/S&BS), Social Sciences Citation Index (SSCI), IBR, IBZ, PsycINFO, PsycLit, PSYNDEX, and Scopus. Impact Factor (2016): 1.057

Ó 2018 Hogrefe Publishing

Contents Editorial

Goodbye and Welcome: Time Again for Editorial Transitions

Nicole Krämer and Christoph Klimmt Original Articles

Interactive Experience and Identiﬁcation as Predictors of Attributing Responsibility in Video Games Nathan Walter and Yariv Tsfati

Serious Efforts at Bias Reduction: The Effects of Digital Games and Avatar Customization on Three Cognitive Biases Adrienne Shaw, Kate Kenski, Jennifer Stromer-Galley, Rosa Mikeal Martey, Benjamin A. Clegg, Joanna E. Lewis, James E. Folkestad, and Tomek Strzalkowski

Effects of Supportive Feedback Messages on Exergame Experiences: A Mediating Role of Social Presence Jihyun Kim and C. Erik Timmerman

Commitment to the Team: Perceived Conﬂict and Political Polarization

Bryan McLaughlin Meeting Calendar

Ó 2018 Hogrefe Publishing

Journal of Media Psychology (2018), 30(1)

Editorial Goodbye and Welcome Time Again for Editorial Transitions Nicole Krämer1 and Christoph Klimmt2 1

Department of Social Psychology – Media and Communication, University of Duisburg-Essen, Duisburg, Germany

Department of Journalism and Communication Research (IJK) of Hanover University of Music, Drama, and Media, Hanover, Germany

In this editorial, we would first like to look back on the past 3 years of the Journal of Media Psychology (JMP) under the editorship of Nicole Krämer and then welcome Christoph Klimmt, in his new role as Editor-in-Chief. The 3 years as editor-in-chief were incredibly exciting and rewarding. They were filled with lots of decisions (most of them, however, difficult because not all papers worthy of acceptance could have been published), with rewarding meetings with our excellent associate editors to discuss the course of the journal, and with numerous friendly interactions with our excellent reviewers who sacrificed their time to give valuable feedback to the authors who were willing to revise their papers according the reviewers’ suggestions. We trust that this process yielded many wonderful papers that help to establish the Journal of Media Psychology as an important and respected journal in the field. One important goal of the editorial term was to improve the scientific publishing in terms of transparency and rigorousness. Most importantly, as the first journal in the area of media psychology and communication, we established the submission of pre-registered reports. By this, we hope to combat publication bias by publishing well conducted but potentially non-significant studies and to fight questionable practices such as HARKing. We were thrilled to see that within these 3 years as many as 21 manuscripts were submitted as pre-registered reports – also thanks to the special issue edited by Malte Elson and Andrew Przybylski and published in 2017. In line with this, we successfully applied to become an open science journal according to level 1 of the TOP guidelines of the Center for Open Science. As reported in the last editorial, we saw a steady increase of submissions in the last 3 years. This is also due to the increasing internationality and visibility of the journal. Each year we received manuscripts from around 35 countries. In result, we now have a low acceptance rate of 10.08% (referring to the period 1/2015–12/2017). In order to be able Ó 2018 Hogrefe Publishing

to give timely feedback to the authors, we increased the rate of desk-rejects from 43% in 2015 to 59% in 2017. Including those desk rejects, we needed on average 51 days to make the first editorial decision. In order to keep the turn-around time low, we need committed experts who review incoming manuscripts and give advice for improvement in a timely manner. In order to secure a basis for this, we significantly extended the Editorial Board from 31 members (in 2015) to 59 members (24 females, 35 males) in 2017. Also, we implemented a measure to reward the reviewers’ commitment and their excellence in helping to improve the journal’s quality: Based on votes among the Associate Editors, we identified approximately 15 people every year as distinguished reviewers (see their names in every year’s last issue). We believe that this is a great instrument to maintain the journal’s excellence and make reviewer service a rewarding experience. I would like to take the opportunity to thank the many contributors to the success of JMP: the readers, the reviewers, the editorial board and the associate editors, and of course the authors of all the interesting papers we were happy to read and evaluate. I thank the team at Hogrefe Publishing (especially Robert Dimbleby, Regina PinksFreybott, and Juliane Munson) for the friendly and trustful cooperation and their truly professional service. My special thanks go out to my editorial assistant, Dr. German Neubaum, who worked incredibly hard to screen all papers, prepare allocations and decisions and to support me regarding decisions on the journal’s course. Without him, the journal would not thrive as much as it does. It is a great pleasure to hand over the journal now to a scholar who is not only an excellent researcher himself, but has also been engaged in developing and fostering new ideas to optimize scientific publishing for years. It is my honor to present Christoph Klimmt as the next Editor-in-Chief of the Journal of Media Psychology. Journal of Media Psychology (2018), 30(1), 1–2 https://doi.org/10.1027/1864-1105/a000242

Official Handover JMP is in good shape and it is both a honor and a pleasure for me to follow Nicole Krämer as Editor-in-Chief. The excellent condition of the journal is the consequence of a flourishing media psychology community, and at the same time it is the result of the remarkably dilligent management excercised by Nicole Krämer and her editorial assistant, German Neubaum. Both deserve our strong appreciation, as they have shown brave and sustained engagement for the journal as a vital element of the international research community of media psychology. The same applies to our current associate editors Nick Bowman, Jesse Fox, Diana Rieger, and, as a new member of the team, Ivar Vermeulen from VU University Amsterdam. The entire community – and me in particular – are thankful for their contribution to keep the publication activities of media psychologists running and to maintain the high standards of academic quality assurance that JMP is dedicated to. Finally, JMP is enjoying its excellent standing because of the many productive authors and thorough, hard-working reviewers – that is, You, dear readers of JMP! A warm expression of gratitude is also indicated with regard to our publisher Hogrefe, whose team has been strongly engaged in running and developing the journal. The transition from one Editor-in-Chief to the next is always a good opportunity to announce changes and innovations. However, I understand that my primary mission as Nicole Krämer’s successor is to achieve a continuation of the excellent service that the journal has been providing in the past to authors and readers and to make sure that JMP will publish many more useful, scientifically excellent, and socially relevant contributions to research in media psychology. Hence, in terms of operations and procedures, there will be little (if any) change associated with the editor transition. However, new people are being involved in the journal’s operations, and I am happy and grateful to introduce our new editorial assistants: Sophie Bruns, Katharina Emde-Lachmund, and Katharina KnopHülß as doctoral students, and Sabine Reich, who is working as post-doctoral researcher at the Department of Journalism and Communication Research (IJK) of Hanover University of Music, Drama, and Media. They will serve as interface between the community and the journal, and I am thankful to each of them for their willingness to engage for the journal and to IJK for enabling and supporting the new editorial team in Hanover. Many issues related to the publication culture in the social sciences are subject to discussions, and the editors of JMP will continue to observe these discourses and to consider useful innovations in the procedures and formats of academic publishing. Our initial experiences with pre-registered reports, for instance, are positive and deserve a Journal of Media Psychology (2018), 30(1), 1–2

Editorial

continuation of this scheme in order to achieve an improved match of publication procedures with the epistemology behind psychological social science. What is more important to keep media psychology flourishing, however, is that You, dear readers, contributors, and reviewers, maintain your high level of innovativeness, intellectual reflection, theoretical rigor, methodological smartness, and vigilance for important developments in contemporary media societies. So, go on conducting and submitting your fine work! The entire JMP team is looking forward to receive your manuscripts and to serve the thriving international community of media psychology for the years to come.

Nicole Krämer Social Psychology – Media and Communication University of Duisburg-Essen Forsthausweg 2 Office LE 243 47057 Duisburg Germany nicole.kraemer@uni-due.de Christoph Klimmt Department of Journalism and Communication Research (IJK) Hanover University of Music, Drama, and Media Expo Plaza 12 30539 Hannover christoph.klimmt@ijk.hmtm-hannover.de

Nicole Krämer is professor for Social Psychology – Media and Communication at the University DuisburgEssen. She has a background in social and media psychology. Her research interests include humancomputer-interaction and computermediated-communication, especially social media. More specifically her research focuses on forms and effects of social media usage, related to impression management, self-disclosure or social comparison. She also analyses social effects of virtual agents and robots.

Christoph Klimmt (PhD, 2004) studied media management at the Department of Journalism and Communication Research (IJK) of Hannover University. From 2007 to 2010, he served as an assistant professor in the Department of Communication, University of Mainz. Since 2010, he has been a professor of communication science at IJK Hannover. His research interests include media effects and processes, entertainment, and digital games.

Ó 2018 Hogrefe Publishing

Original Article

Interactive Experience and Identiﬁcation as Predictors of Attributing Responsibility in Video Games Nathan Walter1 and Yariv Tsfati2 1

Department of Communication, University of Southern California, Los Angeles, CA, USA

Department of Communication, University of Haifa, Israel

Abstract: This study examines the effect of interactivity on the attribution of responsibility for the character’s actions in a violent video game. Through an experiment, we tested the hypothesis that identification with the main character in Grand Theft Auto IV mediates the effect of interactivity on attributions of responsibility for the main character’s antisocial behavior. Using the framework of the fundamental attribution error, we demonstrated that those who actually played the game, as opposed to those who simply watched someone else playing it, identified with the main character. In accordance with the theoretical expectation, those who played the game and came to identify with the main character attributed the responsibility for his actions to external factors such as “living in a violent society.” By contrast, those who did not interact with the game attributed responsibility for the character’s actions to his personality traits. These findings could be viewed as contrasting with psychological research suggesting that respondents should have distanced themselves from the violent protagonist rather than identifying with him, and with Iyengar’s (1991) expectation that more personalized episodic framing would be associated with attributing responsibility to the protagonist. Keywords: video games, interactivity, FAE, human–computer interaction, identification

The flourishing of the multibillion dollar entertainment software industry has fueled social and scientific interest in this phenomenon (Daim, Justice, Hogaboam, Mäkinen, & Dedehayir, 2014). As early as 2001, video games were rated as the most entertaining medium, while television found itself in second place (International Digital Software Association, 2001). Currently, according to the Entertainment Software Association (ESA) 2014 report, 59% of Americans play video games, and contrary to the popular stigma, the average gamer is as diverse as the people who drink milk. Moreover, it seems that video games are not only more entertaining than television, but they are also more violent. Indeed, studies show that violence in video games is more frequent, more brutal, and more often rewarded than violence in primetime television (Greitemeyer & Mügge, 2014; Weber, Ritterfeld, & Kostygina, 2006). However, the extremely violent imagery of popular video games is just one characteristic of these media. In fact, the presumed ability of video games to allow the player to personify immoral characters and act out gruesome and realistic virtual violence is another reason for the moral panic over the use of video games. One of the recent examples Ó 2016 Hogrefe Publishing

of this panic occurred in the wake of the launching of a new version of the popular adventure video game, Grand Theft Auto V. Interestingly, the outrage over this game came about not merely for its drastic violence or even its overt misogyny, but, supposedly, because the game offers the player a first-person experience in objectifying and humiliating women, with a wide plethora of options and scenarios to achieve this goal (Rodenberg, 2013). Hence, it is the coalescence of extreme levels of violence and the embodiment of vicious protagonists that critics find especially disturbing. In addition, media critics and policymakers who are deeply concerned with the deleterious effects of violence and interactivity seem to differentiate between video games and movies that display similar levels of violence, rendering the latter artistic (Martin, 2007). From this perspective, inquiry into the underlying mechanisms that govern antisocial in-game behavior is central for our understanding of whether such media differences in public perceptions are reasonable or not. In the current study, we investigate the effects of interactivity as a major characteristic of video games on the players’ identification with a violent character. We argue that Journal of Media Psychology (2018), 30(1), 3–15 DOI: 10.1027/1864-1105/a000168

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

the ability to control the characters and shape the narrative increases the players’ identification with them and consequently influences the players’ attributions of responsibility for the actions of the violent protagonists. The attribution of responsibility for the characters’ actions is a particularly relevant consequence of interactivity for two reasons. First, in order to enjoy popular video games that are often saturated with virtual violence and antisocial behavior, players need to rationalize their immoral choices. One of the mechanisms that deal with the management of negative behavior is the attribution of responsibility. Thus, the framework of attribution may provide a useful tool for viewing the process of managing socially disapproved behavior (Rudolph, 2003). Second, the interactive quality of video games makes it possible to establish a shared perspective between the player and the character (Klimmt, Hefner, & Vorderer, 2009). In the context of virtual violence, it is interesting to understand the role that identification with the character plays in affecting the attribution of responsibility for antisocial behavior. If players decide to attribute the responsibility to the character, it may force them to acknowledge their own role in it. By contrast, external attribution may serve as a specific strategy to manage moral concerns in order to make violence enjoyable (Klimmt, Schmid, Nosper, Hartmann, & Vorderer, 2006, 2008). This possibility is consistent with the psychological biases associated with the attribution of responsibility such as the fundamental attribution error (FAE). In this investigation, we examined the effects of interactivity on identification with and attribution of responsibility in the context of a popular violent game, Grand Theft Auto IV.

Interactivity in Video Games Interactivity is a feature that is unique to video games, compared with more traditional narrative communication formats, such as books, movies, and television programs. Over the years, researchers have suggested three separate definitions for the concept of interactivity (Lee & Peng, 2006). Technology-oriented definitions see interactivity as a component of communication technologies that allow users to affect and shape the content (Biocca, 1997; Sundar, 2004). Media-oriented definitions suggest that interactive processes happen when both the user and the medium influence each other simultaneously (Chaffee, Rafaeli, & Lieberman, 1985; Kiousis, 2002; Rafaeli, 1988). Finally, user-oriented definitions emphasize the audience’s ability to be both the producers and recipients of information as the purist expression of interactivity (Lee, Park, & Jin, 2006; Leiner & Quiring, 2008). In the context of explaining the appeal of video games to adolescents, Raney, Smith, and Baker (2006) underscored selectivity, modification, tailoring, linearity, and control devices as the essential Journal of Media Psychology (2018), 30(1), 3–15

aspects of the interactive experience. More recently, Weber, Behr, and DeMartino (2014) proposed differentiating between play-related interactivity and non-play-related interactivity, analyzing interactions with the game setting and actual game playing as separate constructs (p. 85). Beyond their theoretical importance, definitions of interactivity are consequential for methodology as well. While most studies focus on the amount of time that players spend on video games as a predictor of their effects (Adachi & Willoughby, 2011; Anderson & Bushman, 2001), this study posits the level of interactivity as a variable that shapes the players’ experience and, as a consequence, its effect on them. Interactivity should be a more accurate predictor of the effects of video games than the amount of time spent on them, because it considers the different levels of participation that stem from this unique technological feature (DeVane & Squire, 2008). In order to conduct a systematic comparison of interactive game play with noninteractive experiences, we followed Klimmt, Vorderer, and Ritterfeld’s (2007) call for a research design that compares participants who either played a game or just watched a recorded sequence of the same game. These artificial conditions highlight the differences between playing and merely watching, crystallizing the essential characteristics of video game interactivity. According to Klimmt, Vorderer, et al., “[we] need to think more carefully about which of interactive media’s properties are susceptible to the uncontrolled variation evoked by interactivity” (p. 174). Hence, watching a narrative action game unfold is an entertaining experience even when someone else is playing the game, because it involves suspense, humor, and transportation into a story, much like watching TV or a movie. However, just watching implies the inability to shape the narrative, which is at the core of interactivity.

Identification With Characters Cohen (2001, pp. 250–251) defined identification as an imaginative process that takes place when an audience member imagines himself or herself being a mediated character, adopts the perspective of the character, and experiences the feelings the character would experience in reaction to the events in the narrative. In the context of video games, laboratory studies have demonstrated that players identify with game characters because, as the result of a role offered by the game, the players change their selfconcept by perceiving themselves as the character (Klimmt et al., 2009; Konijn, Nije Bijvank, & Bushman, 2007). Unlike noninteractive communication technologies, video games not only introduce the characters and their environment, but they also invite players to actively influence and shape the virtual world. Consequently, through interactivity, players are able to bridge the gap between fantasy Ó 2016 Hogrefe Publishing

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

and reality, as they take on an active role in the virtual world and temporarily regard the goals of the character as if they were their own (Li, Liau, & Khoo, 2013). Players not only observe the characters on screen but they can also animate their avatars (Fischer, Kastenmüller, & Greitemeyer, 2010; Klimmt et al., 2009). This unique feature of video games seems to maximize the dyadic identification process explicated by Cohen (2001), because during the game, there are no “two separate entities” consisting of players and characters (Klimmt et al., 2009). In fact, the player and the avatar are inseparable, for they actually share the same “self.” Unlike TV and film viewers or readers of books who may be psychologically active in making sense of the stories but physically passive, players of digital games take on the role of the characters (Cohen, in press; van Reijmersdal, Jansz, Peters, & van Noort, 2013). As in role-playing games, individuals who take on a virtual identity feel they are the character (Yoon & Vargas, 2014) and tend to see their character’s major physical and emotional features as more similar to their own (Klimmt et al., 2009). In addition, Kim, Lee, and Kang (2012) suggest that the users’ identification with their avatars is an important element of their satisfaction with the video game experience. Hence, it is not surprising that players tend to credit the success of their character to themselves, which increases their self-esteem. By contrast, when the character fails, the player’s self-esteem tends to decline (Klimmt et al., 2009; McDonald & Kim, 2001). Thus, although the traditional definition of identification was based on models of the audience–character relationship developed in research on noninteractive entertainment such as novels and television, it has an even better theoretical fit with the context of the video game experience (Downs & Sundar, 2011). Identification implies that the player sees the world through the eyes of the character, feeling that he or she is the character (Cohen, 2001; Downs & Sundar, 2011). In a negative context, identification with a character has to do with the ability to rationalize the character’s actions, establishing a bond of empathy between the player and the character, not necessarily a bond of admiration or resemblance (Funk, Baldacci, Pasold, & Baumgardner, 2004). The player is often the active aggressor, generating negative emotions and trying to kill the enemy (Lin, 2013a, p. 537). Given that the player portrays a protagonist who acts immorally, in order to enjoy the game, the player is encouraged to acknowledge the complexities of the situation and suspend his or her moral judgment. Based on, Klimmt and colleagues’ (2009) explication of identification in the context of video games, we consider identification as a process whereby a player is able to forget himself or herself during the game and adopt the perspective and goals of the character by sharing the same “self” (p. 42). Ó 2016 Hogrefe Publishing

If the interactive nature of video games promotes identification with the characters regardless of their violence or immoral actions, we might ask – what are the consequences of this pairing? Focusing on the notion of the enjoyment of video games, Hartmann and Vorderer (2010) argue that portraying virtual violent characters offers pleasurable gratification (i.e., makes us feel superior and powerful) and/or serves a mood-regulation function (i.e., stimulates excitement). While this line of research has produced numerous insights into the process of moral disengagement in video games, it downplays the role of identification with the immoral character as a primary motivation to shift the focus of attention from the violent actions to the violent environment. Echoing Bandura’s concept of moral disengagement (2002), Klimmt et al. (2008) argue that, “consumers of violent media entertainment apply [different] strategies to cope with moral concerns” (p. 313). Of the cognitive strategies proposed, the diffusion of responsibility (i.e., “the individual responsibility for violence is transferred to others”) and attribution of blame (i.e., “justifies violence by arguing that the target of violent action deserves nothing but violence”) are of particular relevance to video game research. Specifically, both coping strategies can result from the interactive nature of video games and identification with their characters. Keeping in mind that the interactive nature of video games might lead players to identify strongly with their characters, when the latter make immoral choices or behave violently, the player becomes an accomplice rather than a passive bystander. Hence, to manage this immoral behavior, the player shifts the focus away from the character and points the finger of blame at others. In other words, when the responsibility for the behavior is attributed to external factors rather than the actions of the player or the character, video game violence becomes manageable (Hartmann, Toz, & Brandon, 2010; Hartmann & Vorderer, 2010). This conclusion concurs with the theoretical expectations of attribution theory in social psychology, namely, the assumption that the way in which individuals allocate responsibility is a result of personal and environmental factors that do not necessarily rely on an objective judgment of the situation.

Attribution of Responsibility Attribution of responsibility is one of the central mechanisms that help us understand the social reality around us. In fact, by attributing responsibility for actions, we condemn certain behaviors while affirming other social norms. Iyengar (1991) has conducted the most comprehensive work on the attribution of responsibility in media research. He examined the ways in which variations in the framing of events in media lead to different allocations of responsibility. His research found that news coverage Journal of Media Psychology (2018), 30(1), 3–15

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

that focuses on the individual (i.e., episodic framing) rather than on an abstract representation of events (i.e., thematic framing) results in the perception that the individual in the story is responsible for his or her problems. For example, when unemployment was depicted using a “day in the life of the unemployed” frame, subjects tended to think that the unemployed individual was responsible for his situation, but when unemployment was portrayed using abstract statistical data about unemployment, subjects tended to think that the government was responsible. The importance of these studies lies not just in the revelation that a specific frame may affect the attribution of responsibility. This series of studies also underscore the possibility raised earlier that the attribution of responsibility for events portrayed in a narrative may not be a static conception that relies on rational assessment. Instead, it may be the result of biased processing that depends on situational cues and mental shortcuts rather than systematic reasoning. One of the most common attribution biases relates to a situation where people tend to overstate the importance of internal factors, while simultaneously reducing the importance of external factors (Ross, 1977). The fundamental attribution error (FAE) is stronger when it comes to others. However, when it comes to the individual’s own responsibility for the outcome of events, there are two distinct possibilities. In the case of failure, the responsibility lies with external factors, whereas in the case of success, the responsibility is attributed to internal factors. For instance, people tend to attribute the failure of others on a test to personal traits (e.g., “He is stupid”), but when we have to explain our personal failure on a similar test, we tend to blame it on external factors (e.g., “The test was unfair” or “I did not have enough time to study”). One explanation that has been proposed as a mechanism underlying the FAE has to do with attention and cognitive availability. According to this account, the FAE occurs in three stages (Gilbert, Pelham, & Krull, 1988). First, people recognize individual behavior. Second, people attribute this behavior to internal factors. Finally, people take into account the specific characteristics of the situation and change the attribution of responsibility accordingly. Whereas the first two steps do not require much cognitive investment, the last step is not automatic. Thus, taking this step actually forces us to assess the situation in depth. As a result, it is not surprising that without a certain level of motivation, the process of attribution will stop after the second stage, leading to internal attribution. When the object of the behavior is the individual himself or herself, stage three becomes an option that increases the chances of reducing the FAE and ascribing responsibility to external factors rather than to personal traits. The FAE seems particularly well suited to video game research for two reasons. First, quantitative and qualitative Journal of Media Psychology (2018), 30(1), 3–15

analyses of their content suggest that the most popular games are replete with antisocial representations and moral dilemmas that force the player to make unpopular choices (Chan & Vorderer, 2006; Smith, 2006). Players sometimes injure and kill others without any real reason or justification (Hartmann et al., 2010). As mentioned, in order to enjoy violent video games, players need to rationalize or manage moral concerns (Klimmt et al., 2006, 2008) that would otherwise inhibit their enjoyment. Put differently, the question of responsibility is a cardinal component of video games, because the player, occasionally as an observer and sometimes as an active participant, carries out antisocial acts in order to progress in the game. Lack of compliance with the rules of “video game reality” is often not an option. By choosing not to participate in the violence, the players receive immediate negative feedback in the form of failure or boredom (Hefner, Klimmt, & Vorderer, 2007). Second, video games provide an interesting platform for the study of the FAE because they introduce the element of interactivity. By allowing the player to act out the violence physically, rather than passively observing these behaviors, interactivity blurs the boundaries between the player and the character, increasing their shared responsibility, which could perhaps morally implicate the player (Gollwitzer & Melzer, 2012). Thus, the players’ “self” is more relevant when it comes to video games compared with noninteractive media, suggesting that the FAE could potentially play an important role in managing violent experiences.

The FAE and Identification With Video Game Characters Considering that the FAE stems from our lack of motivation to understand others, this study focused on identification as a relevant factor in establishing mutual goals and understanding the behavior of others, even when that behavior is considered socially unacceptable. As in situations where people attribute responsibility for the negative outcomes of their own behavior to external factors, we expect the individuals who personify a violent character to be motivated to attribute responsibility for the game’s violence to external factors. This expectation accords with findings that examined demographic similarity as the basis for a reduction in the FAE. Burger (1981) demonstrated that in the case of a car accident in which there was some resemblance between the subject and the driver, the former tended to hold the driver less accountable for the accident. Nevertheless, the effect in our research should be stronger, because video games encourage higher levels of identification and intimacy that focus not only on demography but also on emotional bonds. In fact, Malle’s (2006) meta-analysis of Ó 2016 Hogrefe Publishing

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

attribution biases suggests a stronger effect when the observer is able to establish an intimate connection with the actor. In other words, while similarity-based identification can also be experienced when just watching, hearing, or reading about a character, the first-person experience with a character through a gaming console could initiate much higher levels of intimacy (Lin, 2013a, 2013b). This expectation is consistent with Peng, Lee, and Heeter (2010), who found that game playing resulted in greater role taking than game watching and text reading. Moreover, the theoretical process posited in this study could serve as an alternative explanation for the moral disengagement effect in video games (for a summary, see Hartmann, Krakowiak, & Tsay-Vogel, 2014). In the context of violence and guilt, Hartmann, Toz, and Brandon (2010) suggested that, “justification of virtual violence and users’ trait empathy determined guilt in a structurally similar way to real-world scenarios. . . people feel guilt if they engage in unjustified virtual violence” (p. 339). According to this approach, moral disengagement cues are embedded in the content itself (Hartmann et al., 2014, p. 310). The current study complements these lines of research by maintaining that similar to real-life attribution of responsibility, the question of justified versus unjustified violence is associated both with the information presented (i.e., situational cues) and the different points of view that the player might adopt. In other words, the unique feature of interactive media is the fact that our reading of the text results from combining the information presented through the narrative with our game-playing experience. For example, when reading about a burglary, we rely only on textual cues to interpret the situation. However, when we actually experience a burglary, what is considered appropriate, largely depends on whether we are the burglar, the property owner, the policeman, or a bystander. As mentioned, when the analyzed behavior holds negative valence, agents exhibit a tendency toward external attribution whereas observers tend to emphasize internal factors. Specifically, in order to rationalize their behavior, perpetrators of crime search for external justifications for their antisocial behavior, while observers reaffirm their morality and virtues by attributing responsibility to internal factors. To this end, interactive media are of particular interests to the FAE, as they have the ability to complicate questions of responsibility by offering multiple angles from which to approach the text (Klimmt, Hefner, Vorderer, Roth, & Blake, 2010). Therefore, it would be particularly interesting to examine the effect of interactivity on the attribution of responsibility with identification as a mediator. While we expect that lower levels of interactivity will result in a stronger motivation to attribute responsibility to internal factors, the interactive experience could evoke identification with the character, which will manifest itself in a deeper Ó 2016 Hogrefe Publishing

understanding of the character’s motivations and in external attributions of responsibility. Given this background, we posit the following hypotheses. Hypothesis 1 (H1): Interactivity will affect the attribution of responsibility for the character’s negative actions, such that participants who interact with a video game (a high degree of interactivity) will tend to attribute responsibility to external factors, while individuals who watch a video recording (no interactivity) will tend to attribute responsibility to internal factors. Hypothesis 2 (H2): Interactivity will affect identification with the character, such that for participants who interact with a video game (a high degree of interactivity), we expect stronger identification between the individual and the character in the game. Hypothesis 3 (H3): Identification with the character will mediate the effect of interactivity on the attribution of responsibility for the character’s negative actions.

Method To examine the effect of interactivity on the players and to measure the role that identification with the character plays in this process, we conducted a post-test-only experiment. We recruited 40 participants by posting adverts throughout the campus of an Israeli university. Although convenience samples may be less representative of the general population, in this case a student sample is actually quite appropriate because the students are compatible with the demographic characteristics of the average gamer, particularly with regard to their age and education (ESA, 2014; Lee & Peng, 2009).

Sample Our sample consisted of 40 students (55% males and 45% females) with a predominantly secular affiliation (secular: 65%, traditional: 27.5%, Orthodox Jewish: 5%, and ultraOrthodox Jewish: 2.5%) and higher education (academic degree: 70%, high school: 25%, and primary school: 5%). The mean age was 25.57 years (SD = 3.66). Participants reported an average of 1.67 weekly hours of exposure to video games (SD = 3.19), and, on average, tended to the center of the Israeli political map (M = 3.82, SD = 1.44, on a scale ranging from 1 = extreme left to 7 = extreme right). Although political affiliation and religiosity are not directly Journal of Media Psychology (2018), 30(1), 3–15

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

related to the gaming experience, the decision to measure these variables stems from the fact that our stimuli focus on immigration, which in the Israeli popular discourse is strongly related to political and religious ideology. The controversy surrounding illegal immigration in Israel has produced two distinct camps, with right-wing conservatives opposing illegal immigration and left-wing liberals willing to consider granting resident or refugee status to illegal immigrants, especially children. With the exception of sex, which we controlled for in the models that follow, the experimental conditions were virtually identical in the distribution of these variables.

Procedure First, our sample was randomly divided into 20 pairs in which each member of the pair was randomly assigned to one of the two conditions: being an active player in a video game or a passive viewer of a segment of a video game. Participants were unaware of the fact that they had been paired with another participant. Each participant was placed in a separate room equipped with a large TV screen. The only difference between the interactive condition and the noninteractive condition was the fact that the former had an active game controller, while the latter found themselves in front of a screen without a controller watching his or her other half of the pair play the game. It is important to note that each pair was exposed to the same content, because both screens were connected to the same gaming console via a wireless transmitter. Indeed, the sole difference between the two groups was the level of interactivity. Participants in the interactive condition were able to play and influence the game actively, while the participants in the noninteractive condition could only watch the game without actively participating in it. We designed this manipulation based on the axiom that the complexity of modern video games turns them into unique and unrepeatable experiences. Simply put, one cannot play the same game twice (DeVane & Squire, 2008). In addition, the fact that the experiment took place simultaneously in two separate rooms meant that the participants were unaware of the fact that they were actually watching a game someone else was playing in the next room. After 25 min of exposure to the game, all of the participants filled out an identical questionnaire that measured the dependent variable (i.e., attribution of responsibility), the degree of their identification with the main character in the game (i.e., mediator) and demographic variables.

Stimulus The game we chose for the experiment was Grand Theft Auto IV (GTA IV) from Rockstar North. It is part of a Journal of Media Psychology (2018), 30(1), 3–15

successful game franchise. Aside from its popularity, we chose GTA IV for several reasons. First, this game belongs to the genre of sandbox games that offer the highest degrees of freedom. Unlike games that have a strict linear plot that leads the player forward, sandbox games allow players to deviate from the main plot and engage in different subplots. Given our definition of interactivity, this feature made GTA IV a good context for examining its role (Steuer, 1992; Weber et al., 2014). Second, despite the high levels of violence, the protagonist of the game, Niko Bellic, is presented as a multilayered character with whom players could potentially sympathize and identify. In fact, the game emphasizes the “impossible” situation in which Bellic has found himself – an immigrant who wants to live a legitimate life, but soon finds himself taking part in criminal activities in order to survive. This form of ambivalence may encourage identification with the character, allowing us to test its role as a mediator. Furthermore, the attribution of responsibility is a cardinal element in the game, as there is an ambiguity about the question of blame. Does responsibility lie with the immigrant who makes violent choices or with the violent society that, by stigmatization, leaves him without any other option? The specific level we chose for the experiment is called “Uncle Vlad.” During this stage, Niko and his cousin, Roman, are kidnapped by the mafia. Niko’s only way to survive and save his relative’s life is to steal a police car, drive it to a particular spot in the virtual city, and rob a van transporting expensive merchandise. This scenario was chosen for several reasons. First, it offers a comprehensive introduction that presents the main characters and the central dilemma of the protagonist. As mentioned, this dilemma highlights the relevance of the game to the issue of responsibility. Importantly, it is virtually impossible to play this part of the game without engaging in or attempting to engage in antisocial behavior. Second, this level presents a relatively simple mission that allows even the most novice player to experience the game, along with its complexities. Finally, the level consists of three distinct sub-missions (stealing, driving, and robbing), which allow us to measure the participant’s level of gaming experience in order to use it as a control variable.

Measures After their experience with the game as a player or a viewer, the participants completed a self-report questionnaire that consisted of four parts. In the first part, an open-ended question asked the participants to describe their experience in the most accurate way they could, by referring to specific events from the narrative. We included the open-ended question to see whether the richness of the interactive experience was manifested in the players’ ability Ó 2016 Hogrefe Publishing

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

to describe the events in greater detail. This assumption was not supported because most of the answers were brief and not revealing (more than half of them consisted of a single sentence). Thus, we decided to drop the question from the analysis and use it only as a limited indicator of boredom. In the second part, the participants answered a series of questions about their level of identification with the main character in the game. Based on Hefner et al. (2007), identification with the game character was assessed with five items: “If Niko were physically wounded, it would make me feel bad,” “I feel close to Niko,” “Niko’s character influenced my personal experience,” “If Niko’s character was replaced, my experience would be different,” and “Niko’s goals became my own goals.” The participants were asked to rate their experience on a 7-point scale ranging from 1 (= completely disagree) to 7 (= completely agree). These five items loaded on a single factor in an EFA (M = 2.89, SD = 1.66, α = 0.87). Further items were used to measure video game identification taken from Van Looy, Courtois, De Vocht, & De Marez (2012), for which validity problems emerged that are discussed in more detail in the limitation portion of the discussion. In the third part of the questionnaire, the participants were asked to assign responsibility by indicating the degree to which they agreed with 11 statements, on a 7-point Likert scale, about responsibility for the situation presented in the game. Examples of these items include: “People like Niko need to use violence in order to survive,” “Niko’s situation is the result of bad governmental policy,” “Niko lives in a world where only the strong survive,” and “ Niko is responsible for his bad situation” (M = 2.65, SD = 0.90, α = 0.85; the items loaded on a single factor in an EFA). In order to determine whether the participants in the interactive condition were more likely to exhibit external attribution (i.e., blaming society), we recoded the scale to assess the degree of external attribution relative to internal attribution (i.e., blaming the character). Higher scores on the scale indicated external attributions, and lower scores indicated internal attributions. The last part of the questionnaire was devoted to an examination of different possible intervening variables. Given that past gaming experience is an important variable that may affect the success of the manipulation because the identification and enjoyment of an experienced player might be different from that of a novice, we measured previous experience with the game in terms of general hours of playing GTA IV (M = 5.42, SD = 17.86). We also measured various demographic and political variables, so that we could ensure that the random assignment to the interactive and noninteractive conditions was successful. Ó 2016 Hogrefe Publishing

Results H1 predicted that interactivity would affect the attribution of responsibility, such that the participants in the interactive condition would attribute more responsibility to external factors, while the participants in the noninteractive condition would attribute responsibility to internal factors. Given that the experimental and control participants were matched, a repeated-measures ANOVA was run with attribution of responsibility as the dependent variable. Consistent with the expectation of H1, the external attribution scores were significantly higher, Wilks’s Λ = .23, F(1, 18) = 59.16, p < .001, ηp2 = .77, in the interactive condition (M = 3.01, SE = .71) than in the noninteractive condition (M = 2.28, SE = .92). The difference between the experimental conditions remained significant even when controlling for the covariates (sex and level of gaming experience). We decided to control for sex because in the experimental group males slightly outnumbered females (12 and 8, respectively), as opposed to the control group in which there were an equal number of people of each sex (10 and 10). H2 predicted that interactivity would increase the participants’ identification with Niko. To test this hypothesis, a repeated-measures ANOVA was run with identification as the dependent variable. Consistent with the expectation of H2, the participants’ identification with Niko was significantly higher, Wilks’s Λ = .43, F(1, 18) = 24.09, p < .001, ηp2 = .57, in the interactive condition (M = 3.09, SE = 1.54) than in the noninteractive condition (M = 1.88, SE = 1.06). For H1 and H2, we re-ran the analysis as a between-subjects ANOVA, with the purpose of making sure that differences at the pair-level also persist when grouping the cases into experimental condition and control condition. Both hypotheses were supported; H1: F(1, 39) = 7.77, p < .05, ηp2 = .17; H2: F(1, 39) = 23.38, p < .05, ηp2 = .38. Again, the difference between the experimental conditions remained significant even when controlling for the covariates (sex and level of gaming experience). H3 predicted that the effect of interactivity on the attributions of responsibility would be mediated by identification with the character. To test this hypothesis, we used PROCESS, an SPSS macro utilizing OLS regression models and bootstrap estimation to test for the significance of the mediated effects (Hayes, 2013). PROCESS provides a bootstrap estimate of this indirect effect, together with a 95% confidence interval. The estimate for the mediated effect of the manipulation through identification was b = .48 (SE = .27, p < .05, LLCI = .01, ULCI = 1.11). Overall, holding sex and level of gaming experience constant, the mediation model accounted for 55.08% of the variance in the attributions of responsibility, F(4, 35) = 3.81, p < .05. As predicted, when adding identification to the repeated-measures Journal of Media Psychology (2018), 30(1), 3–15

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

Identification with a character .43** H2/H3

Interactive condition

.48* H3

H1 .44**

Attribution of responsibility

(.17) Figure 1. Beta coefficients for the relationship between interactivity and attribution of responsibility for the character’s actions as mediated by identification with the character, controlling for sex and level of gaming experience. We include the coefficients for the direct effect of interactivity on attribution of responsibility as they appear before (.44) and after the inclusion of the mediator. *p < .05. **p < .001.

model, the direct effect of interactivity became insignificant, whereas the indirect sequence from the manipulation to the attribution of responsibility through identification became significant. Taken together, our hypotheses yielded the theoretical model presented in Figure 1.

Discussion Attribution of Responsibility as an Outcome of Video Game Interactivity The prevalence of graphic violence in popular video games raises important questions regarding the psychological processes that underlie the mechanisms of moral disengagement and attribution of responsibility. When portraying a violent man, the players could have tried to distance themselves from the character by attributing more, not less, responsibility to Niko (because unlike in the control condition, they shared some of the blame for his violent and morally dubious deeds), or they could empathize and understand Niko because they were complicit in his actions. The players could have also distanced themselves from Niko by reporting less, not more, identification than the control group. According to research in social psychology, individuals distance themselves from people who display characteristics they fear in themselves (Schimel et al., 2000). Having just behaved aggressively could have induced the fear in the participants that they are aggressive (see Schimel et al., 2000). Consequentially, players would have reported feeling distance from rather than affinity and identification with Niko. However, our findings show that despite these psychological principles, the participants in the interactive condition reported a stronger identification with Niko. They adopted the relatively complex perspective of this poor, socially marginalized, aggressive immigrant and reported feeling closer to him. By contrast,

Journal of Media Psychology (2018), 30(1), 3–15

the participants in the noninteractive group reported much lower levels of identification with the character. Our findings not only suggest that simply by playing (rather than watching) a popular video game, the participants increased the attribution of responsibility for the protagonist’s violent behavior to external factors. We also show that the process underlying this effect has to do with the players’ identification with the character. The interactive experience not only places the player in the shoes of the character (similar to what happens when reading a good book or watching an engaging movie), but the complicity with their avatar is also part of the mechanism allowing the players to experience the reality of their avatar in first person. The players see the world through the character’s eyes, and they also attribute the same type of responsibility for the character’s behavior that they would make for their own.

Different Experiences Across Different Media Iyengar’s (1991) work on the attribution of responsibility in the context of mass media suggests that a focus on the individual (i.e., episodic framing), rather than on an abstract representation of events (i.e., thematic framing), leads to the perception that the individual has personal responsibility for his or her state. Our findings suggest a contrary pattern to the theoretical expectations that emerge from Iyengar’s study. While Iyengar demonstrates that the personal framing of unemployment leads to the attributing of responsibility to the unemployed (i.e., internal attribution), our findings show that the personal engagement with the violent immigrant leads to the attributing of responsibility to society (i.e., external attribution). One explanation for this difference has to do with the theoretical mechanism behind each process. In Iyengar’s case, it is the textual element (the frame) that encourages

Ó 2016 Hogrefe Publishing

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

a different assessment of the topic. In our study, it is the interactive experience that creates the connection between the player and the character. Another explanation for the apparent inconsistency between our findings and Iyengar’s work lies in the difference between watching the news and playing a video game. Even the most concentrated and focused viewer does not experience the news in the same way that a player experiences a video game. Although most of us can relate to the distress that unemployed people experience, watching a 5-min broadcast about unemployment, presumably, will not make us feel like “the other.” Nonetheless, that is exactly what video games seem to do through interactivity. This gap is illustrated in the distinction proposed by Klimmt et al. (2009) between “dyadic” and “monadic” identification. Simply put, a news story about an immigrant could make us feel close to the character on the screen, whereas interactive gaming experience would lead to “monadic” identification, which suggests the merging of the character and the players’ self into one perceived social entity (i.e., “I understand Niko Bellic” vs. “I am Niko Bellic”). To that end, the situation that Iyengar describes mirrors the process that occurred in our control group. In the noninteractive condition, the viewers’ identification with the immigrant was much lower, and so they tended to attribute responsibility to the character’s internal characteristics. Furthermore, while Iyengar compares different forms of framing (episodic and thematic), he does not differentiate between different modes of focus on the individual. Thus, although a variety of narratives may be used to depict unemployment, variations among the unemployed individuals are not considered. The results of our study show that within episodic framing, we must differentiate between different modes of individual foci. The fact that attribution of responsibility is experienced differently by viewers and players should also be consequential to the ways in which we address and label video game violence. Admittedly, the results of this study imply that critics of video games are not entirely wrong to denounce violent video games, while being much more tolerant toward movies that display similar levels of violence. Though it would be unfair to refer to every violent movie as “a work of art” and every violent video game as “a killing simulator,” it is hard to argue with the assertion that video games provide much more opportunities to rationalize violence, making it manageable, and even enjoyable.

External Attribution of Responsibility as Strategy of Moral Management As popular video games grow to be graphically realistic and imbued with violence than ever before, moral concerns gradually become an integral part of gaming experience

Ó 2016 Hogrefe Publishing

(Klimmt et al., 2006). The theoretical framework of moral disengagement outlines two mechanisms that help players attenuate their moral concerns (Hartmann & Vorderer, 2009). Specifically, the mechanisms include moral disengagement cues (e.g., “It’s OK to kill him because he enslaves people”) and reflexive processing (e.g., “This is merely a game”). Though these strategies can reduce negative emotions of guilt without damaging the enjoyment of the experience, they do not directly confront the question, “What is unique in moral management of violent video game experience?” In fact, these strategies equally pertain to any other entertainment media experience. The current study advances the literature on moral management by suggesting that the same interactivity that facilitates in-game immoral decision making also offers a way out. Indeed, by inviting players to share the experience with a violent character, interactivity prompts players to attribute responsibility to the character as if they were attributing responsibility for their own behavior. In the case of antisocial behavior, the end result of this process should be external attribution of responsibility, which alleviates negative emotions associated with violence. Accordingly, external attribution of responsibility for in-game immoral behavior functions as a moral management strategy that both justifies violence and preserves enjoyment. From this perspective, it would be interesting to examine whether user factors influencing guilt responses such as trait empathy can predict the tendency to attribute video game violence to external factors. Moreover, whereas past literature on gaming morality has emphasized disengagement cues that reside inside the text (Hartmann et al., 2014), our study demonstrates that moral disengagement may occur as a result of the shared perspective between the player and character, irrespective of the text. In fact, textual cues are less important in our framework, as they are kept constant between the conditions.

Limitations and Future Directions Although our design very cleanly parcels out the effect of interactivity while controlling for all other factors, this study is not free of methodological limitations. One could argue that the findings may be explained by boredom in the noninteractive control group, as opposed to the heightened enjoyment in the interactive condition. While we cannot rule out this possible criticism, our data provide some evidence against it. All of our participants were requested to answer an open-ended question asking them to describe their experience in the lab. Only three participants of the 20 in the interactive condition directly or indirectly expressed boredom, and removing these three did not change the pattern of results reported in this study.

Journal of Media Psychology (2018), 30(1), 3–15

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

Furthermore, three out of 20 was also the number of control (noninteractive) group participants who expressed boredom in the experimental session (probably as a result of watching someone having difficulty playing the game at a level that would allow them to enjoy it). Thus, boredom – at least boredom expressed by the responses to the openended questions – was equally distributed across groups. Nevertheless, with a growing body of research demonstrating the important role that enjoyment plays in shaping the video game experience (Klimmt, Hartmann, & Frey, 2007; Skalski, Tamborini, Shelton, Buncher, & Lindmark, 2010; Trepte & Reinecke, 2010), the confounding effect of boredom on our research design is quite possible. A second criticism deals with the small and homogeneous sample we used. Although student samples may be regarded as unrepresentative, we note Shapiro’s (2002) argument that convenience samples should not be automatically ruled out. Instead, we must consider whether the use of students poses any problems for testing our specific research questions. Keeping in mind that our study focuses on video games, students seem to constitute a very relevant social group as they correspond to the gamer population with respect to important variables such as age and education (Chan & Vorderer, 2006; Lee & Peng, 2006). Furthermore, at the end of each experimental trial, the participants were asked to guess the purpose of our experiment. Although some responses indicated a highly developed imagination, none of the participants guessed the real purpose of the experiment. An additional limitation of our design is related to the definition of identification. As mentioned, we based our operational definition of identification on a single-dimension scale (Hefner et al., 2007), corresponding in large part to avatar identification. Conversely, Van Looy et al. (2012) developed a more nuanced scale, emphasizing subscales of similarity, embodied presence, and wishful identification. Although the scale of Van Looy et al. (2012) offers a wider understanding of various sub-dimensions of identification, we decided to use the more parsimonious scale for several reasons. First, the scale of Hefner and colleagues (2007) is more relevant to cases where players identify with a single character (i.e., monadic identification), whereas the more nuanced scale is relevant (but obviously not restricted to) for measuring identification in massively multiplayer online games (MMOGs). Second, Van Looy and coworkers’ (2012) measurement raised many difficulties when trying to translate the individual items to Hebrew. In particular, in Hebrew, the concept of “embodied presence” is not associated with events occurring in a physical or virtual space (as the scale suggests) but rather with chronological relationships. Likewise, the concept of similarity is strongly associated in Hebrew with “imagination,” a process that was not emphasized in our stimuli. Nonetheless, in the original Journal of Media Psychology (2018), 30(1), 3–15

design of this study, we decided to use the scales of Hefner et al. (2007) and Van Looy et al. (2012) to test our hypotheses. Unfortunately, when testing our hypotheses with the Van Looy et al. (2012) scale, our prior concerns were confirmed, since embodied presence was the only subdimension that loaded separately from the other measures of identification. In addition, while embodied presence was positively associated with interactivity, the other associations were only borderline significant. Another important limitation deals with lack of recorded in-game players’ behavior. The availability of actual ingame behavior could have given us an important insight into the information on how antisocial the experimental game content was. Future research should try to record or observe actual in-game behavior for subsequent correlational analysis with outcome variables, such as identification, attribution of responsibility, or level of violence. Another methodological issue that we need to mention is the fact that our experiment focused only on a short-term effect (immediately after exposure to the game). It would have been interesting to assess whether interactivity influences our assessments in the long-term as well. We know that avid gamers devote many hours each day to the same characters. Therefore, it is likely that the pattern that we identified in our experiment could have an even stronger influence outside the laboratory. In order to confirm this hypothesis, we need to conduct further research that focuses on long-term effects and exposure in natural surroundings. The final limitation relates to the operationalization of interactivity. The pairing of the viewers and players was designed to keep the stimulus constant for each pair, while accounting for the interactive elements of the video game. However, it can be argued that our comparison demonstrated a difference between two starkly distinct conditions (playing vs. watching). One must raise the question of whether these results would hold with smaller degrees of difference, such as variations in the sense of control, customization, or other factors that might influence identification. Undoubtedly, answers to this question would promote a multidimensional approach to interactivity and a better understanding of the relation between interactive experience and identification.

Conclusion Beyond the new context for examining the FAE and interactivity, this study also introduces an additional element into the process of the attribution of responsibility. Specifically, there has not been enough emphasis on identification as a critical variable in the mechanism of the attribution of Ó 2016 Hogrefe Publishing

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

responsibility for other people’s behaviors. Almost a decade ago, Goldstein and Cialdini (2007) proposed the concept of vicarious self-perception to account for situations when sharing the same perspective with someone gives us the sensation that we know that person and that we are de facto experiencing his or her realities. Similarly, in our study, the interactive experience was able to bridge major social gaps, helping individuals from one group understand the complexities of individuals from another group. For example, playing an interactive video game might help Israeli undergraduates understand the dilemmas confronting immigrants from Eastern Europe. Given that interactivity allows us to encounter and even identify with characters who are often absent from the mass media or presented only in antisocial contexts, it becomes obvious that video games may have important social implications. This study offers a step forward in our long endeavor to understand the effects of video games on society, but the road is still long, and research in this direction must continue. Acknowledgments The authors would like to thank Jonathan Cohen, Asaf Segal, Nurit Tal-Or, Gabriel Weimann, and the two anonymous reviewers for their valuable comments, and Christoph Klimmt, whose dedication as Associate Editor greatly contributed to this manuscript.

References Adachi, P. J. C., & Willoughby, T. (2011). The effect of violent video games on aggression: Is it more than just the violence? Aggression and Violent Behavior, 16(1), 55–62. doi: 10.1016/ j.avb.2010.12.002 Anderson, C. A., & Bushman, B. J. (2001). Effects of violent video games on aggressive behavior, aggressive cognition, aggressive affect, physiological arousal, and prosocial behavior: A metaanalytic review of the scientific literature. Psychological Science, 12(5), 353–359. Bandura, A. (2002). Selective moral disengagement in the exercise of moral agency. Journal of Moral Education, 31(2), 101–119. Biocca, F. (1997). Cyborg’s dilemma: Progressive embodiment in virtual environments. Journal of Computer Mediated Communication, 3(2). Retrieved from http://www.ascusc.org/jcmc/ vol3/issue2/. doi: 10.1111/j.1083-6101.1997.tb00070.x Burger, J. M. (1981). Motivational biases in the attribution of responsibility for an accident: A meta-analysis of the defensiveattribution hypothesis. Psychological Bulletin, 90(3), 496. Chaffee, S., Rafaeli, S., & Lieberman, S. (1985, May). Human computer interactivity: A concept for communication research. Paper presented at the annual meeting of the International Communication Association, Chicago, IL. Chan, E., & Vorderer, P. (2006). Massively multiplayer online games. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 77–89). New York, NY: Routledge.

Ó 2016 Hogrefe Publishing

Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication and Society, 4(3), 245–264. doi: 10.1207/ S15327825MCS0403_01 Cohen, J. (2014). Identification with media characters. Oxford bibliographies in communication. doi: 10.1093/obo/ 9780199756841-0144. Daim, T., Justice, J., Hogaboam, L., Mäkinen, S. J., & Dedehayir, O. (2014). Identifying and forecasting the reverse salient in video game consoles: A performance gap ratio comparative analysis. Technological Forecasting and Social Change, 82, 177–189. doi: 10.1016/j.techfore.2013.06.007 DeVane, B., & Squire, K. D. (2008). The meaning of race and violence in Grand Theft Auto. Games and Culture, 3(3), 264–285. doi: 10.1177/1555412008317308 Downs, E., & Sundar, S. S. (2011). “We won” vs. “They lost”: Exploring ego-enhancement and self-preservation tendencies in the context of video game play. Entertainment Computing, 2(1), 23–28. Entertainment Software Association. (2014, April). The 2014 essential facts about the computer and video game industry. Retrieved from http://www.theesa.com/wp-content/uploads/ 2014/10/ESA_EF_2014.pdf Fischer, P., Kastenmüller, A., & Greitemeyer, T. (2010). Media violence and the self: The impact of personalized gaming characters in aggressive video games on aggressive behavior. Journal of Experimental Social Psychology, 46(1), 192–195. Funk, J. B., Baldacci, H. B., Pasold, T., & Baumgardner, J. (2004). Violence exposure in real-life, video games, television, movies, and the internet: is there desensitization? Journal of Adolescence, 27(1), 23–39. doi: 10.1016/j.adolescence.2003.10.005 Gilbert, D. T., Pelham, B. W., & Krull, D. S. (1988). On cognitive business: When person perceived meet persons perceived. Journal of Personality and Social Psychology, 54, 733–739. doi: 10.1037/0022-3514.54.5.733 Goldstein, N. J., & Cialdini, R. B. (2007). The spyglass self: A model of vicarious self-perception. Journal of Personality and Social Psychology, 92(3), 402. doi: 10.1037/0022-3514.92.3.402 Gollwitzer, M., & Melzer, A. (2012). Macbeth and the Joystick: Evidence for moral cleansing after playing a violent video game. Journal of Experimental Social Psychology, 48(6), 1356–1360. doi: 10.1016/j.jesp.2012.07.001 Greitemeyer, T., & Mügge, D. O. (2014). Video games do affect social outcomes: A meta-analytic review of the effects of violent and prosocial video game play. Personality and Social Psychology Bulletin. doi: 0146167213520459 Hartmann, T., Krakowiak, K. M., & Tsay-Vogel, M. (2014). How violent video games communicate violence: A literature review and content analysis of moral disengagement factors. Communication Monographs, 81(3), 310–332. doi: 10.1080/ 03637751.2014.922206 Hartmann, T., Toz, E., & Brandon, M. (2010). Just a game? Unjustified virtual violence produces guilt in empathetic players. Media Psychology, 13(4), 339–363. doi: 10.1080/ 15213269.2010.524912 Hartmann, T., & Vorderer, P. (2010). It’s okay to shoot a character: Moral disengagement in violent video games. Journal of Communication, 59, 865–890. doi: 10.1111/j.14602466.2009.01459.x Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford Press. Hefner, D., Klimmt, C., & Vorderer, P. (2007). Identification with the player character as determinant of video game enjoyment. In L. Ma, M. Rauterberg, & R. Nakatsu (Eds.), Entertainment Computing–ICEC 2007 (pp. 39–48). Heidelberg, Germany: Springer.

Journal of Media Psychology (2018), 30(1), 3–15

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

International Digital Software Association. (2001). Economic impact of the demand for playing interactive entertainment software. Washington, DC: Author. Iyengar, S. (1991). Is anyone responsible: How television frames political issues. Chicago, IL: University of Chicago Press. Kim, C., Lee, S.-G., & Kang, M. (2012). I became an attractive person in the virtual world: Users’ identification with virtual communities and avatars. Computers in Human Behavior, 28(5), 1663–1669. doi: 10.1016/j.chb.2012.04.004 Kiousis, S. (2002). Interactivity: a concept explication. New Media & Society, 4(3), 355–383. doi: 10.1177/ 146144402320564392 Klimmt, C., Hartmann, T., & Frey, A. (2007). Effectance and control as determinants of video game enjoyment. Cyberpsychology & Behavior: The Impact of the Internet, Multimedia and Virtual Reality on Behavior and Society, 10(6), 845–847. doi: 10.1089/ cpb.2007.9942 Klimmt, C., Hefner, D., & Vorderer, P. (2009). The video game experience as “true” identification: A theory of enjoyable alteration of players’ self-perception. Communication Theory, 19, 351–373. doi: 10.1111/j.1468-2885.2009.01347.x Klimmt, C., Hefner, D., Vorderer, P., Roth, C., & Blake, C. (2010). Identification with video game characters as automatic shift of self-perceptions. Media Psychology, 13(4), 323–338. Klimmt, C., Schmid, H., Nosper, A., Hartmann, T., & Vorderer, P. (2006). How players manage moral concerns to make video game violence enjoyable. Communications, 31(3), 309–328. doi: 10.1515/COMMUN.2006.020 Klimmt, C., Schmid, H., Nosper, A., Hartmann, T., & Vorderer, P. A. (2008). Moral management: Dealing with moral concerns to maintain enjoyment of violent video games. In A. SudmannJahn & R. Stockmann (Eds.), Computer games as a sociocultural phenomenon: Games without frontiers–wars without tears (pp. 108–118). Hampshire, UK: Palgrave. Klimmt, C., Vorderer, P., & Ritterfeld, U. (2007). Interactivity and generalizability: New media, new challenges. Communication Methods and Measures, 1(3), 169–179. doi: 10.1080/ 19312450701434961 Konijn, E. A., Nije Bijvank, M., & Bushman, B. J. (2007). I wish I were a warrior: The role of wishful identification in the effects of violent video games on aggression in adolescent boys. Developmental Psychology, 43(4), 1038. doi: 10.1037/00121649.43.4.1038 Lee, K. M., Park, N., & Jin, S. (2006). Narrative and interactivity in computer games. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 302–322). New York, NY: Routledge. Lee, K. M., & Peng, W. (2006). What do we know about social and psychological effects of computer games? A comprehensive review of the current literature. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 383–407). New York, NY: Routledge. Leiner, D. J., & Quiring, O. (2008). What interactivity means to the user. Essential insights into and a scale for perceived interactivity. Journal of Computer-Mediated Communication, 14(1), 127–155. doi: 10.1111/j.1083-6101.2008.01434.x Li, D. D., Liau, A. K., & Khoo, A. (2013). Player–avatar identification in video gaming: Concept and measurement. Computers in Human Behavior, 29(1), 257–263. Lin, J.-H. (2013a). Do video games exert stronger effects on aggression than film? The role of media interactivity and identification on the association of violent content and aggressive outcomes. Computers in Human Behavior, 29(3), 535–543. doi: 10.1016/j.chb.2012.11.001 Lin, J.-H. (2013b). Identification matters: A moderated mediation model of media interactivity, character identification, and video

Journal of Media Psychology (2018), 30(1), 3–15

game violence on aggression. Journal of Communication, 63(4), 682–702. doi: 10.1111/jcom.12044 Malle, B. F. (2006). The actor-observer asymmetry in attribution: A (surprising) meta-analysis. Psychological Bulletin, 132(6), 895. doi: 10.1037/0033-2909.132.6.895 Martin, B. (2007). Should videogames be viewed as art? In A. Clarke & G. Mitchell (Eds.), Videogames and art (pp. 201–210). Chicago, IL: The University of Chicago Press. McDonald, D. G., & Kim, H. (2001). When I die, I feel small: Electronic game characters and the social self. Journal of Broadcasting & Electronic Media, 45(2), 241–258. doi: 10.1207/ s15506878jobem4502_3 Peng, W., Lee, M., & Heeter, C. (2010). The effects of a serious game on role-taking and willingness to help. Journal of Communication, 60(4), 723–742. doi: 10.1111/j.14602466.2010.01511.x Rafaeli, S. (1988). Interactivity: From new media to communication. In R. P. Hawkins, J. M. Weimann, & S. Pingree (Eds.), Sage annual review of communication research: advancing communication science (Vol. 16, pp. 110–134). Beverly Hills, CA: Sage. Raney, A. A., Smith, J. K., & Baker, K. (2006). Adolescents and the appeal of video games. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 191– 209). New York, NY: Routledge. Rodenberg, C. (2013, December 27). Grand Theft Auto V makes it cool to pick up – even kill – prostitutes. The Guardian. Retrieved from http://www.theguardian.com/commentisfree/2013/dec/ 27/grand-theft-auto-v-prostitutes-killed Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 10, pp. 173– 220). New York, NY: Academic Press. Rudolph, T. J. (2003). Who’s responsible for the economy? The formation and consequences of responsibility attributions. American Journal of Political Science, 47(4), 698–713. Schimel, J., Pyszczynski, T., Greenberg, J., O'Mahen, H., & Arndt, J. (2000). Running from the shadow: Psychological distancing from others to deny characteristics people fear in themselves. Journal of Personality and Social Psychology, 78(3), 446. Shapiro, M. A. (2002). Generalizability in communication research. Human Communication Research, 28(4), 491–500. doi: 10.1111/j.1468-2958.2002.tb00819.x Skalski, P., Tamborini, R., Shelton, A., Buncher, M., & Lindmark, P. (2010). Mapping the road to fun: Natural video game controllers, presence, and game enjoyment. New Media & Society. doi: 1461444810370949 Smith, S. (2006). Perps, pimps, and provocative clothing: Examining negative content patterns in video games. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 57–76). New York, NY: Routledge. Steuer, J. (1992). Defining virtual reality: Dimensions determining telepresence. Journal of Communication, 42(4), 73–93. Sundar, S. S. (2004). Theorizing interactivity’s effects. The Information Society, 20(5), 385–389. doi: 10.1080/ 01972240490508072 Trepte, S., & Reinecke, L. (2010). Avatar creation and video game enjoyment. Journal of Media Psychology: Theories, Methods, and Applications, 22(4), 171–184. doi: 10.1027/1864-1105/ a000022 Van Looy, J., Courtois, C., De Vocht, M., & De Marez, L. (2012). Player identification in online games: Validation of a scale for measuring identification in MMOGs. Media Psychology, 15(2), 197–221. doi: 10.1080/15213269.2012.674917 van Reijmersdal, E. A., Jansz, J., Peters, O., & van Noort, G. (2013). Why girls go pink: Game character identification and

Ó 2016 Hogrefe Publishing

N. Walter & Y. Tsfati, Interactivity, identification and attribution of responsibility

game-players’ motivations. Computers in Human Behavior, 29 (6), 2640–2649. doi: 10.1016/j.chb.2013.06.046 Weber, R., Behr, K.-M., & DeMartino, C. (2014). Measuring interactivity in video games. Communication Methods and Measures, 8(2), 79–115. doi: 10.1080/19312458.2013.873778 Weber, R., Ritterfeld, U., & Kostygina, A. (2006). Aggression and violence as effects of playing violent video games? In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 408–426). New York, NY: Routledge. Yoon, G., & Vargas, P. T. (2014). Know thy avatar: The unintended effect of virtual-self representation on behavior. Psychological Science, 25(4), 1043–1045. doi: 10.1177/0956797613519271 Received August 9, 2014 Revision received May 5, 2015 Accepted June 20, 2015 Published online April 8, 2016

Nathan Walter is a doctoral student from the Annenberg School for Communication and Journalism at the University of Southern California. In his studies, he focuses on the psychology of video games, narrative persuasion and normative social influence.

Yariv Tsfati (PhD, 2001, University of Pennsylvania) is Associate Professor and Chair at the Department of Communication, University of Haifa. His research focuses on trust in media, the third person effect, and campaign effects. His research was funded by the Israel Science Foundation, the German-Israel Foundation, and other institutes.

Nathan Walter Department of Communication University of Southern California 3502 Watt Way Los Angeles, CA USA Tel. +1 626 390-1833 E-mail nathanw@usc.edu

Ó 2016 Hogrefe Publishing

Journal of Media Psychology (2018), 30(1), 3–15

Original Article

Serious Efforts at Bias Reduction The Effects of Digital Games and Avatar Customization on Three Cognitive Biases Adrienne Shaw,1 Kate Kenski,2 Jennifer Stromer-Galley,3 Rosa Mikeal Martey,4 Benjamin A. Clegg,5 Joanna E. Lewis,6 James E. Folkestad,7 and Tomek Strzalkowski8 1

Department of Media Studies and Production, Temple University, Philadelphia, PA, USA

Department of Communication, University of Arizona, Tucson, AZ, USA

School of Information, Syracuse University, Syracuse, NY, USA

Journalism and Media Communication Department, Colorado State University, Fort Collins, CO, USA

Psychology Department, Colorado State University, Fort Collins, CO, USA

Department of Psychology, Central Florida University, Orlando, FL, USA

School of Education, Colorado State University, Fort Collins, CO, USA

The Institute for Informatics, Logics, and Security Studies, University at Albany SUNY, Albany, NY, USA

Abstract: As research on serious games continues to grow, we investigate the efficacy of digital games to train enhanced decision making through understanding cognitive biases. This study investigates the ability of a 30-minute digital game as compared with a 30-minute video to teach people how to recognize and mitigate three cognitive biases: fundamental attribution error, confirmation bias, and bias blind spot. We investigate the effects of character customization on learning outcomes as compared with an assigned character. We use interviews to understand the qualitative differences between the conditions. Experimental results suggest that the game was more effective at teaching and mitigating cognitive biases than was the training video. Although interviews suggest players liked avatar customization, results of the experiment indicate that avatar customization had no significant effect on learning outcomes. This research provides information future designers can use to choose the best medium and affordances for the most effective learning outcomes on cognitive processes. Keywords: cognitive biases, educational games, game characteristics, character customization, digital games

For nearly as long as digital games have existed, researchers and educators have attempted to leverage the unique qualities of this medium for teaching (e.g., Gee, 2003; Greenfield, 1984), with the research on serious games continuing to grow (see Ritterfeld, Cody, & Vorderer, 2009). Digital games offer learning potential because of their interactivity – that is, being able to perform an action such as pushing a button and seeing the effects on a screen or by engendering interest because players can follow and sometimes control game narratives (Breuer & Bente, 2010). The pleasant frustration (Gee, 2003) of games can offer additional motivations for play and, subsequently, for learning (Przybylski, Deci, & Ryan, 2010). Yet questions remain unanswered as to how and when educational games can be effective training devices (Kirriemuir & McFarlane, 2004; Klopfer, Osterweil, & Salen, 2009; Shaw & Linebarger, 2008) especially for complex knowledge and skills that require transfer of learning beyond the materials encountered. To that end, this study answers the call for researchers to find ways of mitigating cognitive biases that can have deleterious effects on Journal of Media Psychology (2018), 30(1), 16–28 DOI: 10.1027/1864-1105/a000174

decision making, rather than continuing to find ways to elicit such biases (Larrick, 2004). We conducted an experiment to investigate whether a training game, as compared with a training video, could be used to teach people about three cognitive biases: fundamental attribution error (FAE), confirmation bias (CB), and bias blind spot (BBS). To explore the impact of game design decisions on learning, we also tested the impact of including character customization in the game. To complement the quantitative results, we used qualitative interviews to explore if and how character customization mattered to participants. Our study provides evidence that games produce strong learning effects on some cognitive biases, immediately and 8 weeks after game play. We are also able to offer insight into how future designers and researchers might spend resources in developing games for these sorts of learning outcomes and learning contexts.

Cognitive Biases and Serious Games Research on cognitive biases demonstrates that biases can function as useful heuristics when people must sort through Ó 2016 Hogrefe Publishing

A. Shaw et al., Serious Efforts at Bias Reduction

large amounts of information, when certain information primes them to focus on specific types of information, or when they are asked to make quick decisions or analyses (Kahneman, 2011). This has been demonstrated for FAE (Jones & Harris, 1967; Tetlock, 1985) and CB (Fischer & Greitemeyer, 2008). People are generally unaware of their reliance on these biases and thus have a blind spot (BBS; Pronin, Lin, & Ross, 2002). Although use of these heuristics can be adaptive, getting people to overcome the shortcomings such biases can engender in decision-making situations has proven a difficult task. Moreover, training ought not to simply replace one bias with another (e.g., overcoming confirmation bias through over-reliance on disconfirmation bias). Research on bias mitigation techniques generally argues that people must be taught to think critically and to question their initial evaluations of scenarios in order to account more thoroughly for alternative interpretations and unaccounted for variables (Larrick, 2004; Stewart, Latu, Kawakami, & Meyers, 2010). These skills are particularly important in recognizing and adjusting behavior to avoid the three aforementioned biases. FAE refers to the tendency for people to overuse personality-based explanations for the behaviors they observe in others (Jones & Harris, 1967). Correspondingly, situational or environmental factors are overlooked. CB is the tendency to seek out or favor information that supports pre-existing beliefs (Kahneman, 2011; Nickerson, 1998). This is a bias in information processing that has been identified as leading to serious errors in analyses of scientific, political, and social information (Cheikes, Brown, Lehner, & Adelman, 2004). BBS is a phenomenon identified by Pronin and colleagues (2002) as people’s tendency to believe they are less biased than are others. It is pervasive, as Pronin et al. demonstrated that this bias persists even after bias mitigation training. Friedrich (1996) also shows that awareness is not enough to mitigate the bias, similar to what Pronin et al. (2002) demonstrate. Recent research found that not only did self-awareness not mitigate this bias, those with higher intelligence were more susceptible to it than were others (West, Madison, & Stanovich, 2012).

Training With Serious Games Training techniques can be incorporated into digital game scenarios (Rahford, 2009). They have a well-documented potential for offering new and innovative learning environments. As Malone (1981, 1980), Rieber (1996), Gee (2003), Prensky (2001), and Aldrich (2005) identify, good games mirror effective learning models. Games promote flow, as described in Csikszentmihalyi (1990), and can create

Ó 2016 Hogrefe Publishing

engaging, immersive, and intrinsically motivating learning environments. Effective games are intrinsically motivating to players (McGonigal, 2011; Salen & Zimmerman, 2003), and intrinsic motivation generates greater learning outcomes (Deci & Ryan, 1985). Yet, training of cognitive biases has not been brought into the medium of digital games. Although there have been efforts to reduce bias in specific decision-making contexts, these are typically framed at the point of decision making (Larrick, 2004), or are based on tools that can directly influence the processing of information (e.g., decision support systems) within the actual task environment (Arnott, 2006). Although reducing individuals’ overall tendencies toward bias through training was proposed by Fischhoff (1982), implementing such training has had limited success (Sanna, Shorts, & Small, 2002). In addition, although some bias elicitation and training techniques have been found to be reliable in non-game settings (Dale, Kehoe, & Spivey, 2007), the use of these techniques in a game environment has rarely been evaluated. The emergence of “serious games” as a training tool offers a unique opportunity to use transformative learning for effective bias mitigation. Digital educational games can potentially provide experiential learning opportunities that allow players to recognize and mitigate their reliance on biases. The interactive and responsive nature of games can provide players feedback on how their behaviors demonstrate or avoid biases. Some scholars, however, suggest that educational videos can be powerful learning tools as well (Karppinen, 2005). This study, thus, tested the following hypotheses: Hypothesis 1 (H1): An educational digital game will result in significant bias reduction of three cognitive biases: fundamental attribution error, confirmation bias, and bias blind spot. Hypothesis 2 (H2): The delivery of cognitive bias training via a digital game will result in better bias reduction than will content delivered via an educational video. In addition, we sought to test whether the learning effects lasted. Therefore, we hypothesized: Hypothesis 3 (H3): The effects of training will persist for game conditions 8 weeks after the training has been completed. In addition to testing the overall impact of the digital game on learning, we were also interested in the impact of specific game features on learning outcomes. It has been suggested that if players identify with their avatars, greater

Journal of Media Psychology (2018), 30(1), 16–28

learning effects can occur (Gee, 2003; Shaw & Linebarger, 2008). One possible reason for such effects is that players form attachments to game characters because they are actively using those avatars to navigate the game space (Klimmt, Hefner, & Vorderer, 2009; Murphy, 2004; Rehak, 2003). Klimmt and Vorderer (2003) suggest that the high level of involvement in games produces a feeling of presence, which serves as a mechanism to transport the player into the narrative world (Green, Brock, & Kaufman, 2004). It is possible that the work of customizing a player character further involves the player in the game, which in turn enhances learning. Indeed, studies of virtual worlds suggest that when people customize or create characters or virtual identities as a form of self-representation, they identify more strongly with their avatars (Boellstorff, 2008; Mortensen, 2007; Taylor, 2006; Turkle, 1995). Some research suggests that customization even when the player character does not represent the player visually still produces greater levels of player–avatar connection (Shaw, 2014). Bailey, Wise, and Bolls (2009) found in a study of children that customizing characters led to greater feelings of presence and psychophysiological indicators of emotion. Other researchers have looked at the relationship between avatar customization and player enjoyment of games (Trepte & Reinecke, 2010) and player aggression after playing a game (Organ, 2009). Some research suggests that players will take on behaviors and attitudes based on what their avatars look like, suggesting avatar presentation can have significant effects on players (Yee, Bailenson, & Ducheneaut, 2009). Given that prior research suggests that character customization may have an effect on learning outcomes, we hypothesized the following: Hypothesis 4 (H4): A game with a customizable character will produce better bias reduction between preand posttests than will a game using an assigned character. In addition, we used qualitative and survey responses to ascertain if the different media experiences lead to differences in participant evaluations of the game experience. This helped us answer the following research question: Research Question 1 (RQ1): Does the use of a customizable character appeal to participants more than does the use of an assigned character? By studying significant effects of conditions on participants as well as participants’ qualitative assessments of conditions, we argue that we are better able to parse how certain game variables come to be important in educational games.

Journal of Media Psychology (2018), 30(1), 16–28

A. Shaw et al., Serious Efforts at Bias Reduction

Method Stimuli To examine character customization and the efficacy of a training game compared with a training video, we built an educational game working with a professional educational game development company. The core game design approach was based on Mezirow’s (1991) transformative learning theory (TLT). We also harnessed a classic learning taxonomy by Bloom, Engelhart, Furst, Hill, and Krathwohl (1956; see also Anderson & Sosniak, 1994) to help hone in on the learning objectives for each element of the game, particularly within the realms of understanding identification of the biases and demonstration of bias mitigation techniques. Our aim was to help players understand how to mitigate the bias as measured in postsession surveys by allowing them to practice bias mitigation in the game. The core of our game was an austere training facility that was reminiscent of a bland institutional building, somewhat futuristic with plaster walls. Modular rooms with sparse furnishings and objects were meant to evoke a training environment. Each room was a closed, modestly sized space in which players could move their character. They clicked objects to solve puzzles that related to the cognitive biases. Each room had a marked exit that was inaccessible until the task or puzzle was completed (see Figure 1). A computerized voice (with captions) provided instructions and information to players as they went through the rooms. Transition spaces between the rooms provided the core teaching about the cognitive bias as well as mitigation techniques. Infographics were also used to help underscore the learning content (see Figure 2). The game was designed to be played for approximately 30 min. In the assigned character condition, players were given a predesigned androgynous character. This avatar wore a space-suit that obscured all but the character’s body type, to ensure that we minimized the extent to which disidentification on the basis of age, race, and other identifiers might occur (Muñoz, 1999). We did not want a sense of disidentification to affect player performance. In the customized character condition, players could choose to play as either a male or female avatar. They then had the option to customize the body type (slim, medium, and large), suit pattern, and helmet of their avatar (see Figure 3 for the male avatar options). They could also name their avatar. These customization options were developed to focus specifically on the effect of customization, rather than the effect of self-representation, on outcome measures. Although customization options were basic, they were selected to complement the limited graphic detail and simple art style of the game.

Ó 2016 Hogrefe Publishing

A. Shaw et al., Serious Efforts at Bias Reduction

Figure 1. Puzzle room about Bias Blind Spot.

Figure 2. Infographic of Fundamental Attribution Error.

The training video was a professionally produced video that covered cognitive biases by looking at a series of scenarios involving a college student during his interactions with friends who engaged in cognitive biases. The vignettes exemplified a bias and were explained by an instructor character afterwards. The instructor character also discussed mitigation strategies. The video was developed using the same definitions and interpretations of the three biases used in the game and was created by the study’s government funders specifically for this project.

Study Procedures Participants were recruited from college classes and psychology pools at three universities from different regions Ó 2016 Hogrefe Publishing

of the United States. After signing up, participants were randomly assigned to one of three conditions: a digital game with character customization, a digital game with an assigned character, or a training video. They came to a computer lab where they gave their consent to participate in the experiment, following standard protections protocols for human subjects. They then answered questionnaires before and after the experimental stimuli were given. Dependent variables – three scales used to assess three different cognitive biases – were measured on pretest and posttest questionnaires along with several demographic and personality variables. After the posttest survey, 27 selectively sampled individuals were interviewed. Eight weeks after the initial posttest, participants were sent email invitations inviting them to answer a follow-up survey that included the dependent measures. Journal of Media Psychology (2018), 30(1), 16–28

A. Shaw et al., Serious Efforts at Bias Reduction

Figure 3. Customized character screen (male options).

Study Participants From the original sample (N = 234), four people were excluded because of noise interference from a nearby construction site or because of participation in the pilot study. Of the remaining 230 participants, 82 (35.6%) were randomly assigned to play the customized character version of the game, 80 (34.8%) to play the assigned character version, and 68 (29.6%) to watch the video). All participants were sent a follow-up questionnaire via e-mail 8 weeks later, and 104 students completed it (45% completion rate). Initial Study Of the initial study participants, 59% reported being female and 41% reported being male. Study participants identified their race/ethnicity by checking all categories that applied to them. Most participants identified as White (64%), 16% as Hispanic/Latino, 10% as East Asian, 8% as Black/African, and 13% as some other racial or ethnic identification. The majority of participants were fourth-year (53%) or third-year students (28%), the rest were first-year (9%) or second-year students (10%), and their average age was 22 years, ranging from 18 to 62. Participants reported the following majors: 35.7% communication, 7.4% business, 5.8% psychology, and 51.2% some other major. When asked to describe their level of experience using computer technologies, 4% of participants described themselves as novices, 43% as intermediate, 28% as advanced, 19% as expert, and 5% as being a master. When asked whether they agreed with the statement: “I like playing video games,” 59% agreed or strongly agreed, 22% were neutral, and 19% disagreed or strongly disagreed. In addition, 22% said they considered themselves to be gamers. Journal of Media Psychology (2018), 30(1), 16–28

Eight-Week Retention Panel Analysis of the 104 participants who took the 8-week follow-up indicated roughly similar demographic profiles to our full sample. Of the retention participants, 66% reported being female and 34% reported being male. Racial breakdowns were: 67% White/Caucasian/European, 15% Hispanic/Latino, 12% East Asian, 6% Black/African, and 8% reported some other racial or ethnic identification. Retention participants had an average age of 23 years (ranging from 19 to 62). About 30% had played the customized version of the game, 37% had played the assigned version, and 34% had watched the video. Qualitative Interviews We selected 27 participants for postsession qualitative interviews. The selection criteria included completion time: who finished their sessions fastest and slowest in each condition. These participants were selected as an attempt to include those who were the most and least adept at playing the game. In order to achieve a gender and racial balance, we also selectively interviewed additional male and nonWhite interviewees in the two game conditions (regardless of their time to completion). Four participants were in the video condition, nine in the assigned character, and 14 in the customized character conditions. We have more interviewees in game conditions for two reasons. First, as we were purposively sampling fastest and slowest finishers, we had more types of players to sample than we did video condition interviewees (who all had to watch the film at the same rate). We also had to additionally oversample in the game conditions to get a better racial and gender balance within each condition. Interviews ranged from 11 to 23 min in length. Approximately 14 of our interview participants were female, 13 were male. Fourteen were White/Caucasian/European, five Hispanic/Latino, Ó 2016 Hogrefe Publishing

A. Shaw et al., Serious Efforts at Bias Reduction

three Black/African, and five of some other racial/ethnic identity. Twelve participants were between the ages of 19 and 21 years, 10 between the ages of 22 and 24 years, and four were between 26 and 38.

Dependent Variables: Measures of Three Cognitive Biases Three scales were developed by the researchers to measure the three cognitive biases that were the focus of study. The original scales for FAE, CB, and BBS were centered on a rating of 4, and average ratings across items were transformed to a scale of 100 (maximum of the opposite bias) to +100 (maximum bias) by subtracting the scale midpoint from the score, dividing by the midpoint –1, and multiplying by 100. This was done in order to center the scores at 0 and for clarity when discussing changes in scores. Thus for positive numbers, lower numbers indicate less bias, negative scores indicate a counter bias (e.g., disconfirmation bias), and 0 indicates no bias. The same measures were used on pre-, post-, and 8-week follow-up tests. In a pilot study of the bias measures, all three measures were examined for practice effects by administering the measures twice (N = 52). This was done with a distractor task in between (solving math problems). No strong practice effects were found (for all paired t tests, p > .2). We also tested discriminant validity of the measures and found only low Pearson correlations among the measures (r < .15). Fundamental Attribution Error (FAE) Brief scenarios were presented about other individuals. Participants were asked to rate the protagonist in scenarios in which some mistake was happening on a 7-point agreeto-disagree scale. If the participants agreed with the description, they were attributing the mistakes to the protagonist’s character rather than to something else (e.g., situation). For example, participants were asked to read the following scenario: “In a class this semester, one of the students, John, almost always arrives late. The best explanation for this behavior is that John is a poor planner. To what extent do you agree or disagree with this assessment?” Participants rated their agreement from 1 (= strongly disagree) to 7 (= strongly agree) of the following statement: “John is a poor planner.” The scale contained 10 such questions with a pretest reliability of α = .70 and a posttest reliability of α = .89. Confirmation Bias (CB) The CB scale was based on the paradigm developed by Cook and Smallman (2008). The questions presented scenarios and asked participants to select a hypothesis. They were then asked to rate pieces of evidence as unimportant Ó 2016 Hogrefe Publishing

to extremely important on a 7-point scale to evaluate their hypothesis. Six pieces of evidence for each scenario were offered: Three pieces confirmed the person’s decision; three disconfirmed the person’s decision. The items were used to measure if the participants gave more weight to confirming evidence or to disconfirming evidence. For example, “Your friends want to adopt a pet. They want an unusual pet, instead of a dog or cat, so they pick a Wolfcat. You don’t know much about them, but you’ve heard Wolfcats are more effort than they are worth. Decide if you think your friends should adopt a Wolfcat.” Participants then chose between two options: “They should adopt a Wolfcat” or “They should not adopt a Wolfcat.” Finally, participants were asked to “rate the importance of considering the following articles in evaluating your decision.” The articles for rating were presented as headlines only, such as “Wolfcats – They can save your life!” and “The notoriously boring Wolfcat and other disappointing pets.” CB was calculated by averaging ratings to items consistent with the hypothesis minus average ratings to items inconsistent with the hypothesis. Five scenarios were used for this scale and had pretest reliability of α = .61 and posttest reliability of α = .79. Bias Blind Spot (BBS) This scale was based on measures used by Pronin et al. (2002). Participants rated the likelihood of their own responses and their estimates of “an average student at their university” to a positive attribute, which reflected the bias Illusion of Superiority on a 7-point scale. Illusion of Superiority refers to an individual seeing themselves as above others on given positive traits. For example, participants were first asked, “How would you rate your ability to be forgiving?” and then, “How would you rate the ability of the average student from your university to be forgiving?” Differences between the self-rating minus the other-rating were used to calculate BBS. The BBS scale was measured with seven questions and had a pretest reliability of α = .88 and posttest reliability of α = .85.

Results Analysis of our pretest bias scales confirmed that our measures detected bias in the study participants. Table 1 shows Ns, means, and standard deviations (SD) for the bias measures. As Table 1 indicates, all of our bias measures showed bias in the pretest, with scores about 0.5 to 1.5 standard deviations above 0 (the designated unbiased level). We examined whether or not the games taught players to mitigate biases (H1) and outperformed the training video (H2) immediately after game play and 8 weeks later (H3).

Journal of Media Psychology (2018), 30(1), 16–28

A. Shaw et al., Serious Efforts at Bias Reduction

Table 1. Bias mitigation descriptive statistics FAE pre

FAE post

FAE 8-week

CB pre

CB post

CB 8-week

BBS pre

BBS post

15.6

7.8

3.2

20.3

13.6

27.1

39.2

41.8

17.8

20.2

15.1

BBS 8-week

16.0

10.2

7.0

12.0

17.2

21.4

19.1

Customized character condition 31

Assigned character condition N

21.3

80 0.4

38 2.1

21.2

13.8

10.3

7.0

6.7

7.1

21.9

36.4

40.7

19.8

19.6

18.6

21.7

21.2

14.0

21.2

4.2

5.0

18.1

20.9

18.8

9.2

12.2

8.2

30.2

34.3

28.9

19.5

23.4

24.4

19.5

21.9

15.8

Video condition

Notes. FAE = Fundamental Attribution Error. CB = Confirmation Bias. BBS = Bias Blind Spot.

To do so, performance on each bias metric was examined using a pair of analyses. First, we used an ANOVA featuring the repeated measures bias score from pretest and posttest by the three conditions (customized, assigned, video) and used planned comparisons to examine posttest-only performance by condition. We also used two planned linear contrasts to explore the effects of game play versus the video (customized and assigned vs. video) and differences between the game conditions (customized vs. assigned). Table 2 shows the test statistics for each bias at immediate posttest. Tests that reach the .05 significance threshold have their significance in bold. FAE showed evidence of equal learning in all conditions, with a main effect of pretest–posttest and significant reductions in bias for all three conditions, supporting H1. There was no significant difference between the games and the video, showing no support for H2, and the two game conditions did not show significant differences on the posttest, suggesting no support for H4. Analysis of CB suggested that the game conditions achieved a significant reduction in this bias and outperformed the video, seen in a main effect of pretest–posttest and a significant difference between the combined game conditions and the video condition. The video condition did not demonstrate a reduction in this bias at posttest. H1 and H2 were supported for CB. Planned linear contrasts did not find a difference between game conditions on the posttest, showing no support for H4. For BBS there was no significant evidence of learning in any condition, demonstrating no support for H1 or H2, and no difference between the game conditions, showing no support for H4. To assess how well this training was retained by participants in the long term, we analyzed the data from the 8-week retention survey. Table 3 shows test statistics for the 8-week retention test for each bias. Tests that reach the .05 significance threshold have their significance in bold. Journal of Media Psychology (2018), 30(1), 16–28

Results should be interpreted with caution owing to the fact that the completion rate was 45% for this follow-up test. Analysis of bias reduction retention for FAE showed a main effect of pretest–posttest, and planned comparisons showed that the assigned game and the video resulted in a long-term reduction of this bias, partially supporting H3 for this bias. The customized game did not show long-term reduction of FAE, but the two game conditions did not differ significantly on the retention test. No difference between the games and the video in the retention test was found using the planned linear contrasts. Participants retained some training in CB in the game conditions, but not in the video condition. Planned comparisons showed that the assigned game reduced CB, but the customized game did not, partially supporting H4. The video condition showed no long-term reduction in bias, and the two game conditions did not differ significantly on the retention test. Planned linear contrasts did not show a difference between the games and video. For BBS there was no significant evidence of long-term change in bias in any condition, and no significant differences between the games or between the games and the video condition. H4 was not supported for this bias. Repeated measures ANOVAs for the immediate posttest and for the 8-week posttest showed none of the following were moderators: gender, handedness, liking video games, and experience using computer technologies.

Qualitative Analysis In order to explore players’ experiences learning from the games and the video in further depth, we also analyzed a series of qualitative interviews conducted after their lab sessions. We aimed to identify whether character customization made the game more or less appealing, Ó 2016 Hogrefe Publishing

A. Shaw et al., Serious Efforts at Bias Reduction

Table 2. Test statistics for repeated measures ANOVAs by bias at immediate posttest FAE

BBS

Test statistics

F(1,227) = 97.61

< .0005

F(1,227) = 9.40

< .005

F(1,227) = .02

.11

F(2,227) = 2.07

.12

F(2,227) = .55

.58

F(2,227) = .80

.45

F(2,227) = .80

.45

F(2,227) = 5.24

< .01

F(2,227) = 2.23

.11

Main effects Main effect pretest–posttest Main effect condition Interaction pretest–posttest* condition Planned comparisons (pretest–posttest) Customized game

t(81) = 6.65

< .0005

t(81) = 3.01

< .005

t(81) = 1.42,

.16

Assigned game

t(79) = 5.32

< .0005

t(79) = 2.95

< .005

t(79) = 0.13,

.90

Video

t(67) = 5.71

< .0005

t(67) =

1.16

.25

t(67) =

1.54,

.13

Planned linear contrasts (posttest only) Customized vs. assigned games

t(227) =

1.28

.13

t(227) =

0.05

.96

Games (combined) vs. video

t(227) =

1.57

.12

t(227) =

2.38

< .05

t(227) = 0.09 t(227) =

1.71,

.93 .09

Notes. FAE = Fundamental Attribution Error. CB = Confirmation Bias. BBS = Bias Blind Spot. Tests that reach the .05 significance threshold are in bold.

Table 3. Test statistics for repeated measures ANOVAs by bias at 8-week retention test FAE

BBS

Test statistics

F(1,101) = 26.85

< .0005

F(1,101) = 8.49

< .005

Test statistics

Main effects Main effect pretest–retention Main effect condition Interaction pre-retention* condition

F(2,101) = .012

.45

F(2,101) = 1.65

F(2,101) = .988

.52

F(2,101) = 3.27

F(1,101) = .014

.91

.65

F(2,101) = 3.44

< .05

F(2,101) = 1.60

.21

Planned comparisons (pre-retention) Customized game

t(30) = 1.68

.10

t(30) = 0.38

.71

Assigned game

t(37) = 2.80

< .01

t(37) = 3.44

< .005

Video

t(34) = 4.82

< .0005

t(34) = 1.02

.32

Customized vs. assigned games

t(101) = 0.12

.91

t(101) = 1.19

Games (combined) vs. video

t(101) = 0.98

.33

t(30) = 0.86 t(37) =

.40

1.29

.21

t(34) = 0.74

.46

.96

t(101) = 1.23

.93

.21

t(101) = 0.41

.73

Planned linear contrasts (retention only) t(101) =

1.19

Notes. FAE = Fundamental Attribution Error. CB = Confirmation Bias. BBS = Bias Blind Spot. Tests that reach the .05 significance threshold are in bold.

how players customized their characters, and whether customization made them feel more or less connected to their character. The interviews took place in private, were digitally recorded, and were anonymized for analysis. Interviews focused on participant reactions to the game/video (what they liked/did not like), if and how participants learned from them (what was effective/ineffective), if the game/video was engaging and why, and if the interviewees felt connected to the character and lessons. We also asked about formal characteristics of both the game and video to see if there were specific areas affecting interviewee reception of the conditions. Overall, most participants liked the game and enjoyed learning from a game more than they would have from other types of materials. A typical reason for this is illustrated by one 20-year-old’s explanation that, “I think playing the game you are a lot more active, and if you’re listening to a lecture, reading a book, you can kinda get Ó 2016 Hogrefe Publishing

away with not having to learn it. Whereas the game – like it makes you learn it, and makes you understand it. I think that was a lot more beneficial.” Players generally found the game engaging because of its interactive format. One interviewee said, for example, “You had to figure out – and I wasn’t just gonna give up and not figure out what I had to do.” Similarly, video condition interviewees generally liked the video and found it humorous and relatable. They felt they learned about cognitive biases from watching it, and generally liked the characters portrayed. Some, however, felt the video was a bit long and not consistently engaging. Responses to the games’ character customization (or lack thereof) were somewhat related to liking the game overall. Interviewees who did not like the game in the customization condition said taking time to customize to their character was an annoyance. On the other hand, those that liked the game said customization was a particularly attractive Journal of Media Psychology (2018), 30(1), 16–28

option. Those that liked the game but could not customize their avatar said they would have preferred to be able to do so, whereas those who did not like the game did not feel customization was needed. There were four identifiable approaches to character customization. Interviewees who did not feel customization mattered to the game clicked options randomly. Some interviewees, regardless of whether they liked the game or customization options, chose options that were the most visually appealing to them. Some interviewees tried to represent themselves, including one player who did not like the game or customization and said, “I don’t know if I really thought about it consciously, but I think, looking back, I just picked one that looked like me sort of, even though it was not human.” Only one interviewee took a more creative route and tried to create a superhero character and referenced comic book art norms for design: [in reference to suit choice] “The solid colors usually represent a solid personality, someone that they can all depend upon. Where it was like the stripeys usually indicates Frost Man or a sidekick. So I was like, ‘I’m going for the big guy. We’re gonna be Superman.’” Despite the rather minimal customization options available in the game, interviewees’ approaches to character customization ran the gamut of video game players’ approaches to representation more generally, as seen in Shaw’s (2014) research.

Discussion This study tested the effects of two versions of an educational digital game and a video designed to reduce FAE, CB, and BBS. Analyses showed that the game could mitigate two of the biases, FAE and CB (H1), and did so better than a training video only for CB (H2). The training video actually increased CB and BBS slightly (but not significantly) in both immediate and 8-week posttests. None of the stimuli materials resulted in reduction of BBS. Overall, the game we developed reduced participants’ cognitive biases by an average across the two games of 42% for FAE and 32% for CB. Percent improvement was calculated as (|Posttest| Pretest)/Pretest 100. This mitigation was retained over 8 weeks for an average of 20% for FAE and 30% for CB when compared with the presession measures, partially supporting H3. Contrary to our hypothesis (H4), we found no statistically significant differences between game conditions, and that the assigned character version of the game was the only condition that produced significant training in mitigating FAE and CB in both the long and short terms. We hypothesized that character customization would have a significant positive effect on cognitive bias reduction. The evidence suggests, however, that such customization Journal of Media Psychology (2018), 30(1), 16–28

A. Shaw et al., Serious Efforts at Bias Reduction

makes no difference in a 30-min play session, although our retention data suggested a slight improvement in learning about CB for players with an assigned character, although this is worth replicating to see if it continues to hold, given our 45% response rate on the retention. Qualitative interviews suggested that participants generally liked all of the media conditions. With regard to the games, they suggested that regardless of condition those who liked the game enjoyed or would have liked to customize their character, but those that did not enjoy the game felt customization was unnecessary or took up too much time (RQ1). Overall, these results are promising and suggest that a game can reduce fundamental attribution error and confirmation bias. Of particular note is the success of the game in reducing confirmation bias. Prior research has demonstrated that although some strategies, such as presenting information in a disfluent (inconsistent and thus less easily processed) format, may reduce this bias (Hernandez & Preston, 2013), generalized training such as that used in the game is rarely effective (George, Duffy, & Ahuja, 2000). Our results suggest that a game format may offer a more successful approach to training against this bias than do linear, narratively based formats, perhaps due to the interactive characteristics of games. Because reductions differed significantly across the three biases, we conclude that learning about one bias does not equate automatically to reduction of another bias. This was particularly clear when examining FAE versus BBS; the latter, as Pronin and colleagues (2002) have argued, can be retained at high levels even when other biases are less evident. Indeed, our results are similar to other research on BBS that finds this bias is resistant to change, even among those who are well trained to recognize biases. This might be because the confidence individuals gain through recognizing and mitigating other biases can translate into overconfidence in the idea that biases in general no longer affect them. This may be heightened by the knowledge of other biases itself. For example, West and coworkers (2012) found that those with higher cognitive ability and more awareness of biases were actually more susceptible to bias blind spot. Similarly, Pronin and colleagues (2002) found that those who were better than average at avoiding biases demonstrated more bias blind spot. Although research suggests that character customization might improve learning, we did not find that in this experiment. Although prior work has suggested that identifying with one’s avatar can increase learning (Gee, 2003; Shaw & Linebarger, 2008), perhaps through an increased sense of involvement in the game (Green, Brock, & Kaufman, 2004), it is possible that in our single lab sitting, players do not develop a strong motivation to invest in their characters, whether they created it or it was given to them. Prior research has shown that having an avatar choice can lead to Ó 2016 Hogrefe Publishing

A. Shaw et al., Serious Efforts at Bias Reduction

greater objectively measured arousal (Lim & Reeves, 2009); however, in our qualitative results participants’ self-reported reactions to customization were relative to their overall assessment of the game. Our controlled experiment, moreover, provides a concrete data point to suggest that all things being equal, customization does not improve learning outcomes immediately after play, although an assigned avatar may improve long-term learning for some types of content. We suggest that the lack of distraction produced from customization at the start of the game may have promoted greater focus to the learning content that was then stored in long-term memory, leading to greater retention effects. Moreover, if involvement is the mediating variable between customization and learning as prior work suggests, further research is needed to tease out that relationship. We identify some limitations with this experiment. First, this project used nonrandom recruitment exclusively among students on university campuses; thus, participants are not representative of the overall US population in terms of education and age, nor of college populations overall. As a younger population, participants may have been more likely to feel comfortable with games and have slightly more experience with them. As college students, these participants may also be more accustomed to learning environments and open to learning activities of different types; as such, they may have learned more than their non-college counterparts would. Participants were also self-selected and largely from the social sciences. Second, our conditions were limited as we had to sacrifice some ecological validity for internal validity. The customization condition we developed was a bit more circumscribed than is typically available in longer or more involved video games with customizable characters. In developing the condition, we focused on the effect of customization in a manner that minimized the possibility of self-representation (e.g., not allowing the choice of facial features often tied to demographic characteristics), as that is a significant hole in the literature on the subject. We believed introducing other types of identities would offer a number of confounds that would require a significantly larger sample to address properly, which was not feasible for this study. For our assigned character condition we attempted to provide gender-neutral avatar options; it is possible that these figures were interpreted as male thus not making them as representationally neutral as intended. Future research should explore more nuanced and complex customization options that still minimize confounding self-representation with customization elements and experiment with additional ways of creating neutrally assigned avatars. Third, the video used for comparison in this study did not present information about biases identically to the language, images, or depth of information used in the games. An important limitation of comparing learning from the Ó 2016 Hogrefe Publishing

game to that from the video is this difference in style; it is possible that the game simply used superior explanations of the biases. Additional research is needed to identify the extent to which the games’ better training resulted from content, format, or a combination of both. Finally, there are limitations associated with the measures of bias used in this study. We had to create measures of FAE and of BBS, because none existed, and we slightly modified Cook and Smallman’s (2008) measures of CB. Test–retest reliability and discriminant validity were assessed, and no problems were detected. It is possible, however, the measures may not fully capture the target biases or be sensitive enough to changes after training. Additional research on survey-based measures of FAE, CB, and BBS are needed. For example, the absence of training effects on BBS might represent a bias that is harder to mitigate, or less effective training within the game, but may also simply indicate that the BBS measure has lower sensitivity to changes in underlying bias than the other measures.

Conclusion In answer to the call for research that focuses on debiasing rather than eliciting bias (Larrick, 2004), this study found significantly greater reductions in the propensity for biased thinking after subjects played our game compared with when they watched a training video, at least on FAE and CB. We found no significant differences at the immediate posttest or at the 8-week retention test between a game with a customized or an assigned avatar. There was positive evidence of retention for the assigned avatar condition and not for the customized avatar, but the difference between conditions was not significant. Notably, however, avatar customization was a decided preference for some interviewees, particularly those who enjoyed the game overall. This suggests that despite a lack of statistically significant differences, it is possible for researchers and designers to leverage this particular game variable to appeal to players. It is important to note that some players found the added time required for customization a source of frustration, thus such additions should be made in consultation with the intended audience. Moreover, in terms of deciding how best to spend resources, these results indicate that customization alone will not result in better learning effects.

Acknowledgments This work was supported by the Intelligence Advanced Research Projects Activity (IARPA) via the Air Force Research Laboratory contract number FA8650-11-C-7176. The views and conclusions contained herein are those of the authors Journal of Media Psychology (2018), 30(1), 16–28

and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of IARPA, AFRL, or the US Government.

References Aldrich, C. (2005). Learning by doing: A comprehensive guide to simulations computer games, and pedagogy in e-learning and other educational experiences. San Francisco, CA: Pfeiffer. Anderson, L. W., & Sosniak, L. A. (1994). Blooms taxonomy: A fortyyear retrospective. Ninety-third yearbook of the Chicago National Society for the Study of Education. Part II. Chicago, IL: University of Chicago Press. Arnott, D. (2006). Cognitive biases and decision support systems development: A design science approach. Information Systems Journal, 16, 55–78. doi: 10.1111/j.1365-2575.2006.00208 Bailey, R., Wise, K., & Bolls, P. (2009). How avatar customizability affects children’s arousal and subjective presence during junk food-sponsored online video games. CyberPsychology & Behavior, 12, 277–283. doi: 10.1089/cpb.2008.0292 Boellstorff, T. (2008). Coming of age in Second Life: An anthropologist explores the virtual human. Princeton, NJ: Princeton University Press. Bloom, B. S., Engelhart, M. D., Furst, E. J., Hill, W. H., & Krathwohl, D. R. (1956). Taxonomy of educational objectives: The classification of educational goals. London, UK: Longmans. Breuer, J., & Bente, G. (2010). Why so serious? On the relation of serious games and learning. Eludamos. Journal for Computer Game Culture, 4, 7–24. Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York, NY: Harper Collins. Cheikes, B. A, Brown, M. J, Lehner, P. L, & Adelman, L. (2004). Confirmation bias in complex analyses (MITRE Technical Report MTR 04B0000017). Bedford, MA: MITRE. Retrieved from http:// www.mitre.org/publications/technical-papers/confirmationbias-in-complex-analyses Cook, M. B., & Smallman, H. S. (2008). Human factors of the confirmation bias on intelligence analysis: Decision support from graphical evidence landscapes. Human Factors, 50, 745–754. Dale, R., Kehoe, C. E., & Spivey, M. J. (2007). Graded motor responses in the time course of categorizing atypical exemplars. Memory & Cognition, 35, 15–28. Deci, E., & Ryan, R. (1985). Intrinsic motivation and self-determination in human behavior. New York, NY: Springer. Fischer, P., & Greitemeyer, T. (2008). A new look at selectiveexposure effects: An integrative model. Current Directions in Psychological Science, 19, 384–389. doi: 10.1177/ 0963721410391246 Fischhoff, B. (1982). Debiasing. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp. 422–444). Cambridge, UK: Cambridge University Press. Friedrich, J. (1996). On seeing oneself as less self-serving than others: The ultimate self-serving bias? Teaching of Psychology, 23, 107–109. doi: 10.1207/s15328023top2302_9 Gee, J. P. (2003). What video games have to teach us about learning and literacy? New York, NY: Palgrave/St. Martin’s. George, J. F., Duffy, K., & Ahuja, M. (2000). Countering the anchoring and adjustment bias with decision support systems. Decision Support Systems, 29, 195–206. doi: 10.1016/S01679236(00)00074-9 Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment: The role of transportation into narrative worlds. Communication Theory, 14, 311–327. doi: 10.1111/ j.1468-2885.2004.tb00317

Journal of Media Psychology (2018), 30(1), 16–28

A. Shaw et al., Serious Efforts at Bias Reduction

Greenfield, P. M. (1984). Mind and media: The effects of television video games, and computers. Cambridge, MA: Harvard University Press. Hernandez, I., & Preston, J. L. (2013). Disfluency disrupts the confirmation bias. Journal of Experimental Social Psychology, 49, 178–182. doi: 10.1016/j.jesp.2012.08.010 Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3, 1–24. doi: 10.1016/0022-1031(67)90034-0 Kahneman, D. (2011). Thinking fast and slow. New York, NY: Macmillan Publishers. Karppinen, P. (2005). Meaningful learning with digital and online videos: Theoretical perspectives. AACE Journal, 13, 233–250. Kirriemuir, J., & McFarlane, A. (2004). Literature review in games and learning. Bristol, UK: NESTA Futurelab. Klimmt, C., Hefner, D., & Vorderer, P. (2009). The video game experience as ‘true’ identification: A theory of enjoyable alterations of players’ self-perception. Communication Theory, 19, 351–373. doi: 10.1111/j.1468-2885.2009.01347.x Klimmt, C., & Vorderer, P. (2003). Media psychology “is not yet there”: Introducing theories on media entertainment to the presence debate. Presence: Teleoperators and Virtual Environments, 12, 246–259. doi: 10.1162/105474603322391596 Klopfer, E., Osterweil, S., & Salen, K. (2009). Moving learning games forward: Obstacles, opportunities & openness. Cambridge, MA: MIT The Education Arcade. Retrieved from http://web.mit.edu/21w.789/www/papers/ MovingLearningGamesForward_EdArcade.pdf Larrick, R. P. (2004). Debiasing. In D. J. Koehler & N. Harvey (Eds.), Blackwell handbook of judgment and decision-making (pp. 316– 338). Malden, MA: Blackwell Publishing. Lim, S., & Reeves, B. (2009). Being in the game: Avatar choice and point of view on psychophysiological responses during play. Media Psychology, 12, 348–370. doi: 10.1080/15213260903287242 Malone, T. (1980). What makes things fun to learn?: A study of intrinsically motivating computer games. Proceedings of the 3rd ACM SIGSMALL symposium and the first SIGPC symposium on small systems (pp. 162–169). New York, NY: ACM. Retrieved from http://dl.acm.org/citation.cfm?id=802839 Malone, T. W. (1981). Toward a theory of intrinsically motivating instruction. Cognitive Science, 5, 333–369. doi: 10.1207/ s15516709cog0504_2 McGonigal, J. (2011). Reality is broken: Why games make us better and how they can change the world. New York, NY: Palgrave Press. Mezirow, J. (1991). Transformative dimensions of adult learning. San Francisco, CA: Jossey-Bass. Muñoz, J. E. (1999). Disidentifications: Queers of color and the performance of politics. Minneapolis, MN: University of Minnesota Press. Mortensen, T. (2007). Me, the other. In P. Harrigan & N. WardripFruin (Eds.), Second person, role-playing and story in games and playable media (pp. 297–307). Cambridge, MA: MIT Press. Murphy, S. C. (2004). ‘Live in your world, play in ours’: The spaces of video game identity. Journal of Visual Culture, 3, 223–238. doi: 10.1177/1470412904044801 Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2, 175–220. Organ, A. (2009). Avatar identification: How similarity in appearance influences aggressive responses (Unpublished master’s thesis). Michigan State University, East Lansing, MI. Prensky, M. (2001). Digital game-based learning. New York, NY: McGraw-Hill. Pronin, E., Lin, D. Y., & Ross, L. (2002). The bias blind spot: Perceptions of bias in self versus others. Personality and Social Psychology Bulletin, 28, 369–381. doi: 10.1177/ 0146167202286008

Ó 2016 Hogrefe Publishing

A. Shaw et al., Serious Efforts at Bias Reduction

Przybylski, A. K., Deci, E. L., & Ryan, R. M. (2010). A motivational model of video game engagement. Review of General Psychology, 14, 154–166. doi: 10.1037/a0019440 Rahford, N. (2009). Overcoming cognitive bias in group decisionmaking [White paper]. Retrieved from http://www.slideshare. net/PihMacho/raford-overcomingbias Rehak, B. (2003). Playing at being: Psychoanalysis and the avatar. In M. J. P. Wolf & B. Perron (Eds.), The video game theory reader (pp. 103–127). New York, NY: Routledge. Rieber, L. (1996). Seriously considering play: Designing interactive learning environments based on the blending of microworlds, simulations, and games. Educational Technology Research and Development, 44, 43–58. Ritterfeld, U., Cody, M., & Vorderer, P. (Eds.). (2009). Serious games: Mechanisms and effects. New York, NY: Routledge. Salen, N., & Zimmerman, E. (2003). Rules of play. Cambridge, MA: MIT Press. Sanna, L. J., Schwarz, N., & Small, E. M. (2002). Accessibility experiences and the hindsight bias: I knew it all along versus it could never have happened. Memory & Cognition, 30, 1288–1296. Shaw, A. (2014). Gaming at the edge: Gender and sexuality at the margins of gamer culture. Minneapolis, MN: University of Minnesota Press. Shaw, A., & Linebarger, D. L. (2008). The influence of computers, interactive games, and the internet on a child’s multicultural worldview. In J. K. Asamen, M. L. Ellis, & G. L. Berry (Eds.), The handbook of child development, multiculturalism, and media (pp. 333–348). Thousand Oaks, CA: Sage. Stewart, T. L., Latu, I. M., Kawakami, K., & Myers, A. C. (2010). Consider the situation: Reducing automatic stereotyping through Situational Attribution Training. Journal of Experimental Social Psychology, 46, 221–225. doi: 10.1016/j.jesp.2009.09.004 Taylor, T. L. (2006). Play between worlds: Exploring online game culture. Cambridge, MA: MIT Press. Tetlock, P. E. (1985). Accountability: A social check on the fundamental attribution error. Social Psychology Quarterly, 48, 227–236. Trepte, S., & Reinecke, L. (2010). Avatar creation and video game enjoyment: Effects of life-satisfaction, game competitiveness, and identification with the avatar. Journal of Media Psychology: Theories, Methods, and Applications, 22, 171–184. doi: 10.1027/1864-1105/a000022 Turkle, S. (1995). Life on the screen. New York, NY: Simon and Schuster. West, R. F., Meserve, R. J., & Stanovich, K. E. (2012). Cognitive sophistication does not attenuate the bias blind spot. Journal of Personality and Social Psychology, 103, 506–519. doi: 10.1037/a0028857 Yee, N., Bailenson, J. N., & Ducheneaut, N. (2009). The Proteus effect: Implications of transformed digital self-representation on online and offline behavior. Communication Research, 36, 285–312. doi: 10.1177/0093650208330254 Received June 4, 2014 Revision received June 26, 2015 Date of acceptance July 13, 2015 Published online May 20, 2016 Adrienne Shaw Department of Media Studies and Production Temple University 2020 N. 13th St. Department of MSP Philadelphia, PA 19122 USA E-mail adrienne.shaw@temple.edu

Ó 2016 Hogrefe Publishing

Adrienne Shaw is Assistant Professor in the Department of Media Studies and Production at Temple University and a Media and Communications PhD program, Philadelphia, PA. She is the author of Gaming at the Edge (University of Minnesota Press, 2014) as well as of numerous journal articles and book chapters on digital games.

Kate Kenski is Associate Professor of Communication at the University of Arizona, Tucson, AZ, where she teaches political communication, public opinion, and research methods. Her current research focuses on incivility in online forums and multimedia teaching strategies to mitigate cognitive biases.

Jennifer Stromer-Galley is Associate Professor in the School of Information Studies at Syracuse University, NY, and Director for the Center for Computational and Data Sciences. She has published over 40 journal articles, proceedings, and book chapters on the social and political uses of digital media and its effects.

Rosa Mikeal Martey is Associate Professor in the Department of Journalism and Communication at Colorado State University, Fort Collins, CO. Her research focuses on games, identity, social interaction and digital media, and can be found in New Media & Society, Information, Communication & Society, and Games Studies, among others.

Benjamin Clegg is Professor of Cognitive Psychology at Colorado State University, Fort Collins, CO. He obtained his PhD in Psychology from University of Oregon. Dr. Clegg’s research looks at: training, learning, and skill acquisition; the impact of automation on performance; the application of principles from cognitive psychology to real-world tasks and skills.

Joanna E. Lewis is a doctoral student and NSF GRFP fellow in the Applied Experimental and Human Factors Psychology program at University of Central Florida, Orlando, FL. She graduated with a BS in Psychology from Colorado State University.

Journal of Media Psychology (2018), 30(1), 16–28

A. Shaw et al., Serious Efforts at Bias Reduction

James Folkestad’s research agenda is technology-enhanced learning (TEL). His innovative publications include his work on computer-aided design and screencapture technology and most recently on serious video game-based training. He has over 50 publications and has been recognized nationally through invited presentations and with an outstanding professor award.

Journal of Media Psychology (2018), 30(1), 16–28

Tomek Strzalkowski is Director of Institute for Informatics, Logics, and Security Studies and Professor of Computer Science at SUNY Albany, NY. His research interests span computational linguistics and sociolinguistics, information retrieval, question answering, human– computer dialogue, virtual environments, and serious games. He has published over 150 scientific papers and several edited volumes.

Ó 2016 Hogrefe Publishing

The positive aspects of digital game play and its impact Topics covered include • Children’s reasons and motivations for getting involved in game play and how they benefit from such play • Exergames, designed to encourage physical activity • Games played by young people with learning difficulties • The importance of educational games and how they can benefit children’s knowledge and learning • A game application designed to help middle school students learn algebra

Mark Blades / Fran C. Blumberg / Caroline Oates (Editors)

Children and Digital Games (Series: Zeitschrift für Psychologie – Vol. 221/2) 2013, iv + 56 pp., large format US $49.00 / € 34.95 ISBN 978-0-88937-446-1 Children’s and adolescents’ exposure to and use of digital media has steadily increased in recent years, in both entertainment and education contexts. The growth of the internet and the development of interactive media such as computer games have opened up new issues that are as yet only partly understood. Furthermore, existing research has often been about the negative impact of media on young people. However, an activity that engages and motivates, such as game play, is one that can have a positive

www.hogrefe.com

impact, e.g., by stimulating knowledge and skills acquisition. This volume investigates the positive aspects of digital game play and its impact on children’s and adolescents’ learning, development, and physical activity. It contains contributions on the design, the use, and the benefits of digital games and provides a general overview of the appeal and educational ramifications of digital game play for youths.

Using movies to help learn about mental illness

“I have been a fan of Movies and Mental Illness from the first edition.” Steven Pritzker, PhD, psychology professor (Saybrook University) and former Hollywood script writer

Danny Wedding / Ryan M. Niemiec

Movies and Mental Illness

Using Films to Understand Psychopathology 4th edition 2014, xviii + 456 pp. US $59.00 / € 42.95 ISBN 978-0-88937-461-4 Also available as eBook Films can be a powerful aid to learning about mental illness and psychopathology – for students of psychology, psychiatry, social work, medicine, nursing, counselling, literature or media studies, and for anyone interested in mental health. Movies and Mental Illness, written by experienced clinicians and teachers who are themselves movie aficionados, has established a great reputation as a uniquely enjoyable and highly memorable text for learning about psychopathology. The new edition has been fully updated to include DSM-5 and ICD-10 diagnoses.

www.hogrefe.com

The core clinical chapters each use a fabricated case history and MiniMental State Examination along with synopses and discussions about specific movies to explain, teach, and encourage discussion about all the most important mental health disorders. Each chapter also includes: Critical Thinking Questions; “Authors’ Picks” (Top 10 Films); What To Read if You Only Have Time to Read One Book or Article; and Topics for Group Discussions.

Original Article

Effects of Supportive Feedback Messages on Exergame Experiences A Mediating Role of Social Presence Jihyun Kim1 and C. Erik Timmerman2 1

School of Communication Studies, Kent State University, Kent, OH, USA

Department of Communication, University of Wisconsin-Milwaukee, WI, USA

Abstract: Since their inception, the popularity of exercise-based video games, or exergames, has continued to grow. Acknowledging the rapid growth of exergames, scholars have investigated exergame play experiences with a focus on the feedback a player receives about performance (e.g., progress bar, score) and how this information can influence game play experiences. To extend the existing exergame research, this study has two primary goals: (a) to test how the supportiveness of verbal feedback messages from an exergame can influence players’ exergame experiences and (b) to evaluate the way that a player’s sense of social presence may serve as a mechanism for understanding the link between feedback and experiences. An experiment (N = 47) was conducted using a between-subjects design. Two groups were compared: highly supportive versus lowly supportive feedback condition. The findings reveal that players who receive highly supportive feedback report greater exergame experience (e.g., enjoyment) than players who receive lowly supportive feedback. Further, social presence is a key variable that mediates the relationship between feedback and exergame experiences. Although the current research findings are important, they are limited to explaining short-term, immediate responses after exergame play among college students. The current research provides implications for research and practice. In particular, the study contributes to an extended understanding of feedback in exergame research and social presence research. At a practical level, the findings suggest that exergame designers should consider incorporating more supportive social feedback messages into a game. Keywords: Exergame, presence, social presence, feedback, mediation effect

Computer and video games provide one of the most popular forms of entertainment (de Kort, IJsselsteijn, & Poels, 2007). One particular form of video games that has become increasingly popular is the exercise video game, or exergame (Lieberman, 1997). To better understand why these forms of video games have increased in popularity, a number of scholars have investigated exergame play experiences (e.g., Fogel, Miltenberger, Graves, & Koehler, 2013; Limperos, 2014; Limperos & Oliver, 2012; Lin, 2015; Lwin & Malik, 2011; Mhurchu et al., 2008; Peña & Kim, 2014; Peng, Lin, Pfeiffer & Winn, 2012; Song, Kim, & Lee, 2014; Song, Kim, Tenzek, & Lee, 2013; Song, Peng, & Lee, 2011). Among other topics, the exergame literature has paid keen attention to players’ assessments of enjoyment (e.g., Peng et al., 2012; Song, Kim, Tenzek, et al., 2013; Song et al., 2014), selfefficacy (e.g., Limperos, 2014; Song et al., 2011), game rating (e.g., Peng et al., 2012), and learning (Limperos, 2014) in diverse contexts. Perhaps even more importantly, research has indicated that exergame play may also result in positive health outcomes (Peng, Lin, & Crouse, 2011). Thus, understanding why players may enjoy exergames could be used to encourage further play and, ultimately, have a positive impact on players’ health. Ó 2016 Hogrefe Publishing

As noted in a theoretical review of exergame research (Lyons, 2015) and empirical studies (e.g., Limperos, 2014; Limperos & Oliver, 2012; Lin, 2015; Peng et al., 2012), feedback is a feature of exergames that can contribute to positive experiences. A recent content analysis of exergames reported that games provide some types of feedback during game play by its nature of the media (Lyons & Hatkevich, 2013). For example, depending on game players’ skills and progress, exergames provide players with performance feedback by elements such as a progress bar, voice feedback from virtual trainers, and numeric scores. To extend our understanding of exergames, this investigation explores the effects of feedback on exergame experiences. Using social cognitive theory (Bandura, 1986, 1997) as a framework, the current study examines how supportive feedback messages from an exergame influence players’ exergame experiences (e.g., degree of enjoyment). Additionally, given that exergame play occurs in the virtual context (game context), the current study also examines players’ feeling of presence, particularly social presence, to better understand players’ exergame experiences. Journal of Media Psychology (2018), 30(1), 29–40 DOI: 10.1027/1864-1105/a000175

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

Effects of Feedback According to SCT, verbal persuasion focuses on providing individuals with encouraging and supportive messages/ remarks about their behaviors (Bandura, 1986, 1997, 2001). Essentially, Bandura argues that people receiving encouraging messages increase effort to accomplish their objectives, which consequently result in positive outcomes, such as self-efficacy. Thus, the purpose of verbal persuasion is to get people to believe they possess capabilities that will enable them to accomplish a task, whether a project in the office, an assignment for a class, or a stage in an exergame. These messages often serve the function of providing individuals with forms of direction and support for their efforts. When verbal persuasion is framed in terms of encouragement and support, the notion can be understood as supportive messages. Supportive messages can be a form of feedback. These messages inform recipients that their performance has achieved a standard and that they have been making progress (Ilgen, Fisher, & Taylor, 1979; Latham & Locke, 1991). In the context of exercise, there is empirical evidence that links supportive feedback to more favorable exercise experiences and outcomes (Deci, Koestner, & Ryan, 1999). Specifically, participants who receive supportive feedback during physical activities experience heightened self-efficacy and task performance (Escarti & Guzman, 1999; Fitzsimmons, Landers, Thomas, & van der Marsh, 1991) as well as increased competence and intrinsic motivation (e.g., Bindarwish & Tenenbaum, 2006; Gernigon & Delloye, 2003). Although the source of feedback may be another person, such as physical trainer or teacher (e.g., Bindarwish & Tenenbaum, 2006; Deci et al., 1999; Escarti & Guzman, 1999; Fitzsimmons et al., 1991; Gernigon & Delloye, 2003), research has also demonstrated effectiveness of supportive feedback from nonhuman actors such as computers. Bracken and Lombard (2004) examined the effects of computer feedback on task performance. Computer users receiving supportive comments (e.g., “You are doing great”) perceived greater ability, recall, and memory recognition compared with those receiving less supportive comments (e.g., “Okay”). Similarly, people being flattered (i.e., social praise) while performing a computer-based task reported positive experiences. In comparison to participants who did not receive positive comments, individuals who received positive comments indicated that they felt better after the performance (positive mood), expressed a more favorable evaluation about the interaction with the computer, and reported that the computer was more attractive (Fogg & Nass, 1997). In sum, there is solid empirical support for the notion that supportive feedback leads to improvements in perceived outcomes. Journal of Media Psychology (2018), 30(1), 29–40

Directly germane to the current investigation, exergame research has also examined the role of feedback (e.g., Kim, Prestopnik, & Biocca, 2014; Limperos, 2014; Limperos & Oliver, 2012; Lin, 2014; Peng et al., 2012). Recent work (Limperos & Oliver, 2012) explored effects of exercising via an exergame, which provides performance feedback to players in an interactive game format, versus exercising via watching an exercise video. Results indicated that receiving feedback from an exergame led to stronger feelings of presence and competence than watching an exercise video. Separately, Peng and colleagues (2012) investigated effects of performance feedback by adjusting difficulty levels in an exergame depending on a player’s performance. In the feedback-on condition, players received feedback in the form of either increasing or decreasing task difficulty. If players were performing well, the game would increase difficulty of the task; likewise, if they were performing poorly, difficulty would decrease. Compared with the feedback-off position (difficulty did not vary with performance), the participants who received feedback reported more favorable game play experiences such as enjoyment, motivation for future play, self-efficacy, and game rating. Further, Kim and colleagues (2014) found that exergame feedback delivered with multiple channels (e.g., visual, audio) contributed to greater enjoyment and energy expenditure compared with feedback from fewer channels. To date, empirical work has effectively documented the positive effects of feedback in exergames. However, what has not been fully addressed is whether the type of feedback matters. As discussed earlier, existing studies indicate that use of feedback in exergames facilitates favorable experiences. Particularly, receiving feedback versus not receiving feedback (e.g., Peng et al., 2012) and receiving a higher amount versus a lower amount of feedback (e.g., Kim et al., 2014) increase favorable exergame perceptions. However, it is not known whether variation in the content of feedback, in particular a degree of supportiveness in feedback messages, matters in players’ exergame experiences. In line with published work finding clear positive associations between supportive feedback and perceptions of exercise (Bindarwish & Tenenbaum, 2006; Deci et al., 1999; Escarti & Guzman, 1999; Fitzsimmons et al., 1991; Gernigon & Delloye, 2003), the contention of this investigation is that supportive feedback will yield similarly positive perceptions of exergame participation. Consistent with Limperos’s (2014) SCT-based argument, feedback from exergames can have a direct influence on a variety of assessments of the exergame experience. We see the game play experience as composed of a variety of variables and identify three as particularly relevant for understanding the impact of feedback on exergames. Thus, the current study focuses on self-efficacy for exergame play, Ó 2016 Hogrefe Publishing

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

perceived usefulness of the exergame, and exergame play enjoyment. First, self-efficacy is a driving force in SCT and refers to an individual’s perceived belief about the capability to control motivation, emotional states, and behaviors (Bandura, 1986, 2001). Although not yet studied in exergames, research in other contexts links variations in feedback to self-efficacy. For instance, supportive feedback has been positively associated with medical students’ self-efficacy for medical education (van de Ridder, Peters, Stokking, de Ru, & ten Cate, 2015) and organizational employees’ self-efficacy for performing a task (Schul & Ganzach, 1995). Research in gaming contexts emphasizes self-efficacy as an important game play experience, indicating that those with stronger self-efficacy invest more effort to overcome barriers and continue playing a game (e.g., Klimmt & Hartmann, 2006). This notion has received considerable attention and has been used as an important variable for understanding exergame experiences (e.g., Limperos, 2014; Limperos & Oliver, 2012; Peng, 2009; Song et al., 2011). In particular, Peng and colleagues (2012) tested how performance feedback influences players’ self-efficacy in an exergame play context. The study found that players who received feedback through variations of game play’s difficulty levels reported stronger self-efficacy than did players who did not receive feedback (no adjustment on the game level regardless of their performance). Thus, based on extant research suggesting that general forms of feedback have positive impacts on exergame play, the SCT argument that supportive messages enhance self-efficacy, and findings from an array of non-exergame settings that link supportive feedback to self-efficacy, the current study predicts that highly supportive feedback increases players’ self-efficacy. A second assessment of game play experience is perceived usefulness (Davis, 1989), which reflects an overall favorability toward the exergame and a view that engaging in game play has benefits for the player. Although the term is not directly associated with SCT, other research points to perceived usefulness as being an important assessment of exergame play experience (Limperos, Downs, Ivory, & Bowman, 2013; Song, Kim, Tenzek, et al., 2013; Song et al., 2014). In an extensive review of game research, Limperos and colleagues (2013) argued that while assessing behavioral intentions and attitudes of a particular activity in the game have received a good deal of attention, assessing intentions or related variables for future use of the particular game (e.g., usefulness, favorability, or evaluation of the game) has been neglected. Essentially, understanding what motivates game players to engage in future game play is important (Limperos et al., 2013) and players are more likely to play again when they perceive a game as useful. Thus, exploring the way that feedback influences players’ Ó 2016 Hogrefe Publishing

perceived usefulness of the game helps address a need for additional information about potential future use of the game. Although their exergame study only focused on feedback quantity (rather than variations in supportiveness), Kim and colleagues (2014) found that greater amounts of feedback induced stronger intention to play the game in the future. Although not identical to usefulness, a player’s intention for future use and perceived usefulness might be understood in a similar vein given that both assess the evaluation of the game and future intention. Although there is no direct empirical research that supports the effect of supportive feedback on usefulness in an exergame yet, published exergame research establishes the idea that exergame characteristics can have an influence on perceptions of usefulness and that, indeed, this is a key variable for learning more about players’ future intentions to pursue exergame activities (Limperos et al., 2013). Thus, based on SCT support for the effect of encouraging messages, the current study predicts that highly supportive feedback increases perceived usefulness of the game. Third, enjoyment is the sum of favorable reactions toward an activity (Ritterfeld & Weber, 2006). Indeed, some scholars argue that enjoyment may be the core experience of game play (Vorderer, Klimmt, & Ritterfeld, 2004). And, because enjoyment is closely related to intrinsic motivation (Ritterfeld & Weber, 2006), this outcome has been examined as one of the most important variables in exergame research. As emphasized in a theoretical review of exergame research (Lyons, 2015) and empirical research (Kim et al., 2014), feedback is one of the key factors that can influence enjoyment. Thus, the current study predicts that supportive feedback would enhance game play enjoyment. In sum, by focusing on these three major outcome variables (self-efficacy, usefulness, enjoyment), the current study proposes the following set of hypotheses. Hypothesis 1 (H1a–c): Individuals receiving highly supportive feedback messages during exergame play report higher levels of (a) self-efficacy for exergame play, (b) perceived usefulness of the exergame, and (c) game enjoyment than do individuals receiving lowly supportive feedback messages.

Social Presence When engaged in exergame play, the perceptual boundary that exists between the physical world and virtual world (game context) may become blurred. That is, although game players are physically present in their own places, their psychological senses may be immersed in the gaming Journal of Media Psychology (2018), 30(1), 29–40

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

context. The feeling of “being there” in the virtual world is simply understood as presence (Biocca, 1997). Of various types of presence such as social, physical, and self-presence (Lee, 2004a), the current study’s focus is on social presence. First described as being aware of the other person in online environments in early social presence research (Short, Williams, & Christie, 1976), the notion has been rigorously investigated (Edwards, Edwards, Spence, & Westerman, 2015; Jin, 2009, 2011; Lee & Nass, 2004; Spence, Westerman, Edwards, & Edwards, 2014). In particular, as technology evolves daily, recent studies have suggested broader and more extended descriptions of social presence, which can be applied to a variety of technology domains, such as perceiving virtual social actors (e.g., video game characters) like actual social actors (Lee, 2004a). For a more organized and unified understanding of social presence, Biocca, Harms, and Burgoon (2003) have classified social presence into categories such as co-presence (e.g., feeling of being together with the virtual social actors) and psychological involvement (e.g., social richness), which originated from early social presence research (Short et al., 1976). From a theoretical perspective, social presence (or presence) is related to Reeves and Nass’s (1996) the media equation (TME) phenomena (Lee, 2004b). The core of TME is that people perceive that media equal real life and that technology users apply social interaction and categorization rules to computers, although they may acknowledge that it is somewhat absurd (Reeves & Nass, 1996). In other words, while people are engaged in virtual experiences, they may respond to the cues from the technology as if the cues were from people (e.g., perceiving virtual objects as actual objects). TME phenomena occur as long as people feel social presence (presence) during media experiences (Lee, 2004b). That is, people would respond to computers and technologies as they do to other social actors when people feel social presence (Lee, 2004b). In this regard, the current study argues that the reason exergame players respond to feedback from an exergame and the way feedback from an exergame influences exergame play outcomes is via social presence. That is, from a theoretical perspective, social presence functions as a mediating mechanism that eventually leads to effective outcomes during exergame experiences. This argument is also supported by empirical studies that have found a mediating role of social presence (or presence in a broad sense) in technology user experiences (e.g., Jin, 2009, 2011; Jin & Park, 2009; Lee & Nass, 2004; Lee, Park, & Song, 2005; Lee, Peng, Yan, & Jin, 2006). Thus, a second hypothesis is proposed: Hypothesis 2 (H2a–c): Social presence mediates the relationship between feedback messages and

Journal of Media Psychology (2018), 30(1), 29–40

exergame experiences such as (a) self-efficacy for exergame play, (b) perceived usefulness of the exergame, and (c) enjoyment.

Method Participants A total of 47 undergraduate students at a large public Midwestern university in the US participated in an experiment. The average age of the participants was 21.75 years (SD = 3.46). Slightly more females (n = 25, 53.2%) participated in the study than males (n = 22, 46.8%). The sample consisted of Caucasian (n = 30, 63.8%), African American (n = 2, 4.3%), Hispanic (n = 4, 8.5%), and other ethnic groups (n = 11, 23.4%). All participants were randomly assigned into one of the two conditions (explained in next section): highly supportive feedback condition (n = 23) or lowly supportive feedback condition (n = 24).

Procedure Following approval by the university’s Institutional Review Board (IRB), initial recruitment was announced in face-toface and online (e.g., e-mail) undergraduate communication courses. Individuals who agreed to participate in the study were contacted via e-mail to arrange a time for participation in the experiment. During this time, basic information about each participant was obtained for the purpose of creating an avatar with an appearance similar to the participant in terms of biological sex (male vs. female), perceived height (tall vs. average vs. short), and perceived weight (thin/slim vs. average vs. heavy). Then, participants were invited to a research lab at a scheduled date and time. Upon arrival, each participant was provided with instructions explaining how to play the game and was allowed a 70-s practice session to ensure that they understood the basic features of the game. Participants were then informed that they would need to play the game for a total of 12 min, at which point the researcher would notify them that the time had elapsed. Then, the researcher left the lab to go to an observation room while the participants started playing the game. Upon completion of the game, participants were immediately taken to a different room where they completed a set of questionnaires. Participants were thanked and asked to avoid discussing the experiment with others. They were then allowed to leave the lab.

Ó 2016 Hogrefe Publishing

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

Materials and Manipulation The exergame used in the study was the Nintendo Wii Fit Hula Hoop game. All features of the original hula hoop game were left intact but augmented by additional social feedback messages. In the highly supportive condition, participants received feedback such as, “You are doing great! Keep it up!” and “Wow! You are making good progress!” In the lowly supportive condition, participants received feedback such as, “You can do better than this” and “Well, I don’t think you are making very good progress.” The feedback messages were delivered through two channels: voice and text during game play. Voice messages were recorded and then manipulated using audio software to mimic the Wii system’s built-in voice messages (e.g., prompts telling the user that the game is about to begin and a countdown). Next, a PowerPoint presentation was created that contained separate slides with the text of each feedback message and an embedded audio file containing the prerecorded spoken message. As the hula hoop game was projected from a ceiling-mounted LCD projector, the text-based version of the feedback messages was beamed from a separate LCD projector that was hidden. At appropriate times during game play, the hidden projector beamed the feedback messages onto an open space on the game screen and the audio messages were played over the same speaker system as the game’s audio. This configuration made it possible to project the game itself as well as the text and audio messages simultaneously. Presentation of the text and audio feedback messages was triggered via remote control from a separate control room equipped with a one-side mirror. The researchers made sure that each participant received an equal administration of 25 messages during the 12-min of game play. The playing time for each set of the hula hoop game was 70 s. Although there were some minor variations depending on how fast or slow a participant would restart a new set, most participants would play approximately six to eight sets during the 12 min of game play. Thus, researchers made sure to deliver approximately three to four feedback messages for each set of the game to spread feedback delivery relatively evenly throughout the entire game play. To ensure the equal administration of 25 messages, researchers kept a track of the number during each game play. Participants were randomly assigned to either the high or low supportive feedback condition. Within each condition (either high or low supportiveness) feedback messages that were either high or low in supportiveness (according to treatment condition) were delivered in a way to keep the content of the feedback consistent with the quality of the player’s performance. Participants who were assigned to the highly supportive condition received 25 highly supportive messages during game play and participants assigned to Ó 2016 Hogrefe Publishing

the lowly supportive condition received 25 lowly supportive messages. In the highly supportive condition, messages were delivered in response to positive aspects of performance. Specifically, the highly supportive messages were provided (a) when participants were succeeding at the game’s primary performance criteria (e.g., catching hula hoops), (b) when they were maintaining good performance (e.g., continuing to spin a hula hoop), and/or (c) when they were continuing to make progress without error. In the lowly supportive condition, messages were provided in response to negative performance elements. Specifically, lowly supportive messages were provided (a) when participants were making errors (e.g., failing to catch hula hoops), (b) when they were not maintaining good performance (e.g., dropping spinning hula hoops), and/or (c) when they were not making progress. Because participants were randomly assigned to either a highly or lowly supportive condition, regardless of hula hoop skill level, the researchers anticipated that either participants in the lowly supportive condition could perform well or that participants in the highly supportive condition might not excel in performance. An issue was how to time the delivery of either highly supportive or lowly supportive messages at times that appeared (to players) to be consistent with other aspects of game play. More specifically, we sought to avoid situations in which players who were performing well would receive messages stating that they were not making good progress or the inverse. To address this, messages were delivered in a manner so that the feedback was consistent with both objective and subjective elements of game performance. Specifically, the hula hoop game has two major tasks that players complete: (a) catching hula hoops thrown by other game characters and (b) spinning. Mostly, the primary focus of feedback delivery was on the task of catching or missing hula hoops. However, when dealing with a mismatch of performance and feedback (e.g., good performance in a lowly supportive condition), feedback delivery was mostly focused on the spinning task. For example, when a participant in a lowly supportive condition performed well and did not make a mistake of dropping a hula hoop, lowly supportive messages were delivered while the participant was spinning a hula hoop. Although catching hula hoops can be evaluated more objectively (catching vs. missing), spinning performance can be evaluated somewhat subjectively, as the criterion of the speed of spinning is a relevant quality that players can perceive while playing the game. So, in the case in which a participant assigned to the low supportiveness condition was successfully catching hula hoops, feedback was delivered when the player seemed to only be maintaining the same speed and/or when there was a decrease in spinning speed during the performance. Journal of Media Psychology (2018), 30(1), 29–40

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

The same approach was used when participants in a highly supportive condition did not perform well. In that condition, when a participant kept missing a hula hoop, feedback was typically delivered when the participant was maintaining the spinning action, which can be evaluated more subjectively. Given that perceived similarity to an avatar influences user experiences (e.g., Fox & Bailenson, 2009; Song, Kim, Kwan, & Jung, 2013), the current study took steps to avoid any potential avatar effects. Thus, the study ensured that all the participants played with an avatar that was similar to their physical attributes. Specifically, based on information collected before the experiment (e.g., biological sex, height, weight), an avatar that matched each participant’s physical attributes was created. For example, a tall and thin female participant was assigned to play with a tall and thin female avatar. All participants agreed to play with the given avatar. Prior to conducting the experiment, the researchers contacted several colleagues who were blind to the experiment and procedures. These individuals were asked to play the exergame that provided manipulated feedback messages made for the experiment. All reported that the procedure was smooth and they did not realize manipulated feedback messages (audio or visual) were separate from the game’s built-in features until they were informed of the manipulation.

Measures A set of questionnaires contained measures that were designed to assess the effectiveness of the manipulations (highly supportive vs. lowly supportive feedback) and participants’ exergame experiences. Responses were obtained on a 7-point Likert-type scale (1 = strongly disagree, 7 = strongly agree). First, to assess whether participants experienced the supportive and nonsupportive feedback conditions as intended (manipulation check), feedback message perception (α = .95) was measured with three items (e.g., “I received supportive messages when I was playing the game”). Other variables were assessed in the same manner. Social presence (α = .76), particularly focusing on psychological involvement (social richness; Biocca et al., 2003; Lombard & Ditton, 1997; Short et al., 1976), was measured with five items, adapted from previous social presence research (Lombard & Ditton, 2000; Short et al., 1976; e.g., “The experience was sociable”). Self-efficacy for game play (α = 0.94) was measured with six items, modified from Kroll, Kehn, Ho, and Groah (2007; e.g., “I am confident that I can play the exergame well even when I am tired”). Perceived usefulness of the exergame (α = .91) was measured with three items, modified from Davis (1989; e.g.,

Journal of Media Psychology (2018), 30(1), 29–40

“Using the exergame would increase the effectiveness of exercise”). Exergame enjoyment (α = .92) was measured with four items, modified from Song Kim, Tenzek et al. (2013; e.g., “This game experience was fun”).

Results Manipulation Check An initial independent samples t test was performed to confirm the feedback manipulation. As expected, participants in the highly supportive feedback condition (M = 6.23, SD = 0.64) reported that the feedback was significantly more supportive and encouraging, t(45) = 17.70, p = .000, d = 5.17, than did those in the lowly supportive feedback condition (M = 1.93, SD = 0.98). Thus, the manipulation was successful.

H1a–c: Effects of Feedback Hypotheses 1a–c predicted that, compared with participants in the lowly supportive feedback condition, participants in the highly supportive feedback condition would report higher levels of (a) exergame self-efficacy, (b) perceived exergame usefulness, and (c) enjoyment. Because H1a–c are directional hypotheses, a p value from SPSS was divided by 2 in order to get the adequte directional p value. For H1a, an independent samples t test indicated that there is no significant difference in exergame self-efficacy between the highly supportive condition (M = 5.60, SD = 0.87) and lowly supportive condition (M = 5.67, SD = 1.20), t(45) = 0.24, p = .41, d = .07. Thus, H1a was not supported. With regard to perceived exergame usefulness (H1b), participants in the highly supportive condition (M = 5.83, SD = 1.29) reported significantly higher perceived usefulness of the exergame than did participants in the lowly supportive condition (M = 5.22, SD = 1.26), t(45) = 1.64, p = .05, d = 0.48. Thus, H1b was supported at a marginal level. The comparison of enjoyment (H1c) across the highly supportive (M = 6.23, SD = 0.79) and lowly supportive condition (M = 5.60, SD = 1.02) was significant, t(45) = 2.37, p = .01, d = .69, indicating support for H1c (see Table 1).

H2a–c: Mediating Role of Social Presence Hypotheses 2a–c predicted that participants’ feeling of social presence would mediate the relationships between feedback type (highly supportive vs. lowly supportive) and (a) exergame self-efficacy, (b) perceived usefulness of the exergame, and (c) enjoyment. Each hypothesis was tested

Ó 2016 Hogrefe Publishing

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

Table 1. The effects of feedback messages on manipulation and outcome variables (H1a–H1c) Outcome variables

Feedback condition

Feedback perception

H L

6.23 0.64 1.93 0.98

17.70*

Exergame self-efficacy (H1a)

H L

5.60 0.87 5.67 1.20

0.24

Usefulness (H1b)

H L

5.83 1.29 5.22 1.26

1.64+

Enjoyment (H1c)

H L

6.23 0.79 5.60 1.02

2.37*

Notes. Directional hypotheses: H = highly supportive; L = lowly supportive. + p = .05. *p < .05.

with a series of bootstrapping procedures for estimating indirect effects in simple mediation models (Preacher & Hayes, 2004). With this procedure, an SPSS macro produces output for both the bootstrap test and regression coefficient on each path in the model simultaneously. Bootstrapping involves the simulation of multiple iterations of sampling from the data set to estimate the indirect effect in each sample. For these analyses, the procedure was based on 5,000 bootstrap samples, as recommended by Preacher and Hayes. In each bootstrapping test, results were interpreted based on 95% of confidence interval (CI). Although mediation effects and indirect effects are often used interchangeably, there is a distinction between them (Holmbeck, 1997). A mediation effect is considered a special case of indirect effect. A mediation effect implies that the total effect of an independent variable (X) on a dependent variable (Y) is initially present. In other words, in order to be a mediator, the effect of X on Y should be significant. However, an indirect effect, a broader conceptualization of a mediation effect, does not assume that the total effect of X on Y should be present. Although there may not be evidence for a significant total effect of X on Y, a significant indirect effect between X and Y through M (mediator) may exist (Holmbeck, 1997). Further, there are two types of mediation: full and partial mediation. Full mediation occurs when the direct effect of X on Y becomes statistically nonsignificant when controlling for M. Partial mediation occurs when the direct effect of X on Y still remains significant but decreases in size when controlling for M. Thus, bootstrap tests were interpreted to identify potential indirect effects, full mediation, or partial mediation. The first test was performed to assess indirect effects of social presence on exergame self-efficacy. The mean score for the indirect effect of social presence was 0.07 (SE = 0.29; CI = 0.63–0.53), indicating that there was no significant indirect effect of social presence between feedback and self-efficacy for exergame play. H2a was not supported. Regarding perceived usefulness of the exergame, a significant mediation effect was present. The mean score for the indirect effect of social presence was 0.94 (SE = 0.32; Ó 2016 Hogrefe Publishing

CI = 0.34–1.60). Subsequent evaluation of regression results indicated a significant bivariate relationship between feedback and perceived usefulness at a marginal level, b = .61, p = .05. When controlling for social presence, the relationship was no longer significant, b = .36, p = .20. This result indicates a full mediation effect of social presence. That is, highly supportive feedback increased the perception of social presence, which enhanced perceived usefulness of the exergame. H2b was supported. Finally, a bootstrap test was performed on enjoyment, which also presented a significant mediation effect. The mean score for the indirect effect was 0.71 (SE = 0.23; CI = [0.29–1.18), indicating a significant indirect effect for social presence. Subsequent evaluation of regression results indicated the bivariate relationship between feedback and enjoyment was significant, b = .64, p = .01. When controlling for social presence, the relationship was no longer significant, b = .10, p = .37, which indicates a full mediation effect. Therefore, highly supportive feedback messages increased a sense of social presence that, in turn, facilitated game enjoyment, which supports H2c (see Table 2 and Figure 1).

Discussion This study examined the effect of feedback messages on participants’ exergame experiences. Overall findings indicate that supportive feedback increases perceptions of exergame usefulness and enjoyment. One reason feedback works this way is because it impacts players’ perceptions of social presence, which mediates the relationship between feedback and exergame experiences (perceived usefulness and enjoyment).

Primary Findings and Implications for Research One of the primary purposes for this investigation was to assess whether participants’ exergame experiences could be influenced by supportive, social feedback from the exergame. Data suggest that supportive feedback enhances favorable exergame experiences. These effects of supportive feedback on people’s experiences are consistent with previous findings in a variety of non-exergame contexts (e.g., Bracken & Lombard, 2004; Deci et al., 1999; Viciana & Cervello, 2007). Although feedback has been studied in exergame research (e.g., Kim et al., 2014; Limperos, 2014; Limperos & Oliver, 2012; Lin, 2014; Peng et al., 2012), most of the previous work has focused on effects of existence of feedback (feedback vs. no feedback) and the amount of feedback (high vs. low). Until the present investigation, very Journal of Media Psychology (2018), 30(1), 29–40

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

Table 2. Bootstrap results: mediating role of social presence between feedback and outcome variables (H2a–H2c) Outcome variable

Mean for indirect effect

Lower CI

Upper CI

Social presence effect

Exergame self-efficacy

0.07

0.29

0.63

0.53

No effect

Usefulness

0.94*

0.32

0.34

1.60

Full mediation

Enjoyment

0.71*

0.23

0.29

1.18

Full mediation

Note. *p < .05.

H2b: Social presence mediates the relationship between feedback messages and perceived usefulness of the exergame (supported).

H2a: Social presence mediates the relationship between feedback messages and self-efficacy for exergame play (not supported).

1.60*

Feedback

Social Presence

0.03

Exergame Self Efficacy

-0.18 (-0.07)

Social Presence

1.60*

Feedback

-0.36 (-0.61*)

0.52*

Usefulness

H2c: Social presence mediates the relationship between feedback messages and enjoyment (supported).

1.60*

Feedback

Social Presence

-0.10 (0.64*)

0.44*

Enjoyment

Figure 1. Illustrations of hypothesis tests (2a–c) assessing the mediating role of social presence. The number inside the parenthesis is the unstandardized coefficient from when the dependent variable is linked to the independent variable alone, without including presence in the equation. The independent variable was dummy coded (0: Lowly supportive feedback, 1: Highly supportive feedback). *p < .05.

little was known about how different types of feedback, particularly different levels of support in feedback messages, could influence players’ exergame experiences. The current study’s finding that variation in the supportiveness of feedback has an impact on the game experience is key for better understanding not just whether the amount of feedback matters, but the actual content matters as well. Unlike the argument of SCT that encouraging and supportive remarks increase people’s self-efficacy, the current study did not find any significant differences in exergame self-efficacy across types of feedback messages. One of the potential explanations may be related to the nature of the task in this exergame. Hula Hoop, the particular game employed in this study, may be one that can be mastered fairly quickly by users. Because experience of mastery is one of the strongest sources influencing self-efficacy (Bandura, 1997), just playing the exergame itself may have made users feel capable of performing it regardless of variations of feedback presented within the game. That is, the simple and repeated act of game play, which was consistent across the feedback conditions, may have helped users Journal of Media Psychology (2018), 30(1), 29–40

establish an adequate level of self-efficacy. Another potential explanation may be related to a college student sample. As previous research noted, college students tend to have high self-efficacy (Limperos, 2014). This exercise self-efficacious group of participants might have resulted in no significant differences between the two groups in this study. Another primary purpose of this study was to examine the role of social presence. The findings here indicate that social presence mediates the relationship between feedback from the game and two measures of a game player’s experiences and perceptions about playing the game. That is, supportive feedback messages increase the feeling of social presence, which leads to more favorable exergame experiences (perceived usefulness of the exergame, enjoyment). The results that reveal presence as a mediator are consistent with the empirical findings from a variety of virtual contexts (e.g., Jin, 2009, 2011; Jin & Park, 2009; Lee & Nass, 2004; Lee et al., 2005, 2006; Song et al., 2014). From a theoretical perspective, TME phenomena (Reeves & Nass, 1996) also support the justifications of the mediating role of social presence. During the virtual Ó 2016 Hogrefe Publishing

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

experiences, technology users may not notice the existence of the medium/technology and feel social presence, thus this experience may allow them to perceive feedback from the exergame as if the cues were from humans (Reeves & Nass, 1996). Overall, the finding on the mediating role of social presence indicates that social presence is a central variable that facilitates effective virtual experiences. Additionally, the study’s finding of a mediating role for social presence in relation to feedback makes further contributions to social presence research. A few recent studies have investigated associations among presence, enjoyment, and exergame experiences. Specifically, Lyons and colleagues (2014) found that presence (named as engagement in their study but assessed with a presence measure) increases exergame play enjoyment, which in turn leads to energy expenditure. Limperos and Oliver (2012) found that perceived feedback increases a feeling of presence. Further, presence increases exergame play enjoyment, which consequently leads to intention for continued use of an exergame (Limperos & Oliver, 2012). In these studies, presence was examined as a key variable; however, the role of presence was limited to an associated outcome in relation to feedback (Limperos & Oliver, 2012) or a variable that indirectly influences exergame play experiences via enjoyment (Limperos & Oliver, 2012; Lyons et al., 2014). In this sense, social presence as a mediating variable in the current study’s exergame context contributes to a further understanding of the concept. Somewhat similar to the current study’s finding, Kim and colleagues (2014) found that spatial presence, another type of presence, mediates the relationship between the amount (quantity) of feedback cues and exergaming intention. However, given that a mediating role of social presence in relation to feedback has not yet been identified in the extant research, this study’s finding is more meaningful. Furthermore, in addition to Kim et al.’s outcome variable, exergaming intention, the current study reveals that other exergame experiences, particularly perceived usefulness of the exergame and enjoyment, can be also explained through presence. In this regard, the current study’s finding contributes to a further understanding of social presence, and presence research as well.

Implications for Practice At a practical level, the findings from this study suggest that supportive social feedback messages produce beneficial effects on user experiences in virtual contexts. Previous research (Midden & Ham, 2008) examined the effects of supportive social feedback about energy use from a social robot compared with factual feedback from the robot during virtual simulation of washing machine use. Participants who received social feedback (e.g., “fantastic” – when Ó 2016 Hogrefe Publishing

the energy level was low) used less electricity than participants who received factual feedback (e.g., actual energy consumed levels). These findings imply that social feedback, particularly supportive social feedback, can more favorably influence technology users’ experiences than factual feedback. Supportive social feedback may also have important implications for target audiences in fitness and exergame settings. Most exercise machines (e.g., treadmills) provide only factual feedback based on a machine user’s performance such as calories consumed, mileage, and heart rate. It may be worth considering the incorporation of supportive social feedback (e.g., “You are doing great!”) as a core part of information the system provides to the user. The same implication can also be suggested to exergame designers. In addition to the game score feedback, which is a common feedback type within the game, game designers may consider incorporating supportive social feedback messages to the game that provide a verbal interpretation about the meaning of the scores or other numerical indicators. As found in this current investigation, this type of messages favorably influences users’ exergame experiences, which may provide incentive for continuing a practice that could eventually produce positive exercise outcomes (Peng et al., 2011).

Limitations and Future Research Directions As with any study, this investigation has limitations that should be kept in mind when interpreting the pattern of results. First, the sample used for the investigation was relatively small and drawn from a largely homogeneous student population. Although neither characteristic necessitates outright rejection of our findings, we do note that a larger sample could increase the potential to detect smaller effects and that increased homogeneity of the sample would introduce greater variation in participants’ perceptions and experiences prior to playing the game. Related to our study design, although this study randomly assigned participants into either a highly or lowly supportive condition, owing to a relatively small sample size, this random assignment might not have effectively taken care of potential individual differences. Second, the current investigation is limited to examining short-term, immediate responses after exergame play. To effectively address this issue, future research should explore how people’s exergame experiences change over a longer period. On a related note, it may be also important to assess how experiences of self-efficacy change over time. Self-efficacy may change over a long period with continued practices and experiences. If this is the case, the finding

Journal of Media Psychology (2018), 30(1), 29–40

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

may provide an effective way for enhancing and maintaining exergame players’ self-efficacy, which might eventually help to develop greater exercise self-efficacy, one of the key factors for exercise behavior (Anderson, Wojcik, Winett, & Williams, 2006; Garcia & King, 1991; McAuley, Blissmer, Katula, & Duncan, 2000; Rodgers, Hall, Blanchard, McAuley, & Munroe, 2002; Rovniak, Anderson, Winett, & Stephens, 2002). Additionally, this investigation only measured one category of social presence, psychological involvement (social richness). Considering various aspects of social presence such as co-presence (Biocca et al., 2003), future research needs to investigate more diverse dimensions of social presence as well as other types of presence (e.g., physical presence, self-presence) for a more complete understanding of presence research. Lastly, because of the random assignment method for feedback types (highly supportive vs. lowly supportive), it might be possible that a person who was performing above the average might have received lowly supportive feedback. In this case, a participant’s perception of the feedback might have influenced responses toward the exergame play experiences. Thus, future studies may wish to take additional steps to separate participants by overall game performance.

Conclusion Current exergame research has documented the important role of the amount of feedback in exergames (e.g., Kim et al., 2014; Limperos, 2014; Limperos & Oliver, 2012; Lin, 2015; Peng et al., 2012). In particular, research has found the effectiveness of including feedback (vs. no feedback) and a high amount of feedback (vs. low) on exergame play experiences. Continuing the research in this realm, the current study further investigated a characteristic of the content of feedback, supportiveness, in exergame play. Findings indicated that supportive social feedback messages during the game play favorably influence users’ exergame experiences. Further, social presence mediates the relationship between feedback and exergame experiences. These findings contribute to a more comprehensive understanding of feedback in exergames and social presence research.

References Anderson, E. S., Wojcik, J. R., Winett, R. A., & Williams, D. M. (2006). Social-cognitive determinants of physical activity: The influence of social support, self-efficacy, outcome expectations, and self-regulation among participants in a

Journal of Media Psychology (2018), 30(1), 29–40

church-based health promotion study. Health Psychology, 25, 510–520. doi: 10.1037/0278-6133.25.4.510 Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: Freeman. Bandura, A. (2001). Social cognitive theory of mass communication. Media Psychology, 3, 265–299. doi: 10.1207/ S1532785XMEP0303_03 Bindarwish, J., & Tenenbaum, G. (2006). Metamotivational and contextual effects on performance, self-efficacy, and shifts in affective states. Psychology of Sport and Exercise, 7, 41–56. doi: 10.1016/j.psychsport.2005.04.001 Biocca, F. (1997). The cyborg’s dilemma: Progressive embodiment in virtual environments. Journal of Computer MediatedCommunication, 3(2). Retrieved from http://jcmc.indiana.edu/ vo/issue2/biocca2.html Biocca, F., Harms, C., & Burgoon, J. (2003). Toward a more robust theory and measure of social presence: Review and suggested criteria. Presence, 12, 456–480. doi: 10.1162/ 105474603322761270 Bracken, C. C., & Lombard, M. (2004). Social presence and children: Praise, intrinsic motivation, and learning with computers. Journal of Communication, 54, 22–37. doi: 10.1111/ j.1460-2466.2004.tb02611.x Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13, 319–340. Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125, 627–668. doi: 10.1037/0033-2909.125.6.627 de Kort, Y. A. W., IJsselsteijn, W. A., & Poels, K. (2007). Digital games as social presence technology: Development of the social presence in gaming questionnaire (SPGQ). Proceedings of the 10th Annual International Workshop on Presence, Barcelona, Spain. Edwards, C., Edwards, A., Spence, P. R., & Westerman, D. (2015). Initial interaction expectations with robots: Testing the humanto-human interaction script. Communication Studies, 1–12. doi: 10.1080/10510974.2015.1121899 Escarti, A., & Guzman, J. F. (1999). Effects of feedback on selfefficacy, performance, and choice in an athletic task. Journal of Applied Sport Psychology, 11, 83–96. doi: 10.1080/ 10413209908402952 Fitzsimmons, P. A., Landers, D. M., Thomas, R. J., & Van der Mars, H. (1991). Does self efficacy predict performance in experienced weight lifters? Research Quarterly for Exercise and Sport, 62, 424–431. doi: 10.1080/02701367.1991.10607544 Fogel, V. A., Miltenberger, R. G., Graves, R., & Koehler, S. (2013). The effects of exergaming on physical activity among inactive children in a physical education classroom. Journal of Applied Behavior Analysis, 43, 591–600. doi: 10.1901/jaba.2010.43-591 Fogg, B. J., & Nass, C. (1997). Silicon sycophants: The effects of computers that fatter. International Journal of HumanComputer Studies, 46, 551–561. doi: 10.1006/ijhc.1996.0104 Fox, J., & Bailenson, J. N. (2009). Virtual self-modeling: The effect of vicarious reinforcement and identification on exercise behaviors. Media Psychology, 12, 1–25. doi: 10.1080/ 15213260802669474 Garcia, A. W., & King, A. C. (1991). Predicting long-term adherence to aerobic exercise: A comparison of two models. Journal of Sport & Exercise Psychology, 13, 394–410. Gernigon, C., & Delloye, J. (2003). Self-efficacy, causal attribution, and track athletic performance following unexpected success or failure among elite sprinter. The Sport Psychologist, 17, 55–76.

Ó 2016 Hogrefe Publishing

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting & Clinical Psychology, 65, 599–610. doi: http://dx.doi.org/10.1037/0022-006X.65.4.599 Ilgen, D. R., Fisher, C. D., & Taylor, M. S. (1979). Consequences of individual feedback on behavior in organizations. Journal of Applied Psychology, 64, 349–371. doi: 10.1037/00219010.64.4.349 Jin, S. (2009). Modality effects in second life: The mediating role of social presence and the moderating role of product involvement. Cyber Psychology & Behavior, 12, 717–721. doi: 10.1089/ cpb.2008.0273 Jin, S. (2011). I feel present. Therefore I experience flow: A structural equation modeling approach to flow and presence in video games. Journal of Broadcasting & Electronic Media, 55, 114–136. doi: 10.1080/08838151.2011.546248 Jin, S., & Park, N. (2009). Parasocial interaction with my avatar: Effects of interdependent self-construal and the mediating role of self-presence in an avatar-based console game, Wii. Cyber Psychology & Behavior, 12, 723–727. doi: 10.1089/cpb.2008.0289 Kim, S. Y., Prestopnik, N., & Biocca, F. A. (2014). Body in the interactive game: How interface embodiment affects physical activity and health behavior change. Computers in Human Behavior, 36, 376–384. doi: 10.1016/j.chb.2014.03.067 Klimmt, C., & Hartmann, T. (2006). Effectance, self-efficacy, and the motivation to play video games. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 133–145). Mahwah, NJ: Lawrence Erlbaum. Kroll, T., Kehn, M., & Ho, P. (2007). The SCI exercise self-efficacy scale (ESES): Development and psychometric properties. International Journal of Behavioral Nutrition and Physical Activity, 4. doi: 10.1186/1479-5868-4-34 Latham, G. P., & Locke, E. A. (1991). Self-regulation through goal setting. Organizational Behavior and Human Decision Processes, 50, 212–247. Lee, K. (2004a). Presence, explicated. Communication Theory, 14, 27–50. doi: 10.1111/j.1468-2885.2004.tb00302.x Lee, K. (2004b). Why presence occurs: Evolutionary psychology, media equation, and presence. Presence: Teleoperators and Virtual Environments, 13, 494–505. doi: 10.1162/ 1054746041944830 Lee, K., & Nass, C. (2004). The multiple source effect and synthesized speech: Doubly-disembodied language as a conceptual framework. Human Communication Research, 30, 182–207. doi: 10.1111/j.1468-2958.2004.tb00730.x Lee, K. M., Park, N., & Song, H. (2005). Can a robot be perceived as a developing creature?: Effects of a robot’s long-term cognitive developments on its social presence and people’s social responses toward it. Human Communication Research, 31, 538–563. doi: 10.1111/j.1468-2958.2005.tb00882.x Lee, K. M., Peng, W., Yan, C., & Jin, S. (2006). Can robots manifest personality?: An empirical test of personality recognition, social responses, and social presence in human-robot interaction. Journal of Communication, 56, 754–772. doi: 10.1111/j.14602466.2006.00318.x Lieberman, D. A. (1997). Interactive video games for health promotion: Effects on knowledge, self efficacy, social support, and health. In R. L. Street, W. R. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 103–120). Mahwah, NJ: Lawrence Erlbaum. Limperos, A. M. (2014). Are “Wii” exercising correctly? Understanding how exergames can be used to increase knowledge of exercise behavior. Games for Health Journal, 3, 25–30. doi: 10.1089/g4h.2013.0066

Ó 2016 Hogrefe Publishing

Limperos, A. M., Downs, E., Ivory, J. D., & Bowman, N. D. (2013). Leveling up: A review of emerging trends and suggestions for the next generation of communication research investigating video games’ effects. In E. Cohen (Ed.), Communication yearbook 37 (pp. 349–377). New York, NY: Routledge. Limperos, A. M., & Oliver, M. B. (2012, May). Assessing the viability of mediated exercise technologies in motivating future exercise intentions. Paper presented at the International Communication Association, Phoenix, AZ. Lin, J.-H. (2015). “Just dance”: The effects of exergame feedback and controller use on physical activity and psychological outcomes. Games for Health Journal, 4, 183–189. doi: 10.1089/g4h.2014.0092 Lombard, M., & Ditton, T. B. (1997). At the heart of it all: The concept of presence. Journal of Computer Mediated Communication, 3(2). doi: 10.1111/j.1083-6101.1997.tb00072.x Lombard, M., & Ditton, T. B. (2000). Measuring presence: A literature-based approach to the development of a standardized paper and pencil instrument. Paper presented at The Third International Workshop on Presence, Philadelphia, PA. Lwin, M. O., & Malik, S. (2011, May). Can exergames impart health messages? Game play, framing and drivers of physical activity among children. Paper presented at the Conference of International Communication Association, Boston, MA. Lyons, E. J. (2015). Cultivating engagement and enjoyment in exergames using feedback, challenge, and rewards. Games for Health Journal, 4, 12–18. doi: 10.1089/g4h.2014.0072 Lyons, E. J., & Hatkevich, C. (2013). Prevalence of behavior changing strategies in fitness video games: Theory-based content analysis. Journal of Medical Internet research, 15(5). doi: 10.2196/jmir.2403 Lyons, E. J., Tate, D. F., Ward, D. S., Ribisl, K. M., Bowling, J. M., & Kalyanaraman, S. (2014). Engagement, enjoyment, and energy expenditure during active video game play. Health Psychology, 33(2), 174–181. doi: 10.1037/a0031947 McAuley, E., Blissmer, B., Katula, J., & Duncan, T. E. (2000). Exercise environment, self-efficacy, and affective responses to acute exercise in older adults. Psychology and Health, 15, 341–355. doi: 10.1080/08870440008401997 Mhurchu, C. N., Maddison, R., Jiang, Y., Jull, A., Prapavessis, H., & Rodgers, A. (2008). Couch potatoes to jumping beans: A pilot study of the effect of active video games on physical activity in children. International Journal of Behavioral Nutrition and Physical Activity, 5, 8. doi: 10.1186/1479-5868-5-8 Midden, C., & Ham, J. (2008). The persuasive effects of positive and negative social feedback from an embodied agent on energy conversation behavior. Proceedings of the Artificial Intelligence and Simulation of Behavior Convention, University of Aberdeen, UK. Peña, J., & Kim, E. (2014). Increasing exergame physical activity through self and opponent avatar appearance. Computers in Human Behavior, 41, 262–267. doi: 10.1016/j.chb.2014.09.038 Peng, W. (2009). Design and evaluation of a computer game to promote a healthy diet for young adults. Health Communication, 24, 115–127. doi: 10.1080/10410230802676490 Peng, W., Lin, J., & Crouse, J. (2011). Is playing exergames really exercising? A meta-analysis of energy expenditure in active video games. Cyberpsychology, Behavior, and Social Networking, 14, 68–1699. doi: 10.1089/cyber.2010.0578 Peng, W., Lin, J.-H., Pfeiffer, K. A., & Winn, B. (2012). Need satisfaction supportive game features as motivational determinants: An experimental study of a self-determination theory guided exergame. Media Psychology, 15, 175–196. doi: 10.1080/15213269.2012.673850 Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models.

Journal of Media Psychology (2018), 30(1), 29–40

J. Kim & C. E. Timmerman, Social Presence and Feedback in Exergames

Behavior Research Methods, Instruments, and Computers, 36, 717–731. doi: 0.3758/BF03206553 Reeves, B., & Nass, C. (1996). The media equation: How people treat computers television, and new media like real people and places. New York, NY: Cambridge University Press. Ritterfeld, U., & Weber, R. (2006). Video games for entertainment and education. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives, responses, and consequences (pp. 399–413). Mahwah, NJ: Erlbaum. Rodgers, W. M., Hall, C. R., Blanchard, C. M., McAuley, E., & Munroe, K. J. (2002). Task and scheduling self-efficacy as predictors of exercise behavior. Psychology & Health, 17, 405–416. doi: 10.1080/0887044022000004902 Rovniak, L. S., Anderson, E. S., Winett, R. A., & Stephens, R. S. (2002). Social cognitive determinants of physical activity in young adults: A prospective structural equation analysis. Annals of Behavioral Medicine, 24, 149–156. doi: 10.1207/ S15324796ABM2402_12 Schul, Y., & Ganzach, Y. (1995). The effects of accessibility of standards and decision framing on product evaluations. Journal of Consumer Psychology, 4, 61–83. doi: 10.1207/ s15327663jcp0401_03 Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommunication. London, UK: Wiley. Song, H., Kim, J., Kwon, J., & Jung, Y. (2013). Anti-smoking educational game using avatars as visualized possible selves. Computers in Human Behavior, 29, 2029–2036. doi: 10.1016/j. chb.2013.04.008 Song, H., Kim, J., & Lee, K. M. (2014). Virtual body vs. real body in exergames: Reducing social physique anxiety in exercise experiences. Computers in Human Behavior, 36, 282–285. doi: 10.1016/j.chb.2014.03.059 Song, H., Kim, J., Tenzek, K. E., & Lee, K. (2013). The effects of competition and competitiveness upon intrinsic motivation in exergames. Computers in Human Behavior, 29, 1702–1708. doi: 10.1016/j.chb.2013.01.042 Song, H., Peng, W., & Lee, K. M. (2011). Promoting exercise selfefficacy with an exergame. Journal of Health Communication, 16, 148–162. doi: 10.1080/10810730.2010.535107 Spence, P. R., Westerman, D., Edwards, C., & Edwards, A. (2014). Welcoming our robot overlords: Initial expectations about interaction with a robot. Communication Research Reports, 31(3), 272–280. doi: 10.1080/08824096.2014.924337 van de Ridder, J. M., Peters, C. M., Stokking, K. M., de Ru, J. A., & ten Cate, O. T. J. (2015). Framing of feedback impacts student’s satisfaction, self-efficacy and performance. Advances in Health Sciences Education, 20, 1–14. doi: 10.1007/s10459014-9567-8

Journal of Media Psychology (2018), 30(1), 29–40

Viciana, J., & Cervello, E. M. (2007). Effect of manipulating positive and negative feedback on goal orientations, perceived motivational climate, satisfaction, task choice, perception of ability, and attitude toward physical education lessons. Perceptual and Motor Skills, 105, 67–82. doi: 10.2466/pms.105.1.67-82 Vorderer, P., Klimmt, C., & Ritterfeld, U. (2004). Enjoyment: At the heart of media entertainment. Communication Theory, 14, 388–408. doi: 10.1111/j.1468-2885.2004.tb00321.x Received May 21, 2015 Revision received September 12, 2015 Accepted September 16, 2015 Published online May 20, 2016 Jihyun Kim School of Communication Studies Kent State University Kent, OH 44242 USA Tel. +1 (330) 672-0282 Fax +1 (330) 672-3510 E-mail jihyunkim218@gmail.com

Jihyun Kim (PhD, University of WisconsinMilwaukee, 2012) is Assistant Professor in the School of Communication Studies at the Kent State University, OH. Her primary research interests are focused on effects and implications of new media/communication technologies and a theoretical notion of social presence in mediatedcommunication contexts. Additionally, her research also examines various issues in health communication.

C. Erik Timmerman (PhD, University of Texas at Austin, 2000) is Associate Professor in the Department of Communication at the University of WisconsinMilwaukee, WI. His research and teaching interests examine communication processes in organizations and applications of technology to instructional settings.

Ó 2016 Hogrefe Publishing

The first structured resource for psychologists that combines mindfulness with character strenghts Ryan M. Niemiec

Mindfulness and Character Strengths A Practical Guide to Flourishing

2014, xx + 274 pp. + CD with meditation exercises US $39.80 / € 27.95 ISBN 978-0-88937-376-1 Also available as eBook At the core of this hands-on resource for psychologists and other practitioners, including educators, coaches, and consultants, is MindfulnessBased Strengths Practice (MBSP), the first structured program to combine mindfulness with the character strengths laid out in the VIA Institute’s classification developed by Drs. Martin E. P. Seligman and Christopher Peterson. This 8-session program systematically boosts awareness and application of character strengths – and so helps people flourish and lead more fulfilling lives. The author’s vast experience working with both mindfulness and character strengths is revealed in his sensitive and clear presentation of the conceptual, practical, and scientific elements of this unique combined approach. It is not only those who are new to mindfulness or to character strengths who will appreciate the detailed primers on these topics in the first

www.hogrefe.com

section of the book. And the deep discussions about the integration of mindfulness and character strengths in the second section will benefit not just intermediate and advanced practitioners. The third section then leads readers step-by-step through each of the 8 MBSP sessions, including details of session structure and content, suggested homework, 30 practical handouts, as well as inspiring quotes and stories and useful practitioner tips. An additional chapter discusses the adaption of MBSP to different settings and populations (e.g., business, education, individuals, couples). The mindfulness and character strengths meditations on the accompanying CD support growth and development. This highly accessible book, while primarily conceived for psychologists, educators, coaches, and consultants, is suitable for anyone who is interested in living a flourishing life.

Clear and compact guidance on integrating mindfulness into practice “Offering conceptual clarity, depth, and analysis, this book skillfully and clearly guides the reader through the practice and implementation of effective mindfulness-based interventions.” Zindel V. Segal, PhD, Distinguished Professor of Psychology in Mood Disorders, Department of Psychology, University of Toronto, ON, Canada; Author of The Mindful Way Workbook

Katie Witkiewitz / Corey R. Roos / Dana Dharmakaya Colgan / Sarah Bowen

Mindfulness (Series: Advances in Psychotherapy – Evidence-Based Practice – Volume 37) 2017, viii + 80 pp. US $ 29.80 / € 24.95 ISBN 978-0-88937-414-0 Also available as eBook This clear and concise book provides practical, evidence-based guidance on the use of mindfulness in treatment: its mechanism of action, the disorders for which there is empirical evidence of efficacy, mindfulness practices and techniques, and how to integrate them into clinical practice. Leading experts describe the concepts and roots of mindfulness, and examine the science that has led to this extraordinarily rich and ancient practice becoming a foundation to many contemporary, evidenced-based approaches in psychotherapy.

www.hogrefe.com

The efficacy of mindfulness-based interventions in conditions as diverse as borderline personality disorder, post-traumatic stress disorder, depression, alcohol and substance use, emotional dysregulation, attentiondeficit hyperactivity disorder, chronic stress, eating disorders, and other medical conditions including type 2 diabetes and rheumatoid arthritis is also described. The book is invaluable reading for all those curious about the current science around mindfulness and about how and when to incorporate it effectively into clinical practice.

Original Article

Commitment to the Team Perceived Conﬂict and Political Polarization Bryan McLaughlin Texas Tech University, College of Media and Communication, Lubbock, TX, USA

Abstract: Scholars have increasingly employed social identity theory to explain how and why political polarization occurs. This study aims to build off of this work by proposing that perception of intergroup conflict serves as a mechanism that mediates the effect of news media coverage on political polarization. Specifically, I argue that the news media’s emphasis on political animosity can cultivate partisans’ perception that the parties are in conflict, which provides a context that makes partisan identity salient and, ultimately, leads to higher levels of affective and ideological polarization. This hypothesis is tested with an experiment using an American national sample of Democrats and Republicans (N = 300). Participants read a news story in which the public believes the parties are in a state of either high or low conflict (or they did not receive a news story). Using mediation analysis, the results of the study provide evidence that news media coverage of political conflict leads to increased perception of intergroup conflict, which then leads to higher levels of (a) partisan identification, (b) affective polarization, and (c) ideological polarization. Keywords: political polarization, social identity theory, political conflict, impersonal influence

Political polarization is the phenomenon where political partisans become more ideologically distinct across groups while becoming more ideologically similar within groups (McCarty, Poole, & Rosenthal, 2006). While political polarization research has historically focused on policy divergence, an alternative framework views polarization from the perspective of social identity theory (SIT) (Iyengar, Sood, & Lelkes, 2012; Lupu, 2013; Nicholson, 2012). SIT argues that humans are driven to act in the interest of the in-group in opposition to a relevant out-group (Tajfel & Turner, 1979). The related self-categorization theory (SCT) proposes that group polarization arises from the cognitive need to minimize differences with in-group members will maximizing differences with the out-group (Turner et al., 1987). Holding political attitudes that are distinct from the out-group can help achieve these goals. Thus, it may be that partisan attitudes in America have polarized so dramatically in recent years (Abromowitz, 2011) because those who identify with a political party have a psychological impetus to distinguish themselves from the opposing party. A key tenet of self-categorization theory is that the salience of a social identity depends upon the specific social context (Turner et al., 1987). Yet, relatively little work has considered the contextual factors that may be driving partisans’ increasing psychological attachment to their political parties. This paper proposes that the perception of intergroup conflict is one factor that plays a key mediating role in exacerbating political polarization. Although conflict is not Ó 2016 Hogrefe Publishing

necessary for a social identity to be salient, the context of intergroup competition creates an increased sense of urgency to commit to the team (Brewer, 2001; Mackie, 1986; Veenstra & Haslam, 2000). It therefore stands to reason that the greater the belief that the political parties are in conflict, the more partisans will behave in a polarized manner. The perception of intergroup conflict represents a subjective assessment based on information gleaned about the social world. The mass media play a fundamental role in shaping social perceptions (Joyce & Harwood, 2014), especially because many peoples’ exposure to other social groups are based largely on vicarious interactions facilitated by the media (Harwood & Vincze, 2012; Mastro, 2015). When it comes to reporting on the political parties, the mainstream news media primarily focuses on electoral competition, contentious policy battles, and general animosity between the parties (Mutz & Reeves, 2005). Thus, even though most citizens typically have limited direct experience with political confrontation (Mutz, 2006), partisans can still receive cues from the news media that heighten their belief that there is an irreconcilable divide between the parties. This study proposes that media portrayals of political conflict can exacerbate the perception that the parties are in conflict. This provides a context that makes partisan identity salient, causing partisans to perceive in-group attitudes to be more extreme and, ultimately, move toward

Journal of Media Psychology (2018), 30(1), 41–51 DOI: 10.1027/1864-1105/a000176

that more polarized position. In order to examine this expectation, I employ an experimental design in which a national sample of Democratic and Republican partisans are presented with a news story in which the public believes the parties to be in a state of either high or low conflict. In doing so, this study examines if perceived conflict acts as a mediating variable that can lead to higher levels of (a) partisan identification, (b) affective polarization, and (c) ideological polarization.

Social Identity and Group Polarization SIT (Tajfel & Turner, 1979) was positioned as a response to “realistic group conflict theory” (Sherif, 1966), which argued that objective conflicts between groups lead to discriminatory behavior. Tajfel and Turner (1979) argued that although actual conflict and/or competition over resources is often sufficient to promote intergroup hostility, it is hardly necessary. In a series of experiments, Tajfel and Turner demonstrated that the simple categorization of belonging to one (trivial) group rather than another results in in-group favoritism and out-group discrimination (e.g., Tajfel, 1970; Tajfel, Billig, Bundy, & Flament, 1971). According to this “minimal group paradigm,” once individuals perceive themselves as belonging to a social group, they will use that social category to define their selfconcept, which, subsequently, influences how they navigate the social world. The theoretical explanation for SIT is that humans are intrinsically driven to hold a positive selfevaluation; and because self-worth is derived, in part, from the groups that help define a person’s self-concept, people are motivated to favorably distinguish the in-group from relevant out-groups (Tajfel & Turner, 1979). Turner and colleagues built upon SIT by developing SCT (Turner & Oakes, 1986; Turner et al., 1987). SCT focused on understanding the process through which individuals shifted from interpersonal to intergroup behavior. SCT conceptualizes the self-concept as containing multiple levels of abstraction, starting from individual identity, moving to classification as a member of a social group, and, finally, classification as a human being (Turner & Oakes, 1986). The use of a social identity for self-definition (its salience) depends upon an interaction between an individual and a specific social context (Oakes & Turner, 1986). Group membership is used to self-categorize when that identity is cognitively accessible and relevant to the particular context (Turner et al., 1987). Once a social identity becomes salient, individuals will seek to minimize in-group differences while maximizing differences from relevant out-groups. During this process, individuals place less emphasis on thinking of themselves as individuals and instead conceptualize themselves as similar to a prototypical representative of the in-group Journal of Media Psychology (2018), 30(1), 41–51

B. McLaughlin, Commitment to the Team

(referred to as depersonalization). The in-group prototype is a mental image of the typical group member – one who holds attitudes, behaviors, and attributes that would reflect the middle position in the range of the group’s distributions on those criteria. When a social identity is salient, people tend to move toward this in-group prototype (Turner et al., 1987). This provides a theoretical explanation for group polarization. When people compare their in-group with a relevant out-group, they become motivated to perceive the in-group characteristics as more distinct from the out-group, resulting in a more extreme in-group prototype. Polarization results when an individual perceives the group norms to be more extreme and then subsequently moves their own self-concept (and thus, attitudes and behaviors) toward those more extreme characteristics (Abrams, Wetherell, Cochrane, Hogg, & Turner, 1990; Hogg, Turner, & Davidson, 1990; Mackie, 1986; Turner & Oakes, 1986).

Perceived Conflict and the Importance of Context Identity salience is best thought of as ranging from fully individualized to fully group oriented, almost always falling somewhere in-between. People rarely, if ever, become completely depersonalized (Oakes, 2002). Many researchers have operated under the assumption that simply exposing partisans to a stimulus in which a relevant social identity is present will fully activate intergroup behavior (see Huddy, 2001). SIT and SCT do not, however, view external categorization as the determining factor in intergroup behavior. Rather, social categories are meaningless unless an individual internalizes those identity categories (Oakes, 2002). Salience is not simply the presence of social identity cues; a social identity only becomes a salient component of an individual’s self-categorization when the fit and accessibility of a social group lead an individual to internalize that identity category (Hogg & Turner, 1985; Oakes & Turner, 1986; Turner et al., 1987). Thus, context plays a central role in determining the degree to which a partisan moves toward depersonalization. Intergroup conflict, while not necessary for the creation of social identity salience, should provide a context that compels individuals to internalize their social identity and adopt a more extreme group position. While SIT was initially intended to explain how discrimination can occur in the absence of objective conflict, this framework does not preclude the role conflict (real or perceived) can have on fostering antagonistic intergroup behavior. There is clear evidence that in the context of intergroup competition people hold higher levels of in-group commitment and are more antagonistic toward the out-group Ó 2016 Hogrefe Publishing

B. McLaughlin, Commitment to the Team

(Brewer, 1979; Tajfel & Turner, 1979). These conflicts do not have to be “objective,” they can also be subjective and symbolic (Brewer, 2001). Because individuals are motivated to maintain a positive in-group status, any threat from an out-group should heighten in-group solidarity (Tajfel &Turner, 1979; Veenstra & Haslam, 2000). Intergroup conflict, therefore, is likely to create a context in which individuals feel an urgency to internalize and act on the relevant social identity (Brewer, 2001; Veenstra & Haslam, 2000). While simply placing an individual in a situation in which an in-group and out-group are externally present might make that identity salient, the presence of conflict should make that salience become more prominent, moving the individual closer to the depersonalization extreme. For example, Mackie (1986) found that introducing competition to a stimulus featuring an in-group and an out-group caused participants to perceive their in-group norms to be more extreme, ultimately becoming more polarized themselves. Thus, when an individual perceives a potential threat from an out-group, they should be more likely to internalize their social identity, causing them to perceive the in-group norm as more polarized, then move themselves in line with that more extreme prototype.

The News Media and Impersonal Influence There is little doubt that the conflict between Democrats and Republicans in America has become particularly contentious in recent years (Gutmann & Thompson, 2012). But how do people know that this is the case? There are numerous sources that work together to create a dynamic environment in which an individual learns about the social world – through interpersonal interaction (face-to-face or online), from trusted leaders (e.g., politicians, teachers, religious leaders), or from mediated communication (e.g., television shows, mainstream new media). This study focuses on one important factor that is likely to exacerbate political polarization, the mainstream news media. Examining the influence of the news media on perceived conflict seems particularly relevant given that the media provide information about intergroup relationships that individuals might not otherwise have (Harwood & Vincze, 2012; Joyce & Harwood, 2014; Mastro, 2015). As a result, media exposure can have important effects on a range of intergroup processes and effects (Mastro & Atwell Seate, 2012). Along those lines, most Americans try to avoid engaging in political discussion with those with whom they disagree (although there are certainly exceptions; Green, Visser, & Tetlock, 2000; Mutz, 2006). People typically form social ties with likeminded individuals (McPherson, Smith-Loving, & Cook, 2001), creating social networks that tend to be politically homogeneous (Colleoni, Rozza, & Arvidsson, 2014). People do, of course, sometimes encounter those Ó 2016 Hogrefe Publishing

with opposing political opinions, but for most citizens these tend to be inadvertent encounters not originally initiated for political purposes (Brundidge, 2010; Mutz, 2006; Wojcieszak & Mutz, 2009). Therefore, although people sometimes experience political conflict, more often than not, this is a peripheral experience. The news media, on the other hand, are much more enamored with conflict, disagreement, and animosity (Jamieson & Cappella, 2008; Mutz & Reeves, 2005). By constantly broadcasting uncivil political campaigns and contentious policy battles, the news media are likely to cultivate the perception that the political parties are in intense intergroup competition. This mediated information about political conflict might include campaign rhetoric, policy debates and coverage of political gridlock, and public opinion polls showing the public’s dissatisfaction with the current political climate. Public opinion polls, in particular, can have a large effect on an individual’s attitudes. This process is what Diana Mutz (1998) refers to as impersonal influence, where people form perceptions about the attitudes, beliefs, and experiences of anonymous others through exposure to various mass communication channels. These impressions about social others, in turn, affect an individual’s own attitudes and behavior. This dovetails well with the SCT account about how people tend to move toward the normative opinions held by members of their in-group (Turner & Oakes, 1986).

Hypotheses When partisans are exposed to public opinion polls suggesting that impersonal others believe political conflict is a problem, they should move toward this mediated opinion, thereby elevating their perception that the political parties are in a state of intergroup conflict. I therefore predict: Hypothesis 1 (H1): When political partisans encounter news stories about political conflict, they will perceive a higher degree of conflict between the political parties. The context of political conflict should cause partisans to become concerned about potential threats to their in-group, leading them to feel a greater need to internalize their political identity. Thus, when citizens get cues from the news media about how polarized the country is, they should be more cognizant of political conflict in the United States, which should, subsequently, make their political identity more salient. I therefore predict: Hypothesis 2 (H2): Perceived conflict will mediate the effect of news coverage of political conflict on partisan identification. Journal of Media Psychology (2018), 30(1), 41–51

B. McLaughlin, Commitment to the Team

When partisan identity is salient, partisans strive to elevate the status of their party, while degrading the rival party. This includes affective polarization, or the process of increasingly viewing the in-group more favorably and preferring social distance between the parties (e.g., objecting to your child marrying someone from the other party; Iyengar et al., 2012). In conditions of intergroup conflict, in-group members should perceive that the prototypical member of the political in-group holds more extreme affective attitudes views toward the opposing party, prompting partisans to adopt those more extreme attitudes. Hypothesis 3 (H3): Perceived conflict will mediate the effect of news coverage of political conflict on affective polarization. Similarly, holding more divergent policy positions helps the political parties differentiate themselves from each other. As with H2, when partisan identity is salient, in-group members should perceive in-group policy attitudes to be more extreme. Perceiving this more extreme norm, partisans should then adopt the more extreme position themselves. I therefore predict: Hypothesis 4 (H4): Perceived conflict will mediate the effect of news coverage of political conflict on the extremity of an individual’s (a) political ideology, (b) economic attitudes, (c) social attitudes, and (d) immigration attitudes.

Method Participants A total of 300 participants were recruited through Survey Sampling International (SSI). SSI provides samples that are “diverse and as representative as possible of the target population.” SSI continuously monitors and calibrates their sampling techniques and their recruiting method includes multiple levels of randomization. SSI also uses quality controls to ensure that all sample populations are consistent and of high quality. Only citizens who identified as Republican or Democrat were eligible to participate in the study. A comparison of study participants with the national population of Republican and Democratic identifiers shows that my sample provided a good representation of Republican and Democratic identifiers (see Table 1).

Journal of Media Psychology (2018), 30(1), 41–51

Experimental Procedure This study employed a three-cell between-subjects design (high conflict vs. low conflict vs. control). After filling out the pretest items, participants were introduced to the stimulus described as a news article recently published in USA Today. Participants were randomly assigned to one of three conditions, after which they were asked about their perceptions of political conflict, levels of partisan identification, as well as affective and ideological polarization. Upon completion of the experiment, participants were debriefed and thanked for their time. In order to help ensure the quality of the data, five quality control measures were included to identify participants who were not sufficiently attentive to the study. Participants who failed two of these questions were removed from the study.

Stimulus Material The primary goal of this experiment was to provide media coverage showing that the American public sees the political parties as in a state of either high or low conflict. To achieve this goal, the stimulus consisted of a constructed news article participants were told “appeared in a recent issue of USA Today.” The focus of the article was on a recent Pew poll about the public’s perception of conflict between Republicans and Democrats. This constructed story was based on an actual news story (see Bass, 2013). In the stimulus, the poll and information were manipulated to either present the perception of political conflict between Republicansand Democrats as very high or lower than expected. The article started with either the headline “Partisan Conflict Tops All Other U.S. Conflicts,” or the headline, “Partisan Conflict No Worse Than Other U.S. Conflicts.” The control condition contained no news article. The high conflict article introduced a “Pew Research” poll that revealed, “the perceived levels of conflict between Republicans and Democrats is far greater than any other conflict in the United States.” The article highlights that 81% of people polled said “they believe there are strong conflicts between followers of the two main political parties.” The story then pointed out that these findings “reflect the feelings of animosity involved in the constant struggle between the two political parties.” The low conflict article started by introducing the poll as evidence that perceptions of conflict between Republicans and Democrats have “been largely exaggerated.” The poll is said to have shown “only 32 percent. . . believe there are strong conflicts between followers of the two main political parties.” As a result, partisan conflict is not seen as “any worse than other conflict.”

Ó 2016 Hogrefe Publishing

B. McLaughlin, Commitment to the Team

Table 1. Comparison of study sample with national population Demographic

National Democrats (%)

Study Democrats (%)

National Republicans (%)

Study Republicans (%)

Male

Female

18–29

30–49

50–64

65+

White

Black

Hispanic

Less than US $30,000

US $30,001–$75,000

US $75,001+

Gender

Age

Racial identity

Income

Note. n = 293.

Measures1 Perceived Conflict Perceived conflict was measured using three items. Participants were asked on a 7-point scale whether they agreed with the following statements (strongly disagree – strongly agree): “There is a strong conflict between Republicans and Democrats,” “the United States is clearly divided into liberals and conservatives,” and “there are irreconcilable differences between liberals and conservatives.” The three items were combined to construct Perceived Conflict (Cronbach’s α = .71, M = 5.52, SD = 1.04). Partisan Identification Partisan identification was measured by adapting Mael and Tetrick’s (1992) 10-item Identification with a Psychological Group scale (e.g., when someone criticizes the Democratic/ Republican Party, it feels like a personal insult). Democrats and Republicans were asked about their respective parties and these scores were merged to create Party Identification (Cronbach’s α = .81, M = 4.51, SD = .97). Affective Polarization Affective polarization was assessed using two measures – in-group favoritism and social distance. First, favorability was measured using a 100-point feeling thermometer. 1

In-group favoritism (M = 47.19, SD = 31.86) was then constructed by subtracting favorability scores of the out-group (M = 27.58, SD = 23.40) from the favorability score for the in-group (M = 75.54, SD = 21.38). Social distance was measured by asking partisans on a 7-point scale (very displeased – very pleased), “Suppose a son or daughter of yours was getting married. How would you feel if he or she married a supporter of (Democratic Party/Republican Party)?” (see Iyengar et al., 2012). Democrats’ feelings toward a child marrying a Republican were matched with Republican feelings toward a child marrying a Democrat. Scores were then reverse coded so that higher scores meant more social distance (M = 3.67, SD = 1.41). In-group favoritism and social distance were then standardized (z-scores) and combined to create Affective Polarization (Spearman–Brown = .58, M = 0.00, SD = 0.89). Political Ideology Participants were asked, “When it comes to politics, how would you describe each person or group?” Political ideology was assessed by where participants placed “yourself” on 7-point ideological continuum (very liberal – very conservative). Democratic scores were reversed coded then merged with Republican scores to create a measure that reflected how ideologically extreme individuals perceived themselves to be (M = 5.17, SD = 1.54).

Prior to finalizing the dependent variables, exploratory factor analysis was performed using principal axis factoring with promax rotation (an oblique rotation method) on eight dependent variables. Results showed three factors with eigenvalues greater than 1. The first factor included social distance (.765), in-group favorability (.739), and partisan identification (.646). The second factor included gay marriage extremity (.720) and abortion extremity (.606). Taxes extremity loaded on the second factor, but only at .368. The third factor included a single item – immigration extremity (.833). In light of these results, I combined social distance and in-group favoritism to create an affective polarization. Additionally, gay marriage extremity and abortion extremity were used to create a social extremity. Partisan identification was considered to be conceptually distinct from affective polarization and was maintained as a separate variable. Taxes and immigration extremity were also kept as distinct dependent variables.

Ó 2016 Hogrefe Publishing

Journal of Media Psychology (2018), 30(1), 41–51

B. McLaughlin, Commitment to the Team

Economic Extremity Participants were asked on a 7-point scale (strongly disagree – strongly agree) to what extent they agreed with the following statement: “In order to balance the federal budget, we should increase taxes on the rich.” Republicans’ scores were reverse coded and merged with Democratic scores. Therefore, as scores go up, partisans are displaying attitudes more closely aligned with the far end of their party’s position on the issue (M = 4.91, SD = 1.94). Social Extremity Social attitudes were assessed using two measures – abortion extremity and gay marriage extremity. Abortion extremity was measured by asking participants on a 7-point scale (strongly disagree – strongly agree) to what extent they agreed with the following statement: “By law, abortion should never be provided.” Democrats’ scores were reverse coded then merged with Republican scores (M = 4.89, SD = 2.12). Gay marriage extremity was measured by asking participants on a 7-point scale (strongly disagree – strongly agree) to what extent they agreed with the following statement: “Gay and lesbian couples should be allowed to get legally married.” Republicans’ scores on this item were reverse coded then merged with Democratic scores (M = 5.02, SD = 2.20). These measures were then combined to create Social Extremity (Spearman–Brown = .46, M = 4.86, SD = 1.86). Immigration Extremity Participants were asked on a 7-point scale (strongly disagree – strongly agree) to what extent that agreed with the following statement: “The federal government should not provide a path to citizenship for undocumented immigrants.” Democrats’ scores on this item were reverse coded then merged with Republican scores (M = 4.77, SD = 2.05). Controls Several demographic variables were used as controls in the analysis, including age, race (White = 1, non-White = 0), gender (male = 1, female = 0), education ((1 = some high school – 5 = postgraduate degree), and income (1 = less than US $15,000 – 8 = US $150,001 or more).2

Results Main Effects of Political Conflict Manipulation First, analysis of covariance (ANCOVA) was performed in order to examine the main effect of the news stories on 2

Figure 1. Main effects of news story on perceived conflict with 95% confidence interval error bars. n = 293. Mean scores were significantly higher in the high conflict condition compared with the low conflict condition and significantly higher in the Control condition compared with the low conflict condition (p < .05).

perceived conflict. Consistent with expectations, a significant main effect of the manipulation was found, F(2, 280) = 9.60, η2 = .06, p < .001. Specifically, those who read the high conflict condition (M = 5.81, SE = .10) perceived significantly more partisan conflict than those who read the low conflict condition (M = 5.18, SE = .10), p < .001. Additionally, participants in the control condition (M = 5.55, SE = .10) perceived significantly higher levels of conflict than those in the low conflict condition, p = .011. There was also a marginally significant difference between the high conflict and control conditions, p = .076 (see Figure 1). H1 was therefore supported. Additionally, using ANCOVA, I examined whether there was a main effect of the manipulation on the outcome variables. Results revealed that there was not a main effect of the manipulation on: Party Identification, F(2, 280) = 1.58, η2 = 0.01, p = .21, Affective Polarization, F(2, 278) = 0.68, η2 = 0.01, p = .51, Political Ideology, F(2, 278) = 0.11, η2 = 0.00, p = .90, or Social Extremity, F(2, 280) = 1.18, η2 = 0.01, p = .31. There was a significant main effect on Economic Extremity, F(2, 278) = 3.46, η2 = 0.02, p = .03, where participants held more extreme economic attitudes in the high conflict condition (M = 5.24, SE = .20) compared with the control condition (Mean = 4.51, SE = .20), p = .01. There was not, however, a significant difference between the high conflict and low conflict conditions (M = 4.96, SE = .20). Additionally, there was a marginally significant main effect of the manipulation on Immigration Extremity, F(2, 280) = 2.40, η2 = 0.02, p = .09. Specifically, participants in the high

The overall patterns of results were similar with or without the control variables. Nevertheless, the control variables were kept in the models in order to minimize any potential confounds.

Journal of Media Psychology (2018), 30(1), 41–51

Ó 2016 Hogrefe Publishing

B. McLaughlin, Commitment to the Team

Table 2. Results of mediation analysis b

95% CI

Direct effects Manipulation ? Perceived conflict

0.62

0.15***

Manipulation ? Party identification

0.11

0.14

Affective polarization

0.27

0.12*

Political ideology

0.13

0.22

Economic extremity

0.15

0.28

Social extremity

0.19

0.29

Immigration extremity

0.63

0.31*

0.35

0.07***

Perceived conflict ? Party identification Affective polarization

0.33

0.06***

Political ideology

0.41

0.11**

Economic extremity

0.22

0.13#

Social extremity

0.26

0.14#

Immigration extremity

0.03

0.15

Party identification

0.11

0.14

Affective polarization

0.06

0.13

Political ideology

0.12

0.22

Total effects

Economic extremity

0.29

0.27

Social extremity

0.02

0.28

Immigration extremity

0.61

0.29*

Party identification

0.22

0.06

0.12

0.37

Affective polarization

0.21

0.06

0.11

0.35

Political ideology

0.25

0.10

0.46

Economic extremity

0.14

0.09

0.00

0.35

Social extremity

0.16

0.10

0.01

0.40

Immigration extremity

0.02

0.09

0.20

0.15

Indirect effects Manipulation ? Perceived Conflict ?

Note. Coefficients are unstandardized. Separate analyses were run for each dependent variable. Sample sizes for each mediation model were: Partisan identification, n = 193; Affective polarization, n = 191; Political ideology, n = 192; Economic extremity, n = 192; Social extremity n = 193; Immigration extremity, n = 193. ***p < .001. **p < .01. *p < .05. #p < .10.

conflict condition (M = 3.60, SE = .21) had significantly more extreme attitudes about immigration policies than those in the low conflict condition had (M = 3.00, SE = .21), p = .04. There were no significant differences from the control condition, however, (M = 3.12, SE = .21).

Indirect Effects Indirect effects were tested using the SPSS Macro PROCESS (Hayes, 2012). Specifically, mediation models for each of the six dependent variables were estimated by examining the relationship between an X variable (conflict 3

story), a Y variable (dependent variables), and an M (mediating) variable (perceived conflict). Several controls were also included in the model (listed earlier). To create the X variable, a Conflict Dummy variable was created with the high conflict condition being scored 1 and the low conflict condition being scored 0.3 Indirect effects were examined using bootstrap analyses with 1,000 bootstrap samples and a 95% confidence interval (CI). Using this method, when an indirect path’s CI does not overlap zero (e.g., .3–.5) it is considered significant, and if it does overlap zero (e.g., .3–.5) it is considered nonsignificant. First, as can be seen in Table 2, there is a significant direct effect of the portrayal of public opinion about

The control condition was not used in any of the mediation models.

Ó 2016 Hogrefe Publishing

Journal of Media Psychology (2018), 30(1), 41–51

political conflict on perceived conflict, t = 4.25, b = .62, SE = .15, p < .001. Interestingly, when looking at the full mediation model, we now see a significant negative direct effect of the conflict manipulation on Affective Polarization and a significant positive direct effect on Immigration Extremity. Additionally, it is worth noting that the direct effects on Party Identification, Affective Polarization, Political Ideology, and Social Extremity are all negative, although only the path to Affective Polarization is significant. This suggests that, when accounting for the mediating role of perceived conflict, news stories about political conflict may slightly attenuate political polarization (see Discussion). Next, there are significant direct paths from perceived conflict to Party Identification, Affective Polarization, Political Ideology, and marginally significant paths to Social Extremity and Economic Extremity. There was not, however, a significant direct path to Immigration Extremity. This provides evidence that increased levels of perceived conflict lead to higher levels of partisan identity salience, affective polarization, and ideological polarization. When looking at the total effects, we see that only the path to Immigration Extremity is significant, while the paths to Party Identification, Affective Polarization, Political Ideology, Economic Extremity, and Social Extremity are all nonsignificant. Finally, consistent with expectations, perceived conflict mediated the effect of the conflict manipulation on Party Identification and Affective Polarization. Thus, I found support for H2 and H3. Additionally, there was an indirect effect of the conflict manipulation on Political Ideology, Social Extremity, and Economic Extremity. There was not, however, a significant indirect path to Immigration Extremity. I therefore found support for H4a and partial support for H4b. The relative lack of significant total effects, but strong evidence for indirect effects, suggests that there may, perhaps, be another mediating path, not accounted for here, that is counter-balancing the effects of perceived conflict on political polarization (see Discussion).

Discussion Scholars have increasingly applied a social identity framework to help explain the proliferation of political polarization in the United States (e.g., Iyengar et al., 2012; Lupu, 2013; Nicholson, 2012). According to SIT (Tajfel & Turner, 1979) and self-categorization theory (Turner et al., 1987), when partisan identity is salient individuals should be motivated to positively differentiate the political in-group from the opposing party. This study sought to build off of these insights by proposing that perceived conflict provides a Journal of Media Psychology (2018), 30(1), 41–51

B. McLaughlin, Commitment to the Team

context that should heighten partisan identity salience, subsequently causing partisans to perceive the in-group norms as more extreme, and, ultimately, leading them to move toward those more polarized political attitudes. Specifically, I argue that the news media’s incessant coverage of political gridlock and the public’s dissatisfaction with political polarization plays a central role in fostering the perception that the political parties are in conflict, which, as a consequence, leads to increased ideological and affective polarization. Results from this experimental study provide evidence that (a) news stories about political conflict increase the perception of political conflict, which, in turn, (b) mediates partisan identification, affective polarization, and ideological polarization. First, the results of this study demonstrate that news coverage of public opinion polling can cause partisans to adjust their perceptions about political conflict in the direction of the popular consensus. This highlights the process of impersonal influence, where individuals rely on cues from others about the current political climate in the United States. When partisans read a news story in which the public thinks the parties are in a state of high conflict they subsequently hold a higher perception that the parties are in conflict. This is consistent with SCT, which argues that when individuals perceive the in-group norm to be more extreme, they will move their own opinions toward that position. Therefore, when public opinion polls make political conflict salient, partisans are more likely to perceive themselves to be at war with the other side. Second, I find support for the claim that perceived intergroup conflict plays a mediating role in activating partisan identity salience, as there was a significant indirect path from the manipulation, through perceived conflict, on partisan identification. Third, the findings also provide evidence that perceived conflict mediates the effect of news media on affective polarization, as there was a significant indirect path from the manipulation through perceived conflict on affective polarization. Fourth, the results largely provide support for the prediction that political conflict leads to ideological polarization, as there was an indirect path from the conflict manipulation on political ideology, social extremity, and economic extremity. There was not, however, a significant indirect effect on immigration extremity. Taken together, these results provide evidence that perceived intergroup conflict serves as an underlying mechanism that helps account for how the news media fuel polarization. Whereas much of the polarization literature has focused on selective exposure to partisan media, this process would reflect a more basic level at which the news media can exacerbate political polarization. By constantly conveying the message that the political parties are polarized, the news media may help construct the symbolic boundaries that subsequently drive polarized behavior. Ó 2016 Hogrefe Publishing

B. McLaughlin, Commitment to the Team

It is important to note that although this paper gives primacy to the role of the news media in fostering perceptions of political conflict, it would be a mistake to suggest that the perceived conflict only results from news media consumption. There are good reasons to believe that the news media play an important role in constructing perceptions of political conflict given that (a) the news media tend to overemphasize conflict and drama (Mutz & Reeves, 2005), (b) people typically try to avoid political confrontation and prefer to fraternize with people who share a similar worldview (McPherson et al., 2001), and (c) the news media play a significant role in shaping the perceptions about intergroup relationships (Mastro & Atwell Seate, 2012). But many people learn about political conflict from other places and/or actually engage in heated political debates. People may hear about the contentious political environment from peers, family members, or respected authority figures. Additionally, people may inadvertently find themselves in a political debate. For example, people may get into political discussions with those with whom they disagree at work (Mutz, 2006) or in online chat rooms/forums that are not meant to be political (Brundidge, 2010; Wojcieszak & Mutz, 2009). These inadvertent discussions can, and often do, turn contentious. The point, therefore, is not that the news media are the only thing that influences perceived conflict. The more important point of emphasis is that perceiving conflict should lead to group polarization. People live in dynamic environments in which many factors influence their subjective perceptions about the social world. Additionally, identities are constructed and embedded with meaning over the long haul (Huddy, 2001). Intergroup conflict is, in large part, constructed and reinforced symbolically. Regardless of how these symbolic divisions are formed, the perception of conflict should provide a crucial contextual cue that compels individuals to internalize relevant social identities. If the trends of increased political polarization are indeed being fueled by the perception that the parties are in a contentious battle, then this is a significant societal construction that warrants further attention. More work is needed to understand which factors contribute to the perception of intergroup conflict. This segues into a potential limitation of this study. It is noteworthy that there were limited direct effects of the manipulation on the polarization variables. One potential explanation is that the effect of the stimulus on perceived conflict was not actually driving the observed indirect effects, but that the direct effects of the manipulation on perceived conflict were occurring independently of preexisting co-variation between perceived conflict and the outcome variables. Because both perceived conflict and polarization were measured after exposure to the stimulus, it is not possible to confirm the entire causal chain of Ă&#x201C; 2016 Hogrefe Publishing

events proposed here. Ideally, this study would have included a pretest measure of perceived conflict in order to isolate the variance in perceived conflict caused by the manipulation, which would allow for a more direct test of the mediation process. Although we cannot be certain that the indirect effects are actually the result of mediation, these results are consistent with the theoretical account presented above and there is reason to trust in their veracity. Another possible explanation for the minimal direct effects of the manipulation on the outcome variables is that there was another mediator, not measured in this study, at play working to suppress polarization. Looking at the direct effects in the mediation models, there is indication that, when accounting for perceived conflict, the news story about political conflict had a slightly negative effect on polarization variables (although these results were largely nonsignificant). There is actually a good explanation for why there might be a slight dampening effect of political conflict news stories on polarization â&#x20AC;&#x201C; many Americans have an aversion to political conflict and are unhappy with the polarized political climate (Harbridge & Malhotra, 2011). Even as partisans continue to polarize, there is still widespread dissatisfaction with the partisan bickering, as most citizens perceive the American political system to be inefficient and unproductive (Gutmann & Thompson, 2012). Thus, public opinion polls have consistently demonstrated that most Americans are fed up with political gridlock (Newport, 2013) and would prefer that politicians reach bipartisan agreement (Harbridge & Malhotra, 2011). What might account for these seemingly contradictory trends of increased polarization, but also growing dissatisfaction with this polarization? One possibility is that this dissatisfaction is working on the more conscious level, where Americans mindfully evaluate the political system and come to the conclusion that they are not particular impressed with the results. Thus, when exposed to a news story about political conflict, many partisans may consciously reject polarization, thinking to themselves that things might be better with a little less animosity and a little more cooperation. At the same time, these stories about political conflict may be activating an unconscious social identity process, which causes partisans to act in a polarized manner, even as they consciously object to such behavior. Social identity categories can activate automatic cognitive processing and, ultimately, biased and discriminatory evaluations and behaviors. People are generally unaware that these cognitive processes are occurring, but they are incredibly powerful in determining how people process and respond to the social world. For example, similar patterns of consciously rejecting discrimination while implicitly employing negative stereotypes are frequently observed when it comes to Journal of Media Psychology (2018), 30(1), 41â&#x20AC;&#x201C;51

Whites’ attitudes about African Americans (Entman & Rojecki, 2000). Unfortunately, the current study lacked the measures necessary to test this hypothesis, but this provides an interesting avenue for future research. Another potential concern is that this study overemphasizes one order of events (perceived conflict leading to policy polarization), when, in fact, policy disagreements can often be the cause of political conflict to begin with. It would, therefore, be a mistake to view the results of this study as an endorsement of the idea that policy polarization can be explained by the context of political conflict alone. Rather, the point is to highlight the important role perception of conflict plays in a larger process in which partisans increasingly dislike and disagree with each other. Political polarization has captured the attention of scholars, media pundits, and American citizens. SIT and SCT offer a theoretical explanation of why partisans are motivated to hold more divergent policy positions and antagonistic views toward the opposing party. This paper builds off of these insights by highlighting a key factor that should heighten partisan identity salience – perceived conflict. Specifically, I have argued that media portrayal of political conflict provides a context that leads to an increased perception of intergroup conflict, which should compel partisans to display higher levels of affective and ideological polarization. Political conflict in the United States is showing no signs of slowing down; neither is the scholarly debate about the causes and consequences of political polarization. This study presents evidence that perceptions of intergroup conflict may play an important mediating role in political polarization, but more work is needed to unpack the various factors that may lead to perceived conflict, as well as to consider how competing forces may be simultaneously affecting polarization on the conscious and unconscious levels.

References Abramowitz, A. (2011). The disappearing center: Engaged citizens polarization, and American democracy. New Haven, CT: Yale University Press. Abrams, D., Wetherell, M., Cochrane, S., Hogg, M. A., & Turner, J. C. (1990). Knowing what to think by knowing who you are: Self-categorization and the nature of norm formation, conformity and group polarization. British Journal of Social Psychology, 29, 97–119. doi: 10.1111/j.2044-8309.1990.tb00892.x Bass, F. (2013, January). Party conflicts in D.C. outstrip all other U.S. conflicts Bloomberg News. Retrieved from http://www. businessweek.com/news/2013-01-10/party-conflicts-in-ddot-c-dot-outstrip-all-other-u-dot-s-dot-conflicts Brewer, M. B. (1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307–324. doi: 10.1037/0033-2909.86.2.307 Brewer, M. (2001). Ingroup identification and intergroup conflict: When does ingroup love become outgroup hate? In R. Ahsmore, L. Jussim, & D. Wilder (Eds.), Social identity, intergroup conflict,

Journal of Media Psychology (2018), 30(1), 41–51

B. McLaughlin, Commitment to the Team

and conflict reduction (pp. 17–41). New York, NY: Oxford University Press.. Brundidge, J. (2010). Encountering “difference” in the contemporary public sphere: The contribution of the Internet to the heterogeneity of political discussion networks. Journal of Communication, 60, 680–700. doi: 10.1111/j.14602466.2010.01509.x Colleoni, E., & Rozza, A., Arvidsson. (2014). Echo chamber or public sphere? Predicting political orientation and measuring political homophily in twitter using big data. Journal of Communication, 64, 317–332. doi: 10.1111/jcom.12084 Entman, R., & Rojecki, A. (2000). The Black image in the White mind: Race and media in America. Chicago, IL: University of Chicago Press. Green, M. C., Visser, P. S., & Tetlock, P. E. (2000). Coping with accountability cross-pressures: Low-effort evasive tactics and high-effort quests for complex compromises. Personality and Social Psychology Bulletin, 26, 1380–1391. Gutmann, A., & Thompson, D. (2012). The spirit of compromise: Why governing demands it and campaigning undermines It. Princeton, NJ: Princeton University Press. Harbridge, L., & Malhotra, N. (2011). Electoral incentives and partisan conflict in Congress: Evidence from survey experiments. American Journal of Political Science, 55, 494–510. doi: 10.1111/j.1540-5907.2011.00517.x Harwood, J., & Vincze, L. (2012). Ethnolinguistic identity and television use in a minority language setting. Journal of Media Psychology, 24, 135–142. doi: .1027/1864-1105/a000071 Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable moderation, mediation, and conditional process modeling [White paper]. Retrieved from http://www. afhayes.com/public/ process2012.pdf Hogg, M. A., & Turner, J. C. (1985). Interpersonal attraction, social identification and psychological group formation. European Journal of Social Psychology, 15, 51–66. doi: 10.1002/ ejsp.2420150105 Hogg, M., Turner, J., & Davidson, B. (1990). Polarized norms and social frames of references: A test of the self-categorization theory of group polarization. Basic and Applied Social Psychology, 11, 77–100. doi: 10.1207/s15324834basp1101_6 Huddy, L. (2001). From social to political identity: A critical examination of social identity theory. Political Psychology, 22, 127–156. Iyengar, S., Sood, G., & Lelkes, Y. (2012). Affect, not ideology: A social identity perspective on polarization. Public Opinion Quarterly, 76, 405–431. doi: 10.1093/poq/nfs038 Jamieson, K. J., & Cappella, J. N. (2008). Echo chamber: Rush Limbaugh and the conservative media establishment. New York, NY: Oxford University Press. Joyce, N., & Harwood, J. (2014). Improving intergroup attitudes through televised vicarious intergroup contact: Social cognitive processing of ingroup and outgroup information. Communication Research, 41, 627–643. doi: 10.1177/ 0093650212447944 Lupu, N. (2013). Party brands and partisanship: Theory with evidence from a survey experiment in Argentina. American Journal of Political Science, 57, 49–64. doi: 10.1111/j.15405907.2012.00615.x Mackie, D. M. (1986). Social identification effects in group polarization. Journal of Personality and Social Psychology, 50, 720–728. doi: 10.1037/0022-3514.50.4.720 Mael, F., & Tetrick, L. (1992). Identifying organizational identification. Educational and Psychological Measurement, 54, 813–824. Mastro, D. (2015). Why the media’s role in issues of race and ethnicity should be in the spotlight. Journal of Social Issues, 71, 1–16. doi: 10.1111/josi.12093

Ó 2016 Hogrefe Publishing

B. McLaughlin, Commitment to the Team

Mastro, D. E., & Atwell Seate, A. (2012). Group membership in race-related media processes and effects. In H. Giles (Ed.), The handbook of intergroup communication (pp. 357–369). New York, NY: Routledge. McCarty, N., Poole, K., & Rosenthal, H. (2006). Polarized America: The dance of ideology and unequal riches. Cambridge, NY: The MIT Press. McPherson, M., Smith-Loving, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27, 415–444. doi: 10.1146/annurev.soc.27.1.415 Mutz, D. C. (1998). Impersonal influence: How perceptions of mass collectives affect political attitudes. New York, NY: Cambridge University Press. Mutz, D. C. (2006). Hearing the other side: Deliberative versus participatory democracy. Cambridge, NY: Cambridge University Press. Mutz, D., & Reeves, B. (2005). The new videomalaise: Effects of televised incivility on political trust. American Political Science Review, 99, 1–15. doi: 10.1017.S0003055405051452 Newport, F. (2013). Americans’ desire for government leaders to compromise increases. Retrieved from http://www.gallup.com/ poll/164570/americans-desire-gov-leaders-compromiseincreases.aspx Nicholson, S. (2012). Polarizing cues. American Journal of Political Science, 56, 52–66. doi: 10.1111/j.1540-5907.2011.00541.x Oakes, P. (2002). Psychological groups and political psychology: A response to Huddy’s ‘critical examination of social identity theory. Political Psychology, 23, 809–824. Oakes, P., & Turner, J. C. (1986). Distinctiveness and the salience of social category memberships: Is there an automatic perceptual bias towards novelty? European Journal of Social Psychology, 16, 325–344. doi: 10.1002/ejsp.2420160403 Sherif, M. (1966). In common predicament: Social psychology of intergroup conflict and cooperation. Boston, MA: Houghton Mifflin. Tajfel, H. (1970). Experiments in intergroup discrimination. Scientific American, 223, 96–102. Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social categorization and intergroup behavior. European Journal of Social Psychology, 1, 149–178. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA: Brooks/Cole.

Ó 2016 Hogrefe Publishing

Turner, J. C., & Oakes, P. J. (1986). The significance of the social identity concept for social psychology with reference to individualism, interactionism and social influence. British Journal of Social Psychology, 25, 237–252. doi: 10.1111/j.20448309.1986.tb00732.x Turner, J. C., Hogg, M., Oakes, P. J., Reicher, P., Reicher, S., & Wetherell, M. (1987). Rediscovering the social group: A selfcategorization theory. Cambridge, MA: Basil Blackwell. Veenstra, K., & Haslam, S. A. (2000). Willingness to participate in industrial protest: Exploring social identification in context. British Journal of Social Psychology, 39, 153–172. doi: 10/1348/ 014466600164390 Wojcieszak, M. E., & Mutz, D. C. (2009). Online groups and political discourse: Do online discussion spaces facilitate exposure to political disagreement? Journal of Communication, 59, 40–56. doi: 10/1111/j.1460-2466.2008.01403.x Received June 1, 2015 Revision received September 17, 2015 Accepted September 23, 2015 Published online May 20, 2016 Bryan McLaughlin Texas Tech University College of Media and Communication Box 43082 Lubbock, TX, 79409 USA Tel. +1 806 834-4873 Fax +1 806 742-1085 E-mail bryan.mclaughlin@ttu.edu

Bryan McLaughlin (PhD, University of Wisconsin, Madison) is an assistant professor in the Department of Advertising at Texas Tech University, Lubbock, TX. His research interests include political polarization, social identity theory, and political advertising.

Journal of Media Psychology (2018), 30(1), 41–51

Meeting Calendar April 5–7, 2018 The 4th World Conference on Media and Mass Communication (MEDCOM ’18) Bangkok, Thailand Contact: Mr. Saranga Meepitiya, http://mediaconference.co/ April 6–7, 2018 MEDIA AND COMMUNICATION ’18/II. International Conference on Media and Communication Studies Conference Istanbul, Turkey Contact: Ozgur Ozturk, https://www. dakamconferences.org/media-andcommunication April 17–19, 2018 6th Annual Conference “Comparative Media Studies in Today’s World” St. Petersburg, Russian Federation Contact: Svetlana Bodrunova, http:// eng.jf.spbu.ru/comparative_media_ studies/284.html

May 19–20, 2018 International Forum For Communication Media, Social Science and Education Research Barcelona, Spain Contact: Dr Carlos, http://cies. education/conferences/barcelona2018may-event/ May 24–28, 2018 68th ICA – International Communication Association Annual Conference Prague, Czech Republic Contact: International Communication Association, Washington, DC, https://www.icahdq.org/conf/

May 21–25, 2020 70th ICA – International Communication Association Annual Conference Gold Coast, Australia Contact: International Communication Association, Washington, DC, https://www.icahdq.org/conf/

July 9–10, 2018 The European Conference on Media, Communication and Film (EuroMedia2018) Brighton, United Kingdom Contact: Kiyoshi Mana, http:// euromedia.iafor.org

April 21, 2018 CHI 2018 Montréal, Canada Contact: https://chi2018.acm.org/

August 9–12, 2018 APA Annual Convention San Francisco, CA, USA Contact: American Psychological Association, http://www.apa.org/ convention/future.aspx

May 9–11, 2018 Annual Conference of the German Communication Association (DGPuK) Mannheim, Germany Contact: German Communication Association, DGPuK, http://www. dgpuk2018.de/

May 23–27, 2019 69th ICA – International Communication Association Annual Conference Washington, DC, USA Contact: International Communication Association, Washington, DC, https://www.icahdq.org/conf/

Journal of Media Psychology (2018), 30(1), 52 https://doi.org/10.1027/1864-1105/a000243

July 2–5, 2019 XVI European Congress of Psychology Moscow, Russia Contact: Russian Psychological Society, https://ecp2019.ru/

Ó 2018 Hogrefe Publishing

Instructions to Authors Journal of Media Psychology (JMP) is committed to publishing original, high-quality papers which cover the broad range of media psychological research. This peer-reviewed journal focuses on how human beings select, use, and experience various media as well as how media (use) can affect their cognitions, emotions, and behaviors. It is also open to research from neighboring disciplines as far as this work ties in with psychological concepts of the uses and effects of the media. In particular, it publishes multidisciplinary papers that reﬂect a broader theoretical and methodological spectrum and comparative work, e.g., cross-media, cross-gender, or crosscultural. As JMP is intended to foster Open Science Practices, authors are offered to publish their data and materials (i.e., stimuli and surveys) as Electronic Supplementary Material on the publisher’s website at http://econtent@hogrefe. com. In line with the Peer Reviewers’ Openness Initiative, authors may be asked by reviewers to share their data and materials at any stage of the reviewing process. Journal of Media Psychology publishes the following types of article: Original Articles, Theoretical Articles, Research Reports, Pre-Registered Reports.

Manuscript submission: All manuscripts should in the first instance be submitted electronically at http://www.editorialmanager.com/ jmp. Detailed instructions to authors are provided at https:// www.hogrefe.com/j/jmp Copyright Agreement: By submitting an article, the author confirms and guarantees on behalf of him-/herself and any coauthors that the manuscript has not been submitted or published elsewhere, and that he or she holds all copyright in and titles to the submitted contribution, including any figures, photographs, line drawings, plans, maps, sketches, and tables, and that the article and its contents do not infringe in any way on the rights of third parties. The author indemnifies and holds harmless the publisher from any third party claims. The author agrees, upon acceptance of the article for publication, to transfer to the publisher the exclusive right to reproduce and distribute the article and its contents, both physically and in nonphysical, electronic, or other form, in the journal to which it has been

Ó 2018 Hogrefe Publishing

submitted and in other independent publications, with no limitations on the number of copies or on the form or the extent of distribution. These rights are transferred for the duration of copyright as defined by international law. Furthermore, the author transfers to the publisher the following exclusive rights to the article and its contents: 1. The rights to produce advance copies, reprints, or offprints of the article, in full or in part, to undertake or allow translations into other languages, to distribute other forms or modified versions of the article, and to produce and distribute summaries or abstracts. 2. The rights to microfilm and microfiche editions or similar, to the use of the article and its contents in videotext, teletext, and similar systems, to recordings or reproduction using other media, digital or analog, including electronic, magnetic, and optical media, and in multimedia form, as well as for public broadcasting in radio, television, or other forms of broadcast. 3. The rights to store the article and its content in machinereadable or electronic form on all media (such as computer disks, compact disks, magnetic tape), to store the article and its contents in online databases belonging to the publisher or third parties for viewing or downloading by third parties, and to present or reproduce the article or its contents on visual display screens, monitors, and similar devices, either directly or via data transmission. 4. The rights to reproduce and distribute the article and its contents by all other means, including photomechanical and similar processes (such as photocopying or facsimile), and as part of so-called document delivery services. 5. The right to transfer any or all rights mentioned in this agreement, as well as rights retained by the relevant copyright clearing centers, including royalty rights to third parties. Online Rights for Journal Articles: Guidelines on authors’ rights to archive electronic versions of their manuscripts online are given in the ‘‘Guidelines on sharing and use of articles in Hogrefe journals’’ on the journal’s web page at https://www. hogrefe.com/j/jmp October 2016

Journal of Media Psychology (2018), 30(1)

Use movies to learn about positive psychology “This is the most important book about movies of our times.” Frank Farley, PhD, L. H. Carnell Professor, Temple University, Philadelphia, Former President of the American Psychological Association (APA)

Ryan M. Niemiec / Danny Wedding

Positive Psychology at the Movies

Using Films to Build Character Strengths and Well-Being 2nd edition 2014, xvi + 486 pp. US $59.00 / € 41.95 ISBN 978-0-88937-443-0 Also available as eBook Positive psychology is regarded as one of the most important developments in the field of psychology over the past century. This inspiring book uses movies as a medium for learning about the latest research and concepts, such as mindfulness, resilience, meaning, positive relationships, achievement, well-being, as well as the 24 character strengths laid out by the VIA Institute of Character. Films offer myriad examples of character strengths and other positive psychology concepts and are uniquely suited to learning about them and inspiring new ways of thinking. This book systematically discusses each of the 24 character strengths, balancing film discussion, related psychological research, and practical applications. Each chapter outlines Key Concepts, Relevant Research, an Exemplar from a key movie, Overuse/Underuse, Key

www.hogrefe.com

Enablers and Inhibitors, Practical Applications, International Cinema, and a Summary. Watching the films recommended in this book will help the reader to practice the skill of strengths-spotting in themselves and others, inspiring self-improvement. Practical resources include a suggested syllabus for a complete positive psychology course based on movies, a list of suitable movies for children, adolescents, and families as well as a list of questions for classroom and therapy discussions. Positive Psychology at the Movies is conceived for educators, students, practitioners, and researchers, but anyone who loves movies and wants to change their lives for the better will find it inspiring and relevant. Read this book to learn more about positive psychology – and watch these films to become a stronger person!

nline free o issue le samp

Journal of Individual Differences Editor-in-Chief Martin Voracek University of Vienna, Austria Associate Editors André Beauducel, Germany Philip J. Corr, UK Sam Gosling, USA Jürgen Hennig, Germany Philipp Y. Herzberg, Germany Aljoscha Neubauer, Austria

ISSN-Print 1614-0001 ISSN-Online 2151-2299 ISSN-L 1614-0001 4 issues per annum (= 1 volume)

Subscription rates (2018) Libraries / Institutions US $324.00 / € 249.00 Individuals US $159.00 / € 114.00 Postage / Handling US $16.00 / € 12.00

www.hogrefe.com

About the Journal Researchers, teachers, and students interested in all areas of individual differences (e.g., gender, temperament, personality, intelligence) and their assessment in human and animal research will find the Journal of Individual Differences useful. It publishes manuscripts dealing with individual differences in behavior, emotion, cognition, and their developmental aspects. This includes human as well as animal research. The Journal of Individual Differences is conceptualized to bring together researchers working in different areas ranging from, for example, molecular genetics to theories of complex behavior. Moreover, it places emphasis on papers dealing with special methodological and conceptual issues in basic science as well as in their applied fields (assessment of personality and intelligence).

Karl-Heinz Renner, Germany Willibald Ruch, Switzerland Astrid Schütz, Germany Andrzej Sekowski, Poland Jutta Stahl, Germany

Manuscript Submissions All manuscripts should be submitted online at www.editorialmanager.com/jindivdiff, where full instructions to authors are also available. Electronic Full Text The full text of the journal – current and past issues (from 1999 onward) – is available online at econtent.hogrefe.com/loi/jid (included in subscription price). A free sample issue is also available here. Abstracting Services The journal is abstracted / indexed in Current Contents / Social and Behavioral Sciences (CC / S&BS), Social Sciences Citation Index (SSCI), Scopus, PsycINFO, and PSYNDEX. Impact Factor (Journal Citation Reports®, Clarivate Analytics): 2016 = 1.123

Social Psychology

ISSN-Print 1864-9335 ISSN-Online 2151-2590 ISSN-L 1864-9335 6 issues per annum (= 1 volume)

Subscription rates (2018) Libraries / Institutions US $478.00 / € 374.00 Individuals US $223.00 / € 159.00 Postage / Handling US $24.00 / € 18.00

www.hogrefe.com

nline free o issue le samp

Editor-in-Chief Kai Epstude University of Groningen, The Netherlands

Editorial Office Wim Meerholz University of Groningen, The Netherlands

Associate Editors Anna Baumert, Germany Marco Brambilla, Italy Adam Fetterman, USA Ilka Gleibs, United Kingdom Michael Häfner, Germany

Hans J. IJzerman, The Netherlands Ulrich Kühnen, Germany Toon Kuppens, The Netherlands Ruth Mayo, Israel Kim Peters, Australia

About the Journal Social Psychology publishes innovative and methodologically sound research and serves as an international forum for scientific discussion and debate in the field of social psychology. Topics include all basic social psychological research themes, methodological advances in social psychology, as well as research in applied fields of social psychology. The journal focuses on original empirical contributions to social psychological research, but is open to theoretical articles, critical reviews, and replications of published research. The journal was published until volume 38 (2007) as the Zeitschrift für Sozialpsychologie (ISSN 0044-3514). Drawing on over 30 years of experience and tradition in publishing high-quality, innovative science as the Zeitschrift für Sozialpsychologie, Social Psychology has an internationally renowned team of editors and consulting editors from all areas of basic and applied social psychology, thus ensuring that the highest international standards are maintained.

Manuscript Submissions All manuscripts should be submitted online at www.editorialmanager.com/sopsy, where full instructions to authors are also available. Electronic Full Text The full text of the journal – current and past issues (from 1999 onward) – is available online at econtent.hogrefe.com/loi/zsp (included in subscription price). A free sample issue is also available there. Abstracting Services The journal is abstracted / indexed in Current Contents / Social and Behavioral Sciences (CC / S&BS), Social Sciences Citation Index (SSCI), PsycINFO, PSYNDEX, ERIH, Scopus, and EMCare. Impact Factor (Journal Citation Reports®, Clarivate Analytics): 2016 = 2.602

Start using strengths today! “The GO-TO book for building character.” Martin E. P. Seligman, the founder of positive psychology

Ryan M. Niemiec

Character Strengths Interventions A Field Guide for Practitioners 2018, xx + 300 pp. US $59.00 / € 46.95 ISBN 978-0-88937-492-8 Also available as eBook This book is the epitome of positive psychology: it takes the “backbone” of positive psychology – character strengths – and builds a substantive bridge between the science and practice. Working with clients’ (and our own) character strengths boosts well-being, fosters resilience, improves relationships, and creates strong, supportive cultures in our practices, classrooms, and organizations. This unique guide brings together the vast experience of the author with the science and the practice of positive psychology in such a way that both new and experienced practitioners will benefit. New practitioners will learn about the core concepts of character and signature strengths and how to fine-tune their approach and troubleshoot. Experienced practitioners will deepen their knowledge about advanced topics such as strengths overuse and collisions, hot button issues, morality, and integrating strengths with savoring,

www.hogrefe.com

flow, and mindfulness. Hands-on practitioner tips throughout the book provide valuable hints on how to take a truly strengths-based approach. The 24 summary sheets spotlighting each of the universal character strengths are an indispensable resource for client sessions, succinctly summarizing the core features of and research on each strength. 70 evidence-based step-by-step activity handouts can be given to clients to help them develop character strengths awareness and use, increase resilience, set and meet goals, develop positive relationships, and find meaning and engagement in their daily lives. No matter what kind of practitioner you are, this one-of-a-kind field guide is a goldmine in science-based applications. You’ll be able to immediately bring the science of well-being into action!

So that’s how my mind works – Now I get it!

“As a fan of PSI theory for more than 20 years, I am very happy seeing it translated for popular consumption! The book makes the theory reasonably simple, with lots of fun illustrations.” Kennon M. Sheldon, PhD, Professor of Psychological Sciences, University of Missouri, Columbia, MO

Johannes Storch / Corinne Morgenegg / Maja Storch / Julius Kuhl

Now I Get It!

Understand Yourself and Take Charge of Your Behavior 2018, vi + 248 pp. US $34.80 / € 27.95 ISBN 978-0-88937-541-3 Also available as eBook Using the example of four colleagues working together in a small company, Now I Get It! shows us the main personality types and their strengths and weakness in such a way that we gain real “now I get it!” insights into what is going on in our own and others’ subconscious. How does my mind work and what kind of personality do I have? When we can answer these questions and have come to terms with who we are, then the solutions to many issues that arise in everyday life will fall into place. What sort of people do I get on with best and how can I

www.hogrefe.com

best deal with the others? Are there recurring stressful situations in my professional or private life, and how do I resolve them? This humorously written and illustrated book, by the world’s leading experts in personality systems interaction (PSI) theory and the Zurich Resource Model (ZRM), gives us profound insights into our and other people’s subconscious thoughts – so we can adapt our own behavior and interactions to improve our quality of life. Cartoons and worksheets help us on our way.