Israel Journal of Psychiatry and Related Sciences by MEDIC

israel journal of

psychiatry

Vol. 51 - Number 1 2014

ISSN: 0333-7308

Special section: Metacognition in schizophrenia 4

Metacognition in Schizophrenia: Introduction to the Special Issue

Paul H. Lysaker and Ilanit Hasson-Ohayon

Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis Sophie Parker et al.

Mentalization-based Treatment for Psychosis Benjamin K. Brent et al.

Chronic Persecutory Delusion and Autobiographical Memories in Patients with Schizophrenia: A Diary Study Fabrice Berna et al.

Metacognition in Non-psychotic Help-seeking Adolescents Ravit Scheyer et al.

Metacognition in Schizophrenia and Schizotypy: Relation to Symptoms of Schizophrenia, Schizotypy and Quality of Life Shai Joseph Rabin et al.

Metacognition in Schizophrenia Spectrum Disorders Paul H. Lysaker et al.

Metacognition in Schizophrenia: A Concept Coming of Age Commentary to the Special Issue Martin Br端ne

Reliability and Cross-Validation of the Night Eating Questionnaire (NEQ): Hebrew Version Yael Latzer et al.

israel journal of

psychiatry

The Official Publication of the Israel Psychiatric Association Vol. 51 - Number 1 2014

and related sciences EDitor

David Greenberg DEPUTY EDITORS

Doron Gothelf Yoav Kohn Ivonne Mansbach Ora Nakash Shaul Lev-Ran David Roe Rael Strous Book reviews editor

Yoram Barak PAst Editor

Eli L. Edelstein Founding Editor

Heinz Z. Winnik Editorial Board

Alean Al-Krenawi Alan Apter Omer Bonne Elliot Gershon Talma Hendler Ehud Klein Ilana Kremer ltzhak Levav Yuval Melamed Shlomo Mendlovic Ronnen Segman Eliezer Witztum Gil Zalsman Zvi Zemishlany International Advisory Board

Paul Appelbaum Yoram Bilu Boris Birmaher Aaron Bodenheimer Stephen Deutsch Carl Eisdorfer Michael First Helen Herrman Julian Leff Ellen Liebenluft John Mann Phyllis Palgi Soumitra Pathare Daniel Pine Bruce Pollock Dan Stein Robert Wallerstein Myrna Weissman Associate editor

Rena Kurs

Assistant Editor

Joan Hooper

Marketing: MediaFarm Group +972-77-3219970 23 Zamenhoff st. Tel Aviv 64373, Israel amir@mediafarm.co.il www.mediafarm.co.il

Special section: Metacognition in schizophrenia Paul H. Lysaker and Ilanit Hasson-Ohayon

4 > Metacognition in Schizophrenia: Introduction to the Special Issue

Paul H. Lysaker and Ilanit Hasson-Ohayon

> Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis

Sophie Parker, Adrian Wells, CPsychol and Anthony P. Morrison

17 > Mentalization-based Treatment for Psychosis: Linking an Attachmentbased Model to the Psychotherapy for Impaired Mental State Understanding in People with Psychotic Disorders Benjamin K. Brent, Daphne J. Holt, Matcheri S. Keshavan, Larry J. Seidman and Peter Fonagy

Chronic Persecutory Delusion and Autobiographical Memories in Patients with Schizophrenia: A Diary Study >

Fabrice Berna, Caroline Huron, Mathilde Kazès, Isabelle Offerlin-Meyer, Dominique Willard, Paulina Verry, Guy Hédelin, Marie-Odile Krebs and Jean-Marie Danion

34 > Metacognition in Non-psychotic Help-seeking Adolescents: Associations with Prodromal Symptoms, Distress and Psychosocial Deterioration Ravit Scheyer, Noa Reznik, Merav Adres, Alan Apter, Larry J. Seidman and Danny Koren

> Metacognition in Schizophrenia and Schizotypy: Relation to Symptoms of Schizophrenia, Traits of Schizotypy and Social Quality of Life

Shai Joseph Rabin, Ilanit Hasson-Ohayon, Moran Avidan, Silvio Rozencwaig, Hadar Shalev and Shlomo Kravetz

54 > Metacognition in Schizophrenia Spectrum Disorders: Methods of Assessment and Associations with Neurocognition, Symptoms, Cognitive Style and Function

Paul H. Lysaker, Bethany L. Leonhardt, Marieke Pijnenborg, Rozanne van Donkersgoed, Steven de Jong and Giancarlo Dimaggio

> Metacognition in Schizophrenia: A Concept Coming of Age Commentary to the Special Issue of the Israel Journal of Psychiatry, edited by Paul H. Lysaker and Ilanit Hasson-Ohayon

Martin Brüne

> Reliability and Cross-Validation of the Night Eating Questionnaire (NEQ): Hebrew Version

Yael Latzer, Orna Tzischinsky, Michal Hason Rozenstein and Kelly Allison

> Book reviews Ricky Finzi-Dottan

> List of reviewers for the Israel Journal of Psychiatry 2013

Hebrew Section

> Abstracts

Whatever is handy (Ma she’ba layad) Danny Polk Danny is a Holocaust survivor, who, at the age of 10 was separated forever from his parents and brother on an Amsterdam train platform, and was later saved by a courageous Dutch farmer who hid him till the end of the war. Danny arranged for the farmer to be honored in his lifetime as a Righteous Gentile at Yad Vashem and still visits the farmer’s family regularly. “I have always collected bits and pieces, and these are parts of children’s games that I never played with. It is what I have been through, part of an extraordinary life.”

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Metacognition in Schizophrenia: Introduction to the Special Issue Abstract Deficits in metacognitive capacity in schizophrenia can be conceptualized as existing along a spectrum from more discrete to more synthetic activities. These capacities may be of great importance in schizophrenia research as a phenomenon which mediates and moderates the impact of illness related factors and general social factors upon outcome. In this introduction to the special issue on metacognition in schizophrenia we will discuss the evolution of the conceptualization of metacognition and its application to schizophrenia. We will focus on the implications of these conceptualizations for emerging models of psychopathology and treatment, and introduce the reader to core concepts to be explored in the articles composing this issue.

Schizophrenia is a complex form of mental illness with potentially devastating emotional, social and financial consequences for patients, their families and members of the larger community. Most current models suggest that interruption of psychosocial dysfunction experienced in schizophrenia is generally the result of an interaction of social and biological factors. Genetic vulnerabilities for schizophrenia have, for instance, been linked with abnormalities in brain development, which may be expressed as both positive and negative symptoms as well as neurocognitive deficits (1-8). Neurocognitive deficits and symptoms may then be linked to reductions in community and work functioning (9, 10). Taking a broader view, social and community factors such as trauma, stigma, poverty, isolation, and attachment patterns exacerbate or can be exacerbated by both brain function and neurocognition on their own, and in conjunction with more biologically based phenomena lead to dysfunction (11-14). One limitation of these current models is that they fail to consider that persons with schizophrenia are not merely passive subjects of social and biological factors. They actively interpret internal and external experiences 4

and respond to them. Thus, the way in which persons with schizophrenia make sense of their experience, in the midst of the influences of social and biological factors, affects their lives (15). They make meaning of events following recursive considerations of their own thoughts and feelings and the thoughts and feelings of others that are available for them. These processes by which meaning is made are also channeled by a personâ&#x20AC;&#x2122;s own history, desires, wishes and fears as they exist in the moment. As such, consistent with self-reports (16, 17), the impact of the social and biological forces which shape schizophrenia may be mediated and moderated by the manner in which the person makes sense of the dilemmas posed by those forces (18). One construct which describes some of the activities which would be involved in making meaning of the social and biological force implicated in schizophrenia is metacognition. Metacognition refers to the capacity to think about thinking. It is necessary for successfully making adaptive and complex sense of evolving social and psychological experience and deciding how to respond to them. Metacognitive capacity has been noted to be impaired in many with schizophrenia (19) and hence may play a role in the path from brain to behavior in schizophrenia as a mediator or moderator of the effects of symptoms, neurocognition and socially based challenges. While the study of metacognition in schizophrenia has important implications for the theoretical understanding of dysfunction in schizophrenia, it may also be of clinical relevance, pointing to a potential target for intervention. If impairments in metacognition, for instance, restrict the ability to make sense of the challenges experienced by persons with schizophrenia, then interventions which enhance metacognition could potentially play a role in assisting persons to more effectively work their way towards wellness (18). These challenges may be related to the illness (e.g., coping with symptoms) or not related to the illness (e.g., asking a friend for a favor). Interventions which enhance metacognition, for instance might assist persons not only to recognize needs for accommodation and potentially useful treatment options but also ways to live with the condition without accepting some of the negative connotations associated with it. In addition, integrating interventions that enhance metacognitive

Paul H. Lysaker and Ilanit Hasson-Ohayon

abilities within different psychotherapeutic approaches may facilitate the therapeutic process as it enables to explore the patient’s goals and desires from new perspectives (20). This exploration, supported by metacognitive abilities, involves a personal meaning reconstruction which is common among different approaches to treatment and psychotherapy (21). One challenge in studying metacognitive deficits in schizophrenia and consequent application to treatment concerns the exact definition of metacognition. The term itself has roots in many different traditions and many different forms of mental activities have been called metacognitive (22). It has been difficult, therefore, to synthesize different studies on metacognition and systematically move the field further ahead. In response to the diversity of definitions and conceptualizations of metacognition, this introduction to the special issue of the Israel Journal of Psychiatry on metacognition in schizophrenia will begin by offering a framework which could be used to understand how different kinds of studies of metacognition may relate to one another. Specifically we will explore how metacognition may be understood as a spectrum of activity which varies according to the extent of complexity of the experience that the metacognitive abilities refer to. This spectrum includes metacognitive activity that concerns a discrete experience vs. one in which many discrete experiences are brought together into a larger whole, and hence could be studied or addressed in treatment using different lenses. We will then introduce the work to be presented in this issue as it speaks to different points on this spectrum. It is hoped that this will position this issue as part of a movement to bring different kinds of metacognitive research together in order to have a more integrative and rich picture of the struggles of schizophrenia as well as the avenues for treatment that truly offer opportunities for the restoration of a meaningful life. Metacognition as a spectrum of activities: As noted above the construct of metacognition has emerged from a range of different traditions. It seems to have first become widely discussed in education literatures (23) and was commonly used to refer to awareness of what conditions best enabled learning (e.g., recognizing one was able to learn one subject better than another or that certain strategies were useful for learning certain kinds of material). The initial studies of metacognition allowed for an understanding of how thinking involves a set of skills that goes beyond the information available to learn. Tarricone (22) suggests that as this work grew,

the construct took on many different meanings through “its many facets and rich conceptual history” (p. 3). In the field of psychopathology metacognition has come to be used in a similar heterogeneous manner. One view of the cognitive tradition has stressed that metacognition involves understanding the kind of thoughts one seeks out which then influence coping strategies in the face of distress (24, 25). A second view though has examined metacognition less as a belief about a specific belief than as a larger process in which information is integrated into complex representations of the self and others that vary according to their effectiveness (26). This tradition stresses the representational nature of thinking, and is more rooted in developmental literature that focuses on how human beings form coherent ideas about themselves and others in close attachments with others (27). Thus, this literature focuses less on the choice of coping strategies and more with the development of the ability to understand that appearances are not reality, that individual people see the world differently and that people may change their views of the world over time (28). This idea is consistent with Dewey’s (29) suggestion to view self-knowledge as “an integration” of multiple facets and not as the sum of data. It is also in accord with a relational approach that emphasizes the dialectical interplay between multiplicity and singularity of the self (30). We suggest that one way to bring the different views on metacognition together is to conceptualize a spectrum which ranges from more discrete to more synthetic activities (31). In this spectrum highly discrete processes might include concern with immediate awareness or accuracy of judgments about one’s own experiences, whereas moderately discrete metacognitive activities by contrast involve less of a focus on a specific experience and more a discrete judgment about an aspect of oneself or another person. Synthetic metacognitive operations, on the other hand, involve the organization and reflect upon increasingly complex and coherent representations of self and others. These activities are not reducible to a matter of being correct or not or about specific perceptions in specific moments and involve the creation and reflection upon plausible and evolving ideas about self and others. In line with Dewey (29), this end of the metacognition spectrum concerns the interplay that occurs as persons perceive or misperceive thoughts and feelings in the moment (e.g., noticing whether an error was made) and then integrate them into something which is ultimately influenced by and linked to, but not the same as, the accuracy or focus of specific thoughts. 5

Metacognition in Schizophrenia: Introduction to the Special Issue

This themed issue: With this spectrum in mind, this special issue will present a range of papers concerned with both the understanding of psychopathology and the suggesting of treatments which involve metacognitive activity that can be located at different points along the metacognitive spectrum. Parker et al. (32, this issue) explore the extent to which changes in more discrete elements of metacognition, namely metacognitive beliefs related to worry, were related to outcome of cognitive therapy. Data are presented suggesting that over the course of cognitive therapy changes in beliefs persons have about beliefs related to worry are related to reduction in hallucinations among persons with schizophrenia. Thus, this work also points that the awareness of discrete beliefs can be a focus of psychotherapy. Continuing with the focus on psychotherapy, Brent et al. (33, this issue) explore how psychotherapy can address metacognition at a different point on the metacognitive spectrum. In contrast to the work focused on beliefs about beliefs, Brent describes how treatment can assist persons to form and reflect increasing complex ideas about themselves and so use enhanced synthetic metacognitive capacity to move towards recovery. Focused more on the links of psychopathology and schizophrenia, Berna et al. (34, this issue) explore how persons remember and think about paranoid beliefs experiences vs. experiences without any paranoid content. Analyzing memories recorded in a diary coupled with a recognition task they found that reflections about persecutory experiences were linked to greater anxiety. Compared to reflections about non-persecutory experiences, reflections about persecutory beliefs were less detailed and nuanced and more likely to be incorrectly remembered. Thus, a potentially unique link is suggested between deficits in discrete forms of metacognition and persecutory delusions. Also considering more discrete forms of metacognition, Scheyer et al. (35, this issue) studied awareness of performance on neuropsychological tests, among non-psychotic adolescents (and their parents) who sought help for a variety of emotional and behavioral difficulties in outpatient units. Awareness of errors was found to moderate the relationship between neurocognition and functional outcome before acute onset of the illness. Turning to the issue of more synthetic forms of metacognition, Rabin et al. (36, this issue) report data replicating links of the complexity of self-reflection with negative symptoms in outpatients with prolonged schizophrenia. Novel findings are also reported linking diminished synthetic metacognitive capacity with schizo6

typy in persons without psychosis. This work thus suggests a link between metacognition and the larger continuum of psychosis as it exists among persons in the general community. Finally, Lysaker et al. (37, this issue) present a summary of roughly eight years of research examining the correlates of metacognitive self-reflectivity and mastery with both symptoms and functional outcomes. This work, along with the piece by Rabin et al. (36, this issue) supports the possibility that deficits in the ability to form and use complex representations of the self and others in daily life may play a key role in determining not only symptom expression but also the way persons are able to live with and function despite the challenges of mental illness. References 1. Braff DL, Freedman R, Schork NJ, Gottesman II. Deconstructing schizophrenia: an overview of the use of endophenotypes in order to understand a complex disorder. Schizophr Bull 2007; 33: 21-32. 2. Green MJ, Cairns MJ, Wu J, Dragovic M, Jablensky A, Tooney PA, Scott RJ, Carr VJ. Genome-wide supported variant MIR137 and severe negative symptoms predict membership of an impaired cognitive subtype of schizophrenia. Mol Psychiatry 2013;18:774-780. 3. Hori H, Yamamoto N, Fujii T, et al. Effects of the CACNA1C risk allele on neurocognition in patients with schizophrenia and health individuals. Sci Rep 2012; 2: 634. 4. LeBlanc M, Kulle B, Sundet K, Agartz I, Melle I, Djurovic S, Frigessi A, Andreassen OA. Genome-wide study identifies PTPRO and WDR72 and FOXQ1-SUMO1P1 interaction associated with neurocognitive function. J Psychiatr Res 2012; 46: 2, 271-278. 5. Light GA, Braff DL. Mismatch negativity deficits are associated with poor functioning in schizophrenia patients. Arch Gen Psychiatry 2005; 62: 127-136. 6. Toulopoulou T, Picchioni M, Rijsdijk, et al. Substantial genetic overlap between neurocognition and schizophrenia. Arch Gen Psychiatry 2007: 64: 1348-1355. 7. Green MJ, Cairns MJ, Wu J, Dragovic M, Jablensky A, Tooney PA, Scott RJ, Carr VJ. Genome-wide supported variant MIR137 and severe negative symptoms predict membership of an impaired cognitive subtype of schizophrenia. Mol Psychiatry, in press. 8. Lipkovich IA, Deberdt W, Csernansky JG, Sabbe B, Keefe RS, KollackWalker S. Relationships among neurocognition, symptoms and functioning in patients with schizophrenia: a path-analytic approach for associations at baseline and following 24 weeks of antipsychotic drug therapy. BMC Psychiatry 2009; 9 doi: 10.1186/1471-244X-9-44. 9. Bowie CR, Leung WW, Reichenberg A, McClure MM, Patterson TL, Heaton RK, Harvey PD. Predicting schizophrenia patientsâ&#x20AC;&#x2122; real-world behavior with specific neuropsychological and functional capacity measures. Biol Psychiatry 2008; 1: 63, 505-511. 10. Lysaker PH, Carcione A, Dimaggio G, Johannesen JK, NicolĂ˛ G, Procacci M, Semerari A. Metacognition amidst narratives of self and illness in schizophrenia: Associations with insight, neurocognition, symptom and function. Acta Psychiatr Scand 2005; 112: 64-71. 11. van Os J, Kenis G, Rutten BP. The environment and schizophrenia. Nature 2010; 468:203-212. 12. Lysaker PH, Roe D, Yanos PT. Toward understanding the insight paradox: Internalized stigma moderates the association between insight and social functioning, hope and self-esteem among people with schizophrenia spectrum disorders. Schizophr Bull 2007; 33: 192-199.

Paul H. Lysaker and Ilanit Hasson-Ohayon

13. Walker EF, Diforio D. Schizophrenia: A neural diathesis-stress model. Psychol Rev 1997; 104: 667-685. 14. Read J, Perry BD, Moskowitz A, Connolly J. The contribution of early traumatic events to schizophrenia in some patients: A traumagenic neurodevelopmental model. Psychiatry 2001; 64: 319-345. 15. Lysaker PH, Lysaker JT. Schizophrenia and the fate of the self. Oxford: Oxford University Press, 2008. 16. Kean CS. Silencing the self: Schizophrenia as a self-disturbance. Schizophr Bull 2009; 35: 1034-1036. 17. Lampshire D. The sounds of a wounded world. In: Geekie J, Randal P, Lampshire D, Read J, editors. Experiencing psychosis: Personal and professional perspectives. New York: Routledge, 2012: pp. 139-145. 18. Lysaker PH, Glynn SM, Wilkness SM, Silverstein SM. Psychotherapy and recovery from schizophrenia: A review of potential application and need for future study. Psychol Serv 2010; 7: 75-91. 19. Brüne M, Dimaggio G, Lysaker PH. Metacognition and social functioning in schizophrenia: Evidence, mechanisms of influence and treatment implications. Curr Psychiatry Rev 2011; 7: 239-247. 20. Hasson-Ohayon I. Integrating cognitive behavioral-based therapy with an inter-subjective approach: Addressing metacognitive deficits among people with schizophrenia. J Psychotherapy Integration 2012; 22: 356-374. 21. Hasson-Ohayon I. Exploring the meaning of visual and auditory hallucinations: A commentary on Pixley’s discussion of the difficulties in addressing hallucinations during psychodynamic therapy. J Psychotherapy Integration 2012; 22: 393-396. 22. Tarricone P . The taxonomy of metacognition. East Sussex: Psychology Press, 2011. 23. Flavell JH. Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. Am Psychol 1979; 34: 906-911. 24. Wells A. Emotional disorders and metacognition: Innovative cognitive therapy. New York: John Wiley, 2000. 25. Bennett H, Wells A. Metacognition, memory disorganization and rumination in posttraumatic stress symptoms. J Anxiety Disord 2010; 24: 318-325. 26. Semerari A, Carcione A, Dimaggio G, et al. How to evaluate metacognitive function in psychotherapy? The metacognition assessment scale and its applications. Clin Psychol Psychother 2003;10: 238-261. 27. Main M. Metacognitive knowledge, metacognitive monitoring, and singular (coherent) vs. multiple (incoherent) models of attachment: Findings and directions for future research. In: Harris P, StevensonHinde J, Parkes C, editors. Attachment across the life cycle. New York: Routledge, New York, 1991: pp. 127-159.

28. Hesse E. The adult attachment interview: Protocol, method of analysis and empirical studies. In: Cassidy J, Shaver PR, editors. Handbook of attachment (2nd Ed.). New York: Guilford, 2008: pp. 552-598. 29. Dewey J. The middle works, 1899-1924, Volume 14. Human nature and conduct (1922). Boydston JA, editor. Carbondale, Illinois: Southern Illinois University, 1983. 30. Mitchell SA. Hope and dread in psychoanalysis. New York: Basic Books, 1993. 31. Lysaker PH, Vohs JL, Ballard R, Fogley R, Salvatore G, Popolo R, Dimaggio G. Metacognition, self reflection and recovery in schizophrenia: Review of the literature. Future Neurology 2013; 8: 103-115. 32. Parker S, Wells A, Morrison, AP. Metacognitive change as a predicator of outcome in cognitive therapy for psychosis. Isr J Psychiatry Relat Sci 2014;51: 8-16. 33. Brent BK, Holt DJ, Keshavan MS, Seidman LD, Fonagy P. Mentalizationbased treatment for psychosis: Linking an attachment-based model to the psychotherapy for impaired mental state understanding in people with psychotic disorders. Isr J Psychiatry Relat Sci 2014;51: 17-24. 34. Berna F, Huron C, Kazès M, Offerlin-Meyer I, Willard D, Verry P, Hédelin G, Krebs MO, Danion JM. Chronic persecutory delusion and autobiographical memories in patients with schizophrenia: A diary study. Isr J Psychiatry Relat Sci 2014;51: 25-33. 35. Scheyer R, Koren D, Reznik N, Adres M, Apter A, Seidman LJ. Metacognition in non-psychotic help-seeking adolescents: Associations with prodromal symptoms, distress and psychosocial deterioration. Isr J Psychiatry Relat Sci 2014;51: 34-43. 36. Rabin SJ, Hasson-Ohayon I, Avidan M, Rozencwaig S, Shalev H, Kravetz S. Metacognition in schizophrenia and schizotypy: Relations to symptoms and social quality of life. Isr J Psychiatry Relat Sci 2014;51: 44-53. 37. Lysaker PH, Leonhardt BL, Pijnenborg M, van Donkersgoed R, de Jong S, Dimaggion G. (in press). Metacognition in schizophrenia spectrum disorders: Methods of assessment and associations with neurocognition, symptoms, cognitive style and function. Isr J Psychiatry Relat Sci 2014;51: 54-62.

Paul H. Lysaker Roudebush VA Medical Center and the Indiana University School of Medicine, Department of Psychiatry, Indianapolis, Indiana, U.S.A.

Ilanit Hasson-Ohayon Department of Psychology, Bar-Ilan University, Ramat Gan, Israel

Guest editors

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis Sophie Parker, ClinPsyD,1,2 Adrian Wells, PhD, CPsychol,1 and Anthony P. Morrison, ClinPsyD1, 2 1

School of Psychological Sciences, University of Manchester, Manchester, U.K. Greater Manchester West NHS Mental Health Foundation Trust, Manchester, U.K.

Background: There is little known about predictors of response to cognitive therapy (CT) for psychosis. This study tests the hypothesis that metacognitive change at both end of treatment and follow-up is associated with positive outcomes in people with psychosis receiving CT. Method: Patients referred for CT for psychosis were offered CT over a maximum of 30 sessions. Assessments, including interview-based measures of psychotic symptoms and a questionnaire assessing metacognitive worry, were performed at pre-CT, post-CT and one-year follow-up. Data from 32 patients were analyzed. Results: Significant positive relationships were found between metacognitive worry change scores at one-year follow-up and both positive symptoms of psychosis and particular dimensions of hallucinations. No significant relationships were found between metacognitive change and delusional symptoms. Conclusions: These results suggest that metacognitive change is associated with symptom change at followup. In particular, these benefits seem most evident in those people experiencing hallucinations. Methodological limitations and clinical implications are discussed.

Introduction Cognitive therapy (CT) is effective when delivered in combination with antipsychotic medication. Several positive meta-analyses support this conclusion (1-3); however, two recent meta-analyses (4, 5) and a Cochrane review (6) suggest CT is not better than nonspecific control interventions or other psychosocial treatments,

although these reviews contain significant problems such as inclusion of inappropriate studies and data extraction errors. Additionally, of those studies indicating that CT is beneficial, few have examined the effects of therapy on specific cognitive processes that are hypothesized to mediate change. Such studies are helpful in order to target treatments, consider suitability, explore mechanisms of change and enhance treatment development work in this area. The Self-Regulatory Executive Function (S-REF) model (7, 8), suggests that vulnerability to psychological dysfunction and the maintenance of disorder is associated with a cognitive-attentional syndrome characterized by extended perseverative processing (such as worry and rumination), threat monitoring, and self-regulation strategies that fail to modify maladaptive self-knowledge. This syndrome is driven by metacognitive beliefs that specify the execution of worry/ruminative processing, guide attention, and serve as a resource for interpreting and controlling cognitive events such as unwanted thoughts. The model predicts an involvement of metacognitive beliefs in vulnerability to and the maintenance of psychopathology. Two subtypes of metacognitive beliefs are implicated in the S-REF model: positive beliefs (e.g., “worrying helps me cope,” “If I worry I can be prepared”) and negative beliefs (e.g., “worrying is uncontrollable,” “worrying thoughts are dangerous”). Both types of belief usually contribute to dysfunction but the stage and precise fashion in which they contribute can vary across disorders. Evidence supports the relationship between metacognitive beliefs and a range of negative emotional outcomes and psychiatric disorders, including generalized anxiety disorder (9), traumatic stress symptoms (10), alcohol use (11), and depression (12). As such a number of metacognitive models have been developed for a variety of psychological problems including generalized anxiety

Address for Correspondence: Sophie Parker, ClinPsyD, Psychology Services, Harrop House, Greater Manchester West NHS Foundation Trust, Bury New Road, Manchester M25 3BL, U.K. sophie.parker@gmw.nhs.uk

Sophie Parker et al.

disorder, social anxiety, depression and PTSD (12-14). One key metacognitive process identified in such models is that of metacognitive worry (or meta-worry), defined as worry about worry which is closely linked to negative beliefs about thinking. This approach to conceptualizing metacognition in psychological disorders is distinct from other approaches to defining metacognition in psychosis, such as Koren’s account of metacognitive monitoring (the subjective evaluation of one’s own neurocognitive functioning) and control (the manner in which one’s behavior is directed by this evaluation) (15), and Lysaker and colleagues’ work examining capacity for self-reflectivity within narrative contexts during psychotherapy (16). There is emerging evidence that a metacognitive approach to understanding psychosis, based upon predictions of the S-REF model, is valid (17, 18). For example, research has shown that metacognition is an important factor in hallucinations (19-21) and delusional beliefs (17, 18, 22-24). Morrison et al. also found that, consistent with the S-REF predictions, people with diagnoses of psychosis and those at ultra-high risk of developing psychosis score higher than non-patients on negative beliefs about unwanted thoughts, whereas only those with established psychotic disorders scored higher on positive beliefs about worry (25). In addition self focused attention (implicated in the S-REF model) is shown to be higher in people with paranoia (26) and those who hear voices (27). There is some support for the involvement of metacognitive worry in the development and maintenance of psychosis. For example, Freeman and Garety (22) found a positive association between delusional distress and metaworry, and a more recent study has also shown metacognitive worry to be associated with cognitive and emotional dimensions of delusional beliefs (28). A study of psychosis proneness in a large non-patient sample found higher levels of paranoid ideation to be associated with greater positive beliefs about worry and negative beliefs about the uncontrollability of thoughts (29). Another study of predisposition to hallucinations found that use of worry as a thought control strategy was associated with higher levels of hallucination-proneness (30). Therefore, there is some indication that meta-worry may be implicated in both the occurrence of psychotic experiences and the distress associated with them; however, much of this research is on non-patient samples and all has been cross-sectional in nature, so clinical research examining change over time in such variables is clearly required to further our understanding of such relationships.

There is emerging evidence that metacognitive approaches to the treatment of psychosis are beneficial. Early case studies suggest that components of metacognitive therapy (MCT), including postponement of perseverative processing and attention training, are promising in the treatment of auditory and visual hallucinations in psychosis (31, 32). A recent trial which focused on changing worry processes in people with persecutory ideas found that reduction of worry was also associated with reduction in paranoia (33), as cognitive and metacognitive models would predict. A small multiple baseline case series of participants with psychosis has shown that structured MCT seemed to demonstrate benefit (34). There is also some evidence that metacognitive changes resulting from CT are associated with symptom change and may be a mechanism of therapeutic change. In a recent trial of CT for people with psychosis without antipsychotic medication, it was found that reductions in metacognitive beliefs regarding uncontrollability of voices and negative beliefs about paranoia were associated with improvements in symptoms (35). This study aims to examine metacognitive worry as a predictor of outcome in CT for psychosis using data from an effectiveness study that confirmed the applicability of CT for psychosis in routine clinical settings (36). The original study aimed to test the feasibility and acceptability of delivering CT for people with psychosis within a community mental health team setting and to compare this to treatment as usual to test hypotheses about CT resulting in symptom reduction and reduced hospitalization. The current study aims to test the hypothesis that reductions in metacognitive worry occurring within CT for people with psychosis will be associated with improvements in psychotic symptoms. We tested two hypotheses: 1) that meta-worry decreases significantly from pre- to post- or pre- to follow-up, and 2) positive associations exist between change (improvement) in meta-worry and change (improvement) in symptoms. Method Participants

Those included in the trial were in contact with mental health services in a geographically defined area and had a diagnosis of schizophrenia, schizophreniform disorder, schizoaffective disorder or delusional disorder. All participants had to be aged 18-65 and currently scoring at least 4 on PANSS delusions or hallucinations. Information regarding exclusion criteria and participants diagnoses can be found elsewhere (36). 9

Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis

Measures:

Case note examination: A demographic and diagnostic checklist examining diagnostic criteria for psychotic disorders (using DSM-IV [37] criteria), in addition to age and gender, was used. A search of psychiatric notes was also made for details of medication and past inpatient admissions examining medication prescribed over the past year and days in hospital over the past year. In addition, an illness variables checklist examining length of illness and number of past relapses and compulsory admissions was used. Semi-structured Interviews

Positive and Negative Syndromes of Schizophrenia Scale (PANSS [38]): A clinician administered 30-item semi-structured interview consisting of seven items assessing positive symptomatology (e.g., hallucinations, delusions, conceptual disorganization), seven items assessing negative symptomatology (e.g., blunted affect, passive/apathetic social avoidance) and 16 items assessing global psychopathology (e.g., depression, anxiety, lack of insight, guilt). All items are scored between 1 (not present) and 7 (severe). A number of studies have demonstrated it possesses good reliability, validity and psychometric properties (39). PSYRATS - Auditory Hallucinations Rating Scale (40): A clinician administered semi-structured interview consisting of 11 items assessing dimensions of auditory hallucinations. All items are scored 0 to 4, with higher scores indicating more severe phenomena. The items assess frequency, preoccupation, location, loudness, conviction, amount of unpleasant content, severity of unpleasant content, amount of distress, intensity of distress, degree of impairment and control. Dimensional scores can be extrapolated relating to: emotional characteristics (i.e., distressing content items), physical characteristics (i.e., descriptions of voices), and cognitive interpretation (i.e., beliefs about the origins of voices, and attributions of control). PSYRATS - Delusions Rating Scale (40): A clinician administered semi-structured interview consisting of six items assessing dimensions of delusional beliefs. All items are scored 0 to 4, with higher scores indicating more severe phenomena. The items assess frequency, preoccupation, conviction, amount of distress, intensity of distress and degree of impairment. This produces two subscales: emotional characteristics and cognitive characteristics. Both PSYRATS scales possess good reliability and validity with sensitivity to change (40). 10

Self-Report

Beck Depression Inventory (41): A 21-item self-report measure used to assess depressive symptomatology where each item is rated 0-3 in terms of intensity. When selfadministered, the instrument generally takes 5-10 minutes to complete and is scored by summing the ratings given to each of the 21 items. Scores of 10 or less are considered normative. A number of studies show the BDI-II has good reliability and validity (42). Anxious Thoughts Inventory (43): The AnTI is a 22-item self-report measure aimed at capturing basic content domains and the distinction between worry and negative meta-appraisal on three subscales. These subscales consist of 1) Social worry (e.g., “I worry about doing or saying wrong things when among strangers”), 2) Health worry (e.g., “I worry about having a heart attack or cancer”) and 3) Meta-worry (e.g., “I worry that I cannot control my thoughts as well as I would like to”). Each items is rated 1-4 on a scale indicating “almost never” to “almost always.” The subscales show good psychometric properties (43). The meta-worry scale was of most relevance to this study. On this subscale student and non-patient groups show average scores of 11.3 and 12.8, whereas patient groups report scores ranging between 15.5 (in social phobia) and 19.7 (in generalized anxiety disorder) (9, 43). Participants were assessed with all of the measures at baseline, end of treatment and then again at 12-month follow-up. End of treatment was agreed collaboratively between therapist and participant with a maximum of 30 sessions delivered. With an average of 18 weekly sessions being delivered across the group, on average this was administered between 4-5 months after treatment began. However, with the range of sessions being between 0-30 and with some participants not always choosing to have weekly sessions this could have been shorter or longer. Follow-up assessments were carried out at 12 months following this end of treatment assessment point. Treatment

CT was delivered over a maximum of 30 sessions (average of 17.7, range between 0-30) with the spacing of sessions agreed collaboratively between therapist and patient (usually on a weekly basis). In addition, patients were able to receive up to three booster sessions following end of treatment. The general principles of cognitive therapy (44) were adhered to including being based on a cognitive model of the disorder in question (17), having a sound therapeutic relationship as a necessary

Sophie Parker et al.

condition for change, and being problem-orientated, time-limited, structured and directive. It was also educational, employing the inductive method and a process of guided discovery and collaborative empiricism. The overall therapy structure involved assessment, development of a problem list, socialization, idiosyncratic case formulation and symptom focused interventions selected on the basis of the case formulation and the goals. Further details regarding therapy and the main trial outcomes can be found elsewhere (36). Although the CT was based on a cognitive model, this included elements from the S-REF model, and the delivery of CT included metacognitive components such as evaluation of metacognitive beliefs (including negative beliefs about worry), postponement of perseverative processing (including worry) and modification of attentional processes. Data analysis

Distributions of the data were inspected for normality using visual inspection and analysis of skewness and kurtosis; all data were normally distributed therefore correlational analyses were conducted as tests of association between change scores and potential predictors using parametric statistics. All tests of significance were 2-tailed, but no correction was made for multiple comparisons given that this is an exploratory study in which we are less concerned about type 1 error. In order to reduce the potential for type 1 error in the correlational analyses, we only examined associations between change in hypothesized mechanisms and change in outcome for those mechanisms for which a significant change pre-post or pre-follow-up were found. The original study (36) found effects on all outcome measures except PANSS negative symptoms; therefore, we will examine those variables found to be significantly different between pre to post treatment in order to examine whether metacognition is associated with change in these domains. Meta-worry change scores were derived by subtracting the scores from the meta-worry subscale of the AnTI at post CT or follow-up time points from the baseline score on this subscale. Results The characteristics of the sample are presented in Table 1 and the characteristics of the main change mechanism (meta-worry) is presented in Table 2. The CONSORT diagram for the study is provided elsewhere (36); participants received a mean of 17.7 sessions (S.D. = 10). Analyses of the effects of CT on our outcome measures

Table 1. Participant characteristics Variable

Mean (SD) or N

Age

36.81 (10.36)

Male: Female ratio

40:19 (68%:32%)

Number of sessions

17.7 (10)

Duration of illness (months)

10.09 (6.98)

Table 2. Meta-worry Scale characteristics Time of completion (n)

Minimum

Maximum

Mean (SD)

Baseline (51)

17.06 (5.3)

Post CT (39)

15.95 (5.8)

Follow-up (22)

15.73 (5.9)

and potential mechanisms of change, at both end of treatment and follow-up, are presented elsewhere (36). Tables 3 and 4 show the correlation matrices for tests of associations between baseline variables, changes in potential mechanisms and changes in outcomes at end of treatment and follow-up. These correlations show that metacognitive change is not associated with improvements at end of treatment but it is at follow-up. More specifically, metacognitive change at follow-up is associated with change (reductions) in PANSS total, PANSS positive, PANSS global, PANSS positive item 3 (hallucinations), PSYRATS Hallucinations total, PSYRATS Hallucinations cognitive and PSYRATS Hallucinations physical. Given the association between metacognitive change and symptom improvement was seen only at follow-up, and particularly within the domain of hallucinations we conducted an additional set of correlational analyses excluding those who did not report any hallucinations at baseline in order to ensure that this lack of change was not inflating the results (akin to a sensitivity analysis). Table 5 shows that the same pattern was found in the smaller group, except that the significant association between meta-worry and PANSS global change was no longer found. Within the hallucinations group the associations between meta-worry and the individual scales on the PSYRATS related to hallucinatory symptoms (excluding emotional change) were all stronger. Discussion In this study we found that changes in scores on a measure of metacognition (meta-worry) were associated with improvements in symptoms at follow-up. This is consistent with Morrisonâ&#x20AC;&#x2122;s model of psychosis (17) which 11

Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis

Table 3. Correlation matrix for baseline variables and change in outcomes and mechanisms at end of treatment (n=33) 1. PANSS Total Change

2. PANSS Positive Change

0.740***

1.000

3. PANSS Global Change

0.930***

0.545***

1.000

4. PANSS Hallucinatory Behavior Change

0.267

0.427**

0.182

1.000

5. PANSS Suspiciousness / Persecution

0.604***

0.716***

0.445**

-0.003

1.000

6. PSYRATS Hallucinations Change

0.291

0.334*

0.248

0.719***

0.007

1.000

7. PSYRATS Hallucinations Cognitive change

0.334*

0.366*

0.287

0.686***

0.063

0.971***

1.000

8. PSYRATS Hallucinations Physical Change

0.265

0.202

0.263

0.708***

-0.069

0.949***

0.921***

1.000

9. PSYRATS Hallucinations Emotional Change

0.270

0.377*

0.221

0.668***

0.033

0.966***

0.910***

0.861***

1.000

10. PSYRATS Delusions Change

0.591***

0.677***

0.465***

0.140

0.573***

0.225

0.273

0.104

0.314*

1.000

11. PSYRATS Delusions Cognitive change

0.554***

0.714***

0.420**

0.175

0.590***

0.205

0.254

0.077

0.308*

0.958***

1.000

12. PSYRATS Delusions Emotional change

0.519***

0.447**

0.440**

0.042

0.406**

0.214

0.252

0.131

0.254

0.842***

0.652***

1.000

13. Meta-worry Change

0.060

0.175

0.052

0.213

-0.059

-0.070

-0.119

0.029

-0.011

-0.039

0.095

-0.223

***Correlation is significant at the 0.001 level ** Correlation is significant at the 0.01 level * Correlation is significant at the 0.05 level

draws from the predictions made within the S-REF model (7, 8) whereby metacognitive factors are implicated in the development and maintenance of psychotic phenomena. Our findings showed that these significant findings were observed on measures of total and psychotic symptoms and, more specifically, on domains of hallucinations; however, there was no association observed between delusional beliefs and meta-worry. This finding is consistent with research which suggests that unhelpful metacognitive beliefs are higher within a hallucination subgroup compared with those with delusions where people with hallucinations had higher levels of concerns about controllability and danger when compared with a group of people with delusions (45). However, some research has shown that dimensions of delusions in 12

general, and paranoia in particular, are associated with meta-worry (22, 28), and it has recently been suggested that metacognition is more associated with delusions than hallucinations (46). Thus, the discrepancy found here between the relationship between reductions in meta-worry and improvements in hallucinations but not delusions require further consideration. However, there was also an effect observed in relation to PANSS total (the most common primary outcome in CT for psychosis trials) and global symptoms, which include anxiety and depression; this would suggest that metacognitive change affects many different clinical outcomes. The treatment protocol for this study indicated that formulations and interventions were based on a cognitive model of psychosis (17). However, at the time of this

Sophie Parker et al.

Table 4. Correlation matrix for baseline variables and change in outcomes and mechanisms at follow up (n=21) 1. PANSS Total Change

2. PANSS Positive Change

0.863***

1.000

3. PANSS Global Change

0.963***

0.773***

1.000

4. PANSS Hallucinatory Behavior Change

0.473**

0.617**

0.397*

1.000

5. PANSS Suspiciousness / Persecution

0.719***

0.760***

0.677***

0.140

1.000

6. PSYRATS Hallucinations Change

0.434*

0.535**

0.414*

0.779***

0.045

1.000

7. PSYRATS Hallucinations Cognitive change

0.461*

0.534**

0.460*

0.736***

0.091

0.940***

1.000

8. PSYRATS Hallucinations Physical Change

0.468*

0.523**

0.438*

0.797***

0.014

0.907***

0.893***

1.000

9. PSYRATS Hallucinations Emotional Change

0.328

0.473*

0.313

0.690***

0.020

0.957***

0.842***

0.757***

1.000

10. PSYRATS Delusions Change

0.425*

0.584**

0.372

0.053

0.600**

0.080

0.135

-0.012

0.110

1.000

11. PSYRATS Delusions Cognitive Change

0.423*

0.588**

0.384

0.109

0.573**

0.045

0.122

-0.019

0.067

0.980***

1.000

12. PSYRATS Delusions Emotional Change

0.461*

0.573**

0.404

0.048

0.610**

0.130

0.146

0.001

0.171

0.945***

0.862***

1.000

13. Meta-worry Change

0.626**

0.570**

0.528*

0.552**

0.428

0.462*

0.508*

0.573*

0.331

0.161

0.224

0.136

***Correlation is significant at the 0.001 level ** Correlation is significant at the 0.01 level * Correlation is significant at the 0.05 level

Table 5. Correlation matrix for baseline variables and change in outcomes and mechanisms at follow-up (Hallucinations only group, n=13) 1. PANSS Total Change

2. PANSS Positive Change

0.822***

1.000

3. PANSS Global Change

0.972***

0.757***

1.000

4. PANSS Hallucinatory Behavior Change

0.488*

0.621**

0.473*

1.000

5. PSYRATS Hallucinations Change

0.524*

0.586*

0.530*

0.767***

1.000

6. PSYRATS Hallucinations Cognitive Change

0.584*

0.626**

0.607**

0.740***

0.906***

1.000

7. PSYRATS Hallucinations Physical Change

0.538*

0.535*

0.537*

0.792***

0.824***

0.832***

1.000

8. PSYRATS Hallucinations Emotion Change

0.327

0.464

0.351

0.601**

0.926***

0.743***

0.565*

1.000

9. Meta-worry Change

0.631*

0.582*

0.514

0.746**

0.689*

0.736**

0.878***

0.451

1.000

***Correlation is significant at the 0.001 level ** Correlation is significant at the 0.01 level * Correlation is significant at the 0.05 level

Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis

study, the role of metacognition was most implicated in specific models of hallucinations (47), rather than delusions. The approach for auditory hallucinations explicitly incorporated modification of negative metacognitive beliefs and attentional processes (see treatment protocol for auditory hallucinations (48). Thus, it is possible that the findings are indicative of greater attention being paid to metacognitive factors in those patients who presented with auditory hallucinations than in those who did not. A very recent study from the same research group, based on the same model, found changes in metacognition were observed in relation to paranoia as well as hallucinations; however, the recent work that has elaborated the role of metacognition in persecutory delusions (18) may have resulted in greater attention being paid to such factors in this patient group. If the finding that patients experiencing hallucinations have higher negative beliefs about worry than patients with delusions (45) is valid, then it may be that the higher presence of these beliefs in the hallucinatory group may provide greater opportunity for change. However, it is important at this stage not to emphasize this differential finding, as there was a strong trend towards a significant relationship between meta-worry change and suspiciousness (p=0.053), which would have been significant using a one-tailed test and would be likely to reach significance in a larger sample with greater statistical power. The other interesting result that we found in this study is that the significant associations between meta-worry and outcome variables were found at the follow-up period but not at the end of treatment. This is not a unique finding, e.g., 49, identified stronger effects of CT for psychosis at follow-up than end of treatment. Given that metacognitive techniques are aimed at learning new skills, it could be hypothesized that these skills get better with practice over time. For example, the application of shifting attention becomes better with practice (50) and the acquisition of new information over longer time periods may decrease patientsâ&#x20AC;&#x2122; conviction in negative metacognitive beliefs. This study has several limitations which mean that these results should be interpreted with caution and require replication in larger, preferably randomized, samples. The application of multiple correlations increases the chance of type 1 error; however, the findings followed a consistent pattern, which can provide some confidence in our findings. The small sample size, which was a convenience sample, clearly limits statistical power; however, we nonetheless found significant effects of our hypothesized 14

mechanisms on several outcome measures. The sample was also diagnostically heterogeneous (schizophrenia spectrum disorders), which could be viewed as a methodological weakness; however, given the development of services for psychosis and the emphasis on diagnostic uncertainty that exists within early intervention services, it should ensure that our findings are potentially generalizable to real world settings. We did not systematically record the use of particular change strategies, including metacognitive techniques; therefore, we are unable to examine the relationship of specific change strategies and symptom change, which could have informed our findings regarding discrepancies in effects on hallucinations in comparison to delusions. Likewise, given that no formal assessment of treatment fidelity was undertaken we cannot be certain that the hypothesized mechanisms were targeted explicitly. However, we anticipate that the delivery of CT within the study was consistent as the supervision and training of therapists should have ensured this. There are also many ways to examine the issue of mediation of treatment effects from a statistical point of view, and the correlation of changes in hypothesized mediators and outcomes is open to criticism for a variety of reasons; most significantly, it is theoretically possible that there is another unmeasured variable that accounts for changes in both that is unrelated to treatment. However, with our small sample these exploratory analyses are all that is possible. Similarly, because our trial was an uncontrolled, open trial without a comparator group, we cannot conclude with any certainly that our hypothesized mechanisms changed as a result of treatment; it is possible that an unmeasured variable was responsible for changes in both mechanism and outcome. Future research could examine these issues using approaches such as traditional mediation analysis (51), which assumes that such confounding is absent, as well as attempting to measure and adjust for all important confounders (52), and attempting to effectively adjust for unmeasured confounders (hidden confounding) using instrumental variable-based methods employing analyses based on principal stratification (53). Clinical implications Given that we had a small non-randomized sample, and that our findings were exclusively at follow-up and mostly for hallucinations, this research requires replication and extensions before we can have robust confidence in the clinical implications. However, within this study of CT,

Sophie Parker et al.

metacognitive change appears to predict outcome; therefore, attention should be given to metacognitive elements of the model and consideration given to using strategies derived from MCT; we hypothesize that MCT may be one way of doing this which has shown encouraging preliminary evidence (34). This may be of particular use to people who experience hallucinations although further research is required before any strong conclusions are made. References 1. Wykes T, Steel C, Everitt B, Tarrier N. Cognitive behavior therapy for schizophrenia: Effect sizes, clinical models, and methodological rigor. Schizophr Bull 2008;34:523-537. 2. Pilling S, Bebbington PE, Kuipers E, Garety PA, Geddes J, Orbach G, et al. Psychological treatments in schizophrenia: I. Meta-analysis of family intervention and cognitive behaviour therapy. Psychol Med 2002;32:763-782. 3. Zimmerman G, Favrod J, Trieu VH, Pomini V. The effect of cognitive behavioural treatment on the positive symptoms of schizophrenia spectrum disorders: A meta analysis. Schizophr Res 2005;77:1-9. 4. Lynch D, Laws KR, McKenna PJ. Cognitive behavioural therapy for major psychiatric disorder: Does it really work? A meta-analytical review of well-controlled trials. Psychol Med 2010;40:9-24. 5. Newton-Howes G, Wood R. Cognitive behavioural therapy and the psychopathology of schizophrenia: Systematic review and meta-analysis. Psychol Psychother Theor Res Pract 2013:86:127-138. 6. Jones C, Hacker D, Cormac I, Meaden A, Irving CB. Cognitive behaviour therapy versus other psychosocial treatments for schizophrenia. Cochrane Database Syst Rev 2012;4. 7. Wells A, Matthews G. Modelling cognition in emotional disorder: The S-REF model. Behav Res Ther 1996;34:881-888. 8. Wells A, Matthews G. Attention and emotion: A clinical perspective. London: LEA, 1994. 9. Wells A, Carter K. Further tests of a cognitive model of generalized anxiety disorder: Metacognitions and worry in GAD, panic disorder, social phobia, depression, and nonpatients. Behav Ther 2001;32:85-102. 10. Roussis P, Wells A. Post-traumatic stress symptoms: Tests of relationships with thought control strategies and beliefs as predicted by the metacognitive model. Pers Individ Dif 2006;40:111-122. 11. Spada M, Wells A. Metacognitions about alcohol use in problem drinkers. Clin Psychol Psychother 2006;13:138-143. 12. Papageorgiou C, Wells A. Positive beliefs about depressive rumination: Development and preliminary validation of a self-report scale. Behav Ther 2001;32:13-26. 13. Wells A. Meta-cognition and worry: A cognitive model of generalised anxiety disorder. Behav Cogn Psychother 1995;23:301-320. 14. Clark DM, Wells A. A cognitive model of social phobia. In: Heimberg RG, Liebowitz MR, editors. Social phobia: Diagnosis, assessment, and treatment. New York, N.Y.: Guilford, 1995: pp. 69-93. 15. Koren D, Seidman LJ, Goldsmith M, Harvey PD. Real-world cognitive - and metacognitive - dysfunction in schizophrenia: A new approach for measuring (and remediating) more “right stuff.” Schizophr Bull 2006;32:310-326. 16. Lysaker PH, Buck KD, Carcione A, Procacci M, Salvatore G, Nicolò G, et al. Addressing metacognitive capacity for self reflection in the psychotherapy for schizophrenia: A conceptual model of the key tasks and processes. Psychol Psychother Theor Res Pract 2011;84:58-69. 17. Morrison AP. The interpretation of intrusions in psychosis: An integrative cognitive approach to hallucinations and delusions. Behav Cogn Psychother 2001;29:257-276.

18. Morrison AP, Gumley AI, Ashcroft K, Manousos R, White R, Gillan K, et al. Metacognition and persecutory delusions: Tests of a metacognitive model in a clinical population and comparisons with non-patients. Br J Clin Psychol 2011;50:223-233. 19. Baker CA, Morrison AP. Cognitive processes in auditory hallucinations: Attributional biases and metacognition. Psychol Med 1998;28:1199-1208. 20. Morrison AP, Nothard S, Bowe SE, Wells A. Interpretations of voices in patients with hallucinations and non-patient controls: A comparison and predictors of distress in patients. Behav Res Ther 2004;42:1315-1323. 21. Morrison AP, Wells A, Nothard S. Cognitive and emotional predictors of predisposition to hallucinations in non-patients. Br J Clin Psychol 2002;4:259-270. 22. Freeman D, Garety PA. Worry, worry processes and dimensions of delusions: An exploratory investigation of a role for anxiety processes in the maintenance of delusional distress. Behav Cogn Psychother 1999;27:47-62. 23. Morrison AP, Gumley AI, Schwannauer M, Campbell M, Gleeson A, Griffin E, et al. The beliefs about paranoia scale: Preliminary validation of a metacognitive approach to conceptualising paranoia. Behav Cogn Psychother 2005;33:153-164. 24. Startup H, Freeman D, Garety PA. Persecutory delusions and catastrophic worry in psychosis: Developing the understanding of delusion distress and persistence. Behav Res Ther 2007;45:523-537. 25. Morrison AP, French P, Wells A. Metacognitive beliefs across the continuum of psychosis: Comparisons between patients with psychotic disorders, patients at ultra-high risk and non-patients. Behav Res Ther 2007;45:2241-2246. 26. Smari J, Stefansson S, Thorgilsson H. Paranoia, self-consciousness and social cognition in schizophrenics. Cognit Ther Res 1994;18:387-399. 27. Morrison AP, Haddock G. Self-focused attention in schizophrenic patients with and without auditory hallucinations and normal subjects: A comparative study. Pers Individ Dif 1997;23:937-941. 28. Morrison AP, Wells A. Relationships between worry, psychotic experiences and emotional distress in patients with schizophrenia spectrum diagnoses and comparisons with anxious and non-patient groups. Behav Res Ther 2007;45:1593-1600. 29. Laroi F, van der Linden M. Metacognitions in proneness towards hallucinations and delusions. Behav Res Ther 2005;43:1425-1441 30. García-Montes JM, Cangas A, Pérez-Álvarez M, Fidalgo ÁM, Gutiérrez O. The role of meta-cognitions and thought control techniques in predisposition to auditory and visual hallucinations. Br J Clin Psychol 2006;45:309-317. 31. Valmaggia L, Bouman T, Schuurman L. Attention training with auditory hallucinations: A case study. Cogn Behav Pract 2007;14:127-133. 32. Hutton P, Morrison AP, Taylor H. Brief cognitive-behavioural therapy for hallucinations: Can it help people who decide not to take antipsychotic medication? A case report. Behav Cogn Psychother 2012;40:111-116. 33. Foster C, Startup H, Potts L, Freeman D. A randomised controlled trial of a worry intervention for individuals with persistent persecutory delusions. J Behav Ther Exp Psychiatry 2010;41:45-51. 34. Hutton P, Morrison AP, Wardle M, Wells A. Metacognitive therapy in treatment-resistant psychosis: A multiple-baseline study. Behav Cogn Psychother 2012 (in press). 35. Morrison AP, Turkington D, Wardle M, Spencer H, Barratt S, Dudley R, et al. A preliminary exploration of predictors of outcome and cognitive mechanisms of change in cognitive behaviour therapy for psychosis in people not taking antipsychotic medication. Behav Res Ther 2012;50:163-167. 36. Morrison AP, Renton JC, Williams S, Dunn H, Knight A, Kreutz M, et al. Delivering cognitive therapy to people with psychosis in a community mental health setting: An effectiveness study. Acta Psychiatr Scand 2004;110:36-44. 37. American Psychiatric Association. Diagnostic and statistical manual for mental disorders, 4th edition. Washington DC: Author 1994.

Metacognitive Change as a Predictor of Outcome in Cognitive Therapy for Psychosis

38. Kay SR, Opler LA. The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull 1987;13:507-518. 39. Kay SR, Opler LA, Fiszbein A. Reliability and validity of the Positive and Negative Syndrome Scale for schizophrenics. Psychiatry Res 1988;23:276-286. 40. Haddock G, McCarron J, Tarrier N, Faragher EB. Scales to measure dimensions of hallucinations and delusions: The psychotic symptoms rating scales (PSYRATS). Psychol Med 1999;29:879-889. 41. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;41:53-63. 42. Beck AT, Steer R, Gabin M. Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clin Psychol Rev 1988;8:77-100. 43. Wells A. A multidimensional measure of worry: Development and preliminary validation of the Anxious Thoughts Inventory. Anxiety Stress Coping 1994;6:289-299. 44. Beck AT. Cognitive therapy and the emotional disorders. New York: International Universities, 1976. 45. Morrison AP, Wells A. Metacognition across disorders: A comparison of patients with hallucinations, delusions, and panic disorder with non-patients. Behav Res Ther 2003;41:251-256. 46. Varese F, Bentall RP. The metacognitive beliefs account of hallucinatory

experiences: A literature review and meta-analysis. Clin Psychol Rev 2011;31:850-864. 47. Morrison AP, Haddock G, Tarrier N. Intrusive thoughts and auditory hallucinations: A cognitive approach. Behav Cogn Psychother 1995;23:265-280. 48. Morrison AP, Renton JC. Cognitive therapy for auditory hallucinations: A theory-based approach. Cogn Behav Pract 2001;8:147-169. 49. Sensky T, Turkington D, Kingdon D, Scott JL, Scott J, Siddle R, et al. A randomized controlled trial of cognitive-behavioral therapy for persistent symptoms in schizophrenia resistant to medication. Arch Gen Psychiatry 2000;57:165-172 50. Wells A. Metacognitive therapy for anxiety and depression. New York: Guilford, 2009. 51. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51:1173-1182. 52. Van der Weele TJ. Marginal structural models for the estimation of direct and indirect effects. Epidemiology 2009;20:18-26. 10.1097/ EDE.0b013e31818f69ce. 53. Emsley R, Dunn G, White IR. Mediation and moderation of treatment effects in randomised controlled trials of complex interventions. Stat Methods Med Res 2010;19:237-270.

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Benjamin K. Brent Et al.

Mentalization-based Treatment for Psychosis: Linking an Attachment-based Model to the Psychotherapy for Impaired Mental State Understanding in People with Psychotic Disorders Benjamin K. Brent, MD, MS,1,3 Daphne J. Holt, MD, PhD,2,3 Matcheri S. Keshavan, MD,1,3 Larry J. Seidman, PhD,1,2,3 and Peter Fonagy, PhD4 1

Department of Psychiatry, Beth Israel Deaconess Medical Center and Massachusetts Mental Health Center, Boston, Mass., U.S.A. Department of Psychiatry, Massachusetts General Hospital, Harvard Medical School, Boston, Mass., U.S.A. 3 Harvard Medical School, Boston, Mass., U.S.A. 4 Psychoanalysis Unit, University College London, London, U.K. 2

Abstract Disturbances of mentalization have been increasingly associated with the symptoms and functional impairment of people with psychotic disorders. It has been proposed that psychotherapy designed to foster self and other understanding, such as mentalization-based treatment (MBT), may play an important part in facilitating recovery from psychosis. Here, we present an attachment-based understanding of mentalization impairments. We then outline a neuropsychological model that links disruptions of mentalization associated with disturbances in the caregiving environment to the pathophysiology of psychosis in genetically at-risk individuals. This is followed by an illustration of some of the core MBT techniques for the rehabilitation of the capacity to mentalize as applied to the treatment of a patient with a psychotic disorder.

Introduction Impaired mentalization (i.e., the capacity to think about mental states in the self and others) is increasingly regarded as an important psychopathological domain in people with psychotic disorders (1-3). Because psychotic symptoms frequently involve misunderstandings of social

situations (e.g., persecutory delusions and hallucinations), or of the person’s self-appraisal with respect to other people (e.g., grandiose or religious delusions), it has been proposed that the disruption of the capacity for social understanding may constitute a key vulnerability to psychosis (4). This hypothesis has received support from studies showing associations between mentalization deficits (e.g., impaired theory of mind [ToM]) and core psychotic symptoms, such as delusions and hallucinations (5). Moreover, growing evidence links aberrant mentalization to the social dysfunction (e.g., inability to work, poor quality of life) that commonly accompanies psychosis (6). The strength of the relationship between impaired mentalization and the socially disabling effects of psychosis is highlighted by a recent meta-analysis that showed that ToM impairments exhibited the strongest association with social dysfunction in schizophrenia of any social or neurocognitive domain tested (7). One question raised by these findings is whether psychotherapy focusing on deficits of social understanding, such as mentalization-based treatment (MBT), can facilitate the recovery of patients with psychosis. Closely related to MBT is metacognitive psychotherapy (8, 9), which targets deficits of “thinking about thinking” in schizophrenia. Metacognition and mentalization have been linked theoretically and empirically, as both involve meta-representational abilities (10). Typically, metacognition is closely associated with self-monitoring one’s cognitive performance (e.g., evaluating how well one

Address for Correspondence: Benjamin K. Brent, MD, Massachusetts Mental Health Center, 75 Fenwood Road - Room 618, Boston, MA 02139, U.S.A. bbrent@bidmc.harvard.edu

Mentalization-Based Treatment for Psychosis

has learned the material for a test), while mentalization focuses more specifically on the representation of states of mind in the self and other people (10). Recently, however, conceptualizations of metacognition have included the integration of complex self and other representations, thus strengthening the theoretical overlap between mentalizing and metacognitive functions (8). In several case reports, people with schizophrenia who received metacognitive psychotherapy have shown significant improvement of social function (11, 12). Here, we provide a conceptual framework to support the adaptation of MBT to the understanding and treatment of individuals with psychosis. We begin by outlining an attachment-based model of mentalization disturbances. This is followed by a theoretical overview regarding how disruptions within the caregiving environment may confer vulnerability to psychosis. We conclude by describing the key features of an MBT approach to the enhancement of self and other understanding in psychotic disorders. Attachment Disturbance, Impaired Mentalization and Psychosis Attachment and Mentalization

A growing literature suggests that understanding mental states is related to the social context in which thinking about minds initially develops, namely attachment relationships involving caregivers (13). During the first year of life, a child and a caregiver develop an emotional bond (attachment), which is thought to create for the child a feeling of safety in proximity to the caregiver and reflect the child’s expectations in turning to the caregiver for comfort during periods of emotional distress (14). Within an attachment theory framework, it is predicted that a caregiver’s attunement to the child’s attachment-seeking behavior (e.g., crying, smiling, clinging), and generally reliable responses to signs of infantile distress, gradually promote the child’s sense of safety with the caregiver (14). Secure attachment relationships are increasingly internalized, leading to an experience of genuine relatedness coupled with appropriate independence and self-sufficiency. Moreover, attachment security has been shown to contribute to the early development of the capacity to link behavior with states of mind – feelings, thoughts and desires (15). On the other hand, greater caregiver misattunement to the child’s efforts to achieve closeness is thought to undermine a child’s expectation of safety from that relationship and lead to attachment patterns indicative of insecurity or disorganization (16). Further, 18

children who experience significant attachment dysfunction (e.g., maltreated children) have shown delays in the acquisition of mental state understanding (15). Insecurity can make itself felt through continued need for physical proximity to the attachment figures, or exaggerated claims of self-sufficiency and pretence of independence. Given these links between attachment and the development of mentalization, the caregiving environment may have a significant moderating influence on the capacity for accurate interpersonal understanding (17). Attachment Dysfunction and Psychosis Risk

Two decades ago, Frith initially theorized that disruption of the ability to represent one’s own mind and the minds of others constitutes a core neuropsychological vulnerability to psychosis (18). There is now consistent evidence for mentalization impairments in people with psychotic disorders (19). However, the extent to which mentalization deficits that arise in the context of dysfunctional attachment relationships contribute to psychosis remains an unresolved question. We recognize that many factors may lead to mentalization deficits in people with psychotic disorders that are not necessarily related to the quality of the caregiving environment, such as temperament, traumatic events independent of experiences with caregivers, or substance misuse. We suggest, however, that evidence from developmental psychology regarding the connection between attachment disruptions and impaired mentalization may provide a valuable additional link for advancing current understanding regarding the contribution of disturbances within the caregiving environment to psychosis (Figure 1). Consistent with contemporary diathesis-stress models (20), we speculate that mentalization impairments arising in the context of aberrant caregiving relationships may interact with dysregulation of the stress-response system and of mesolimbic dopamine to heighten the risk for psychosis in genetically vulnerable people. Aberrant relationships with caregivers are increasingly recognized as an important environmental risk factor for psychosis (20). Chronic exposure to life stressors, such as aberrant attachment relationships, is thought to increase the risk for psychosis in large part because of its disruption of the biological system responsible for regulating stress (the hypothalamic–pituitary–adrenal [HPA] axis) (21). Because the HPA-axis system stimulates dopamine synthesis and release, chronic life stressors that result in HPA-axis overactivity may contribute to the dysregulation of prefrontal and corticolimbic dopamine circuits that

Benjamin K. Brent Et al.

have been linked with key psychotic symptoms, such as delusions and hallucinations (21). Additionally, in the context of attachment insecurity or disorganization, dopamine dysfunction could become amplified through decreased levels of oxytocin (the key neurohormone associated with social attachment). It has been shown, for example, that oxytocin is reduced in children exposed to an aberrant caregiving environment (22). Given the evidence that oxytocin has an inhibitory effect on mesolimbic dopamine (23), lower levels of oxytocin may act synergistically with disruptions of the HPA-axis system to contribute to dopaminergic dysregulation in people at risk for psychosis who experience significant disruptions in relationships with caregivers. According to aberrant salience models of psychosis, dopamine dysfunction may provide a biological vulnerability for “heightened states of awareness” and subsequent misinterpretations of internal and external stimuli (24). Thus, individuals at risk for psychosis who develop mentalization impairments in the context of attachment insecurity may be particularly vulnerable to elaborating abnormal explanations of social experience during periods of acute stress. For example, at-risk individuals with difficulties discerning others’ intentions, or evaluating their position in the world relative to others, may be prone to evolve mild paranoid or grandiose beliefs. Further, difficulties differentiating internal from external

sensory experience, or a compromised sense of agency, might increase the risk of aberrant perceptual experiences among people with risk genes for psychosis. These aberrant beliefs and perceptions could, in turn, become reinforced and maintained via underlying dopamine dysfunction, leading to further anomalies in social and self-understanding. Mentalization Impairments in Psychotic Disorders

Mentalization impairments have increasingly been linked with a broad range of disorders, including borderline (19, 25) and schizotypal (5) personality disorder, and autism (26). There is growing recognition that aberrant “self-experience” occurs across all major mental illness to some degree (27). According to phenomenological models, it has been proposed that anomalous self-experience in schizophrenia is marked by a hyper-focus on inner mental states (“hyper-reflexivity”) together with a loss of the sense of being the subject of one’s experience (i.e., loss of “ipseity”) (28). Accordingly, experiences of caregiver neglect that impede curiosity about others’ minds and contribute to “hypermentalizing” (excessive reflection on the self) may play an important role in the development of psychosis (29). Indeed, “autism” (e.g., preoccupation with fantasy, or the withdrawal of interest in personal hygiene and/or relationships with other people) has long been viewed as a defining aspect of the psychopathology of people with

Figure 1. Model Linking Mentalization Impairments Associated with the Early Caregiving Environment and Psychosis. In genetically predisposed individuals, attachment disturbances may contribute to: 1) impaired self–other understanding (mentalization disturbances) and 2) dopamine dysregulation and heightened states of awareness resulting from chronic stress/HPA-axis dysfunction, combined with reduced oxytocin/mesolimbic dopamine inhibition. Individuals with mentalization impairments may be prone to elaborating abnormal explanations of social experience that, in conjunction with heightened states of awareness, could constitute a psychological/ neurobiological vulnerability for the eventual emergence of psychotic symptoms.

Mentalization-Based Treatment for Psychosis

psychosis and has posed a central challenge to psychotherapeutic interventions (30). The specific mentalization deficits occurring in psychotic disorders, however, remain to be determined empirically (5). A meta-analysis of ToM deficits in schizophrenia showed similar levels of mentalizing impairment across a wide range of ToM tasks (e.g., verbal vs. nonverbal tasks, or first-order vs. second-order ToM) (19). Differentiating the mentalizing disturbances associated with psychotic disorders from those occurring in borderline personality disorder, autism, or other major mental illness is complicated by several factors: 1) the limitations of current assessment methods (5); 2) the extent to which mentalization impairments are linked to specific social contexts (27); 3) the considerable diagnostic heterogeneity of people with psychosis (4); and 4) the high rates of psychiatric comorbidity, including personality disorders, among psychotic disorder patients (31). Future longitudinal studies are needed to determine which attachment contexts and forms of mentalization deficits are specific to people who develop psychosis. Mentalization-Based Treatment for Psychosis MBT is a manualized, evidence-based treatment for addressing the core symptoms (i.e., affect dysregulation, impulsivity, and self-harm behaviors) of borderline personality disorder (32). However, because psychotic symptoms commonly arise situationally within specific social contexts (e.g., misunderstanding other people’s intentions), the rehabilitation of the capacity for self and other understanding may be a critical component of the mechanism of change in the treatment of people with psychosis. Assessment of Mentalization

Two of the central goals of MBT are to foster the capacity for mentalization about the self and others and to facilitate understanding of the way that mentalization is affected by specific interpersonal relationships. Given the links between mental state understanding and the quality of an individual’s social experience, patients may frequently exhibit impairments in some, but not all, aspects of mentalization (17). In the treatment of people with psychosis, developing an understanding of the particular domains of impairment in an individual’s capacity to think about mental states helps to establish a treatment focus. It provides an indication of the social and attachment contexts within which disturbances 20

of mentalization are most likely linked with psychotic symptoms and, therefore, should be addressed. Early in the course of treatment, the quality of mentalization is probed through discussions of the patient’s thinking within interpersonal relationships. The initial presentation of Rachel illustrates the assessment of mentalization in a person with early psychosis: Rachel is a 25-year-old woman, the oldest child of four siblings, with one psychiatric hospitalization for paranoid delusions accompanied by auditory hallucinations. Rachel came to treatment for depression about falling short of her professional goals and not being able to develop lasting intimate relationships. She said she wanted to be able to get closer to other people, but felt that she could not. Rachel described a chaotic early family environment. Her father struggled with alcohol and drug addiction and there was perpetual conflict between her parents. When she was 10 years old, Rachel’s parents divorced. Neither parent felt able to care for her, and she was sent to live with her grandparents. During her first summer away from her mother, Rachel remembers a time when she looked into the night sky and saw one star that seemed to be brighter than all the others. She began to wonder whether this might be a message from God telling her that she was a fallen angel. During high school, Rachel continued to search for clues about God’s plan for her. But she remained uncertain about whether her interpretations of signs regarding God’s intentions were “real” or just in her mind. Rachel graduated from high school with good grades. She had some friends, but never dated. Rachel had difficulty, however, adjusting to college. Her grades began to drop and she increasingly worried about being a failure. Ultimately, Rachel became convinced that her academic problems were an indication that she was being punished by God. A first episode of psychosis ensued, followed by hospitalization and subsequent maintenance on antipsychotic medication. Since that time, Rachel’s acute psychotic symptoms stabilized, but she remained discouraged by her difficulties functioning socially. From an MBT perspective, Rachel’s history provides an example of how psychosis may be linked to attachment contexts (e.g., disturbances in relationships with caregivers) in which the capacity for mental state understanding is particularly likely to become impaired in vulnerable

Benjamin K. Brent Et al.

individuals. For Rachel, both the initial emergence of her psychotic-like beliefs about being a fallen angel and her subsequent episode of full-blown psychosis were associated with periods of significant separation and/or disruption involving important caregivers (i.e., during the time after her parents’ divorce, and then subsequently after leaving home to start college). In people whose capacity to reflect on the self and others has become compromised, maintaining physically proximate relationships with caring others is thought to play an increasingly pronounced role in the regulation of affect and self-experience (25). During periods of separation, such individuals may be particularly vulnerable to the situational breakdown of self-coherence that had been maintained through the attachment relationship, which may lead to an increased risk of impaired reality monitoring and psychosis, especially given an underlying genetic risk. To more fully characterize Rachel’s current ability to mentalize, Rachel was asked how she understood why her mother had sent her to live with her grandparents. Rachel said she thought her mother did not love her. She was unable to entertain alternatives – for example, that her mother may have been overwhelmed trying to raise four children as a single parent. Rachel was also asked to describe her current relationship difficulties. She said she was sure that other people thought there was something wrong with her. Rachel explained that she recently went to a party, but was unable to talk and felt “frozen.” All she could think about was how everyone must be thinking she was a failure. Rachel noticed herself thinking: “Everyone here has done so much more with their lives… I feel like such a failure… Why would anyone want to talk to me?” She assumed that other people must have been thinking the same thing about her. During this initial assessment, Rachel exhibited a tendency toward rigid, inflexible assumptions about other people’s minds (e.g., regarding her mother’s behavior toward her in childhood, or the other people’s thoughts about her at the party), typical of “psychic equivalence mentalization,” where internal and external reality are given equivalent status (25). Additionally, Rachel identified an interpersonal context in which her capacity to mentalize appeared particularly disrupted; namely, in the situation of the party, which presented the possibility of forming new relationships that might activate attachment needs (e.g., the need for emotional closeness). In the setting of increased anxiety about meeting someone new, Rachel’s ability to mentalize became acutely compromised, as she felt “frozen,” preoccupied with her own internal

criticism, and unable to interact with other people. For Rachel, self-critical judgments (e.g., feeling like a failure during college) also appeared closely related to her prior difficulties maintaining a sense of the difference between her own thinking and external reality, and to the phenomenology of her psychotic delusion (i.e., that she was being punished by God). Thus, fostering Rachel’s curiosity about her own and others’ minds, and helping her to understand how interpersonal settings involving the heightening of attachment needs might affect her capacity to think about herself and others, became central foci of the initial phase of treatment. Therapeutic Stance and Basic Interventions

In MBT, the therapist’s focus on the patient’s state of mind is critical to the development of a collaborative mentalizing process. In particular, taking an inquisitive, “not-knowing” therapeutic stance with regard to what a patient is thinking or feeling is viewed as fundamental to the evolution of the patient’s curiosity about how his/ her mind works and the generation of second-order representations in relation to mental states (19). Active questioning about the patient’s mental state and detailed exploration of how the patient’s state of mind is related to particular interpersonal contexts are employed to demonstrate the therapist’s interest in understanding the way that what is going on in the patient’s mind is related to the concerns that have led him/her to seek psychotherapy. For example: After her initial assessment, Rachel began her next therapy appointment by asking: “What should I talk about today?” Rather than assuming to know what Rachel should talk about, or why she was having trouble getting started on this occasion, her therapist began by observing that initially she had spoken very freely, but something seemed different today. “I’m not sure I know what you should talk about today,” he said. “But, maybe we could try to think about what’s making it hard for you to come up with a topic together?” Rachel looked down at the floor, and then said: “I thought you would know what I should talk about.” Rachel then told a story about a recent family gathering. During a conversation with her uncle, Rachel felt she had unintentionally made a critical comment that hurt her uncle’s feelings. After their conversation, Rachel thought her uncle had given her a disapproving look. She felt rejected. In an effort to stimulate Rachel’s curiosity about her state of mind at the beginning 21

Mentalization-Based Treatment for Psychosis

of the hour, Rachel’s therapist noted that after saying she didn’t know what to talk about, she had told a story in which she’d felt rejected after saying something she thought was critical. “If I were in your shoes,” he proceeded, “I might wonder if I would be rejected for saying something critical here.” “I’m always worried I’m going to say the wrong thing,” Rachel said. Rachel and her therapist then returned to her initial uncertainty about what to talk about, explored her concerns about “saying the wrong thing” in therapy, and worked toward developing a clearer focus for her treatment. At the start of the next session, Rachel again said she didn’t know how to begin. However, perhaps reflecting a greater sense of safety in talking about her mental life with her therapist, she continued: “I guess I’ll talk about what we started to discuss the last time…” As patients with psychotic disorders may have difficulty with very basic aspects of mental state understanding (e.g., identifying their thoughts or feelings), complex interpretations about “deep” unconscious motivation, the connection between the remote past and the present, or even nonconscious phenomena are generally avoided and not given privileged status. Instead, a variety of mentalizing techniques are used to promote the awareness and understanding of mental states that can be most readily linked to a patient’s current subjective experience. Some of the core MBT interventions include: the use of short, simple (“soundbite”) observations; focusing on the patient’s mind (particularly affective experience), as opposed to behavior or physical/social circumstances; nonjudgmental active listening; questioning to provoke curiosity about motivations; a dogged determination to fully understand the patient’s point of view; praising positive mentalizing; or using the therapist’s mind as a model. Of particular importance is that any intervention should be tailored to the patient’s mentalizing capacity during a given therapy hour. For example, expressions of intense emotional arousal suggest that simpler, less complex interventions may be called for to support a patient’s sense of safety and avoid the breakdown of the ability to mentalize. We illustrate the use of some of these techniques in a vignette from the treatment of Rachel below: Shortly after beginning therapy, Rachel reported paranoid thoughts – e.g., feeling like she might be being secretly recorded. She said these thoughts didn’t last for very long, and she didn’t think they 22

were true. But they were bothering her. Therapist: Can you tell me a little more about these thoughts? When did you first notice having them? (Active questioning.) Rachel: Well, during the last hour. I remember just looking at how the walls in your office are undecorated, just plain white walls. (Example of concrete mentalizing.) I started thinking: Is this a real doctor’s office? And, then I wondered if you might be recording me… Therapist: I’m not sure that I understand how noticing the white walls in my office was connected with the thought that I was tape recording you. Did you have any particular feelings when you were looking at the walls here? (Focus on affect.) Rachel: (After a long pause.) When I look at the walls, it feels like you’re not really here, like you could be planning to leave any minute. Therapist: So, perhaps, when you feel that someone is going to leave you, you can start to worry about being hurt – like the idea that I’d record you without your permission? (Affect labeling with qualification.) Rachel: Yeah, I think that’s right. Therapist: How has it felt to talk about this just now? (Monitoring patient’s reaction to the process.) Rachel: It’s been okay. I feel more connected with you today. As highlighted in this example, by focusing on the patient’s inner experience, particularly in a specific social setting where the capacity for self and other understanding became lost (e.g., when Rachel felt increasingly disconnected from her therapist), MBT techniques offer a potentially valuable approach to the impairments of social understanding that are increasingly thought to contribute to the functional deficits and symptomatology of people with psychosis. Conclusion Here, we have presented an attachment-based model to support the adaptation of MBT to address impairments of mental state understanding in people with psychotic disorders. A number of case reports have provided evidence that conceptually related metacognitive psychotherapeutic techniques designed to foster self-understanding in schizophrenia may contribute to functional improvements in people with psychosis (8, 9,

Benjamin K. Brent Et al.

11, 12). Further, findings from randomized clinical trials have demonstrated significant long-term improvements in social function among patients with schizophrenia who received cognitive therapies incorporating techniques targeting social cognitive deficits, for example, impaired perspective-taking (33) or ToM (34). Taken together, these findings suggest that MBT techniques to promote the recovery of the capacity to mentalize have the potential to contribute to improvements in social function of people with psychosis, particularly among individuals with disturbances in the caregiving environment. The MBT therapeutic approach and treatment interventions outlined here share much in common with other established treatments for psychosis. For example, CBT (35) and metacognitive psychotherapy (8) for psychosis also emphasize: 1) the importance of taking a structured approach to patient’s psychological capacities; 2) the use of interventions that are simple and easy to understand; and 3) a focus on the patient’s current mental state, as opposed to explorations of the distant past. Some of the distinguishing features of the MBT model, however, include: 1) the focus on the patient’s state of mind as central to the rehabilitation of the capacity for social understanding; 2) the emphasis on the role of affect in disruptions of the ability to mentalize; and 3) the importance given to understanding the links between the quality of mentalization and specific interpersonal/ attachment contexts. The emphasis in MBT on the association between attachment-related dysfunction, the need for a compassionate stance to internal experience, and the recognition of impaired mentalization in people with psychosis is in no way intended to blame caregivers as the cause of these deficits. In the absence of an underlying genetic predisposition to psychosis, disturbances in child–caregiver relationships are highly unlikely to contribute to the development of psychotic symptoms. Further, in our view, the hypothesis that the quality of attachment relationships may be linked with the social understanding of people at genetic risk for psychosis is consistent with the more general evidence regarding the significant value and importance that relationships with caregivers have in the lives of people with psychosis (36). In particular, evidence regarding the beneficial effects of family-based interventions with respect to clinical course and social function in people with psychosis suggests that patients who feel better understood by their caregivers may have a greater capacity to think about themselves and to function in the world (37). Future research, however, is needed to

determine whether taking an MBT approach to deficits of social understanding will lead to improved clinical outcomes for patients with psychotic disorders. Acknowledgements We wish to thank Dr. Christopher Morse for his valuable comments and suggestions throughout the writing process.

References 1. Brent BK. Mentalization-based psychodynamic psychotherapy for psychosis. J Clin Psychol 2009; 65: 803-814. 2. Gumley A. The developmental roots of compromised mentalization in complex mental health disturbances of adulthood: An attachment-based conceptualization. In: Dimaggio G, Lysaker PH, editors. Metacognition and severe adult mental disorders. New York: Routledge, 2010. 3. Liotti G, Gumley A. An attachment perspective on schizophrenia: The role of disorganized attachment, dissociation and mentalization. In: Moskowitz A, Schäfer I, Dorahy MJ, editors. Psychosis, trauma and dissociation: Emerging perspectives on severe psychopathology. Chichester: John Wiley, 2008. 4. van Os J, Kenis G, Rutten BPF. The environment and schizophrenia. Nature 2010; 468: 203-212. 5. Harrington L, Siegert RJ, McClure J. Theory of mind in schizophrenia: A critical review. Cogn Neuropsychiatry 2005; 10: 249-286. 6. Couture SM, Penn DL, Roberts DL. The functional significance of social cognition in schizophrenia: A review. Schizophr Bull 2006; 32: S44-S63. 7. Fett AKJ, Viechtbauer W, Dominguez MG, Penn DL, van Os J, Krabbendam L. The relationship between neurocognition and social cognition with functional outcomes in schizophrenia: A meta-analysis. Neurosci Biobehav Rev 2011; 35: 573-588. 8. Lysaker PH, Gumley A, Dimaggio G. Metacognitive disturbances in persons with severe illness: Theory, correlates with psychopathology and models of psychotherapy. Psychol Psychother 2011; 84: 1-8. 9. Lysaker PH, Buck KD, Caricione A, Procacci M, Salvatore G, Nicolò G, Dimaggio G. Addressing metacognitive capacity for self reflection in the psychotherapy for schizophrenia: A conceptual model of the key tasks and processes. Psychol Psychother 2010; 84: 58-69. 10. Papeleontiou-Louca E. Metacognition and theory of mind Newcastle, U.K.: Cambridge Scholars, 2008. 11. Lysaker PH, Buck KD, Ringer J. The recovery of metacognitive capacity in schizophrenia across thirty two months of individual psychotherapy: A case study. Psychother Res 2007; 17: 713-720. 12. Salvatore G, Lysaker PH, Gumley A, Popolo R, Mari J, Dimaggio G. Out of illness experience: Metacognition-oriented therapy for promoting self-awareness in individuals with psychosis. Am J Psychother 2012; 66: 85-106. 13. Carpendale JIM, Lewis C. Constructing an understanding of mind: The development of children’s understanding within social interaction. Behav Brain Sci 2004; 27: 79-96. 14. Bowlby J. Attachment and loss, Vol. 1. Attachment. London: Hogarth Press and the Institute of Psycho-Analysis, 1969. 15. Sharp C, Fonagy P. Social cognition and attachment-related disorder. In: Sharp C, Fonagy P, Goodyer I, editors. Social cognition and developmental psychopathology. Oxford: Oxford University, 2008. 16. Ainsworth MDS. The development of infant-mother attachment. In: Caldwell BM, Ricciuti HN, editors. Review of child development research Chicago: University of Chicago, 1973: Vol. 3, pp. 1-94. 17. Fonagy P, Luyten P. A developmental, mentalization-based approach to the understanding and treatment of of borderline personality disorder. Dev Psychopathol 2009; 21: 1355-1381. 18. Frith CD. The cognitive neuropsychology of schizophrenia. Hove,

Mentalization-Based Treatment for Psychosis

England: Erlbaum, 1992. 19. Sprong M, Scothorst P, Vos E, Hox J, van Engeland H. Theory of mind in schizophrenia. Br J Psychiatry 2007; 191: 5-13. 20. Howes OD, McDonald C, Cannon M, Arseneault L, Boydell J, Murray RM. Pathways to schizophrenia: The impact of environmental factors. Int J Neuropsychopharmacol 2004; 7(S1): S7-S13. 21. Philips LJ, McGorry PD, Garner B, Thompson KN, Pantelis C, Wood SJ, Berger G. Stress, the hippocampus and the hypothalamic-pituitaryadrenal axis: Implications for the development of psychotic disorders. Aust N Z J Psychiatry 2006; 40: 725-741. 22. Wismer Fries AB, Ziegler TE, Kurian JR, Jacoris S, Pollak SD. Early experience in humans is associated with changes in neuropeptides critical for regulating social behavior. Proc Natl Acad Sci USA 2005; 102: 17237-17240. 23. Baskerville TA, Douglas AJ. Dopamine and oxytocin interactions underlying behaviors: Potential contributions to behavioral disorders. CNS Neurosci Ther 2010; 16: e92-e123. 24. Kapur S. Psychosis as a state of aberrant salience: A framework linking biology, phenomenology, and pharmacology in schizophrenia. Am J Psychiatry2003 ; 160: 12-23. 25. Fonagy, P, Gergely G, Jurist E, Target M. Affect regulation, mentalization, and the development of the self. New York, N.Y.: Other Press, 2002. 26. Hill EL, Frith U. Understanding autism: Insights from mind and brain. Philos Trans R Soc Lond B Biol Sci 2003; 358: 281-289. 27. Bateman A, Fonagy P. Mentalization-based treatment for borderline personality disorder: A practical guide. New York: Oxford University, 2006. 28. Sass LA, Parnas J. Schizophrenia, consciousness, and the self. Schizophr Bull 2003; 29: 427-444.

29. Ase E, Fonagy P. Mentalization-based family therapy. In: Bateman AW, Fonagy P, editors. Handbook of mentalizing in mental health practice. Washington DC: American Psychiatric, 2012: pp. 107-128. 30. Henriksen MG, Skodlar B, Sass LA, Parnas J. Autism and perplexity: A qualitative and theoretical study of basic subjective experiences in schizophrenia. Psychopathology 2010; 43: 357-368. 31. Moore EA, Green MJ, Carr VJ. Comorbid personality traits in schizophrenia: Prevalence and clinical characteristics. J Psychiatr Res 2012; 46: 353-359. 32. Bateman A, Fonagy P. 8-year follow-up of patients treated for borderline personality disorder: Mentalization-based treatment versus treatment as usual. Am J Psychiatry 2008; 165: 631-638. 33. Eack SM, Greenwald DP, Hogarty SS, Cooley SJ, DiBarry AL, Montrose DM, Keshavan MS. Cognitive enhancement therapy for early-course schizophrenia: Effects of a two-year randomized controlled trial. Psychiatr Serv 2009; 60: 1468-1476. 34. Subramaniam K, Luks TL, Fisher M, Simpson GV, Nagarajan S, Vinogradov S. Computerized cognitive training restores neural activity within the reality monitoring network in schizophrenia. Neuron 2012; 73: 842-853. 35. Kingdon D, Turkington GD. Cognitive psychotherapy of schizophrenia New York: Guilford, 2005. 36. McFarlane WR, Dixon L, Lukens E, Lucksted A. Family psychoeducation. and schizophrenia: A review of the literature. J Marital Fam Ther 2003; 29: 223-245. 37. McFarlane WR, Lukens EP. Insight, families, and education: An exploration of the role of attribution in clinical outcome. In: Amador XF, David AS, editors. Insight and psychosis: Awareness of illness in schizophrenia and related disorders. New York: Oxford University, 1998.

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Fabrice Berna et al.

Chronic Persecutory Delusion and Autobiographical Memories in Patients with Schizophrenia: A Diary Study Fabrice Berna, MD, PhD,1,2,3 Caroline Huron, MD, PhD,4 Mathilde Kazès, MD, PhD,5,6 Isabelle Offerlin-Meyer, PhD,1,3 Dominique Willard, MSc,5,6 Paulina Verry, MSc,1,3 Guy Hédelin, PhD,7 Marie-Odile Krebs, MD, PhD,5,6 and Jean-Marie Danion, MD1,2,3 1

Hôpitaux Universitaires de Strasbourg, Clinique Psychiatrique, Hôpital Civil, Strasbourg, France Université de Strasbourg, Faculté de Médecine, Strasbourg, France 3 INSERM U666, Physiopathologie et Psychopathologie Cognitive de la Schizophrénie, Hôpitaux Universitaires de Strasbourg, Strasbourg, France 4 INSERM U992, Cognitive Neuroimaging Unit, CEA Neurospin, Gif sur Yvette, France 5 Service Hospitalo-Universitaire, Hôpital Sainte-Anne, Paris, France 6 INSERM, Laboratoire de Physiopathologie des Maladies Psychiatriques, Centre de Psychiatrie et Neurosciences, Paris, France 7 INRS, Epidémiologie en Entreprise, Vandoeuvre, France 2

Abstract Background: While chronic persecutory delusions are typically anchored into patients’ everyday life situations, no investigation has ever looked at how situations associated with a feeling of persecution are recorded and later retrieved. Method: A diary methodology combined with a recognition task involving ten patients with schizophrenia who presented chronic persecutory delusions and ten control participants. Diaries of everyday persecutory events (PE) and non-persecutory events (NPE) were kept. Results: In both groups, 1) PE were associated with higher anxiety scores than NPE, 2) PE were experienced as less distinctive and more stereotyped than NPE, 3) the frequency of incorrect recognition of altered descriptions of PE was higher than that of NPE. Limitations: Because high levels of motivation are required of the diarists, our sample size was small. Conclusion: Memories of persecutory events were highly emotional and semanticized. They were frequently incorrectly recognized, suggesting the existence of bias resulting from interactions between their processing and persecutory delusions.

Introduction Metacognition refers to a large spectrum of cognitive processes ranging from noticing specific or discrete things about oneself to the integration of more complex perceptions or experiences into larger images of oneself and others (1). Metacognitive dysfunction can also lead to the formation and persistence of false or delusional beliefs (2) as observed in patients with schizophrenia who frequently present with persecutory delusions (3, 4). These patients exhibit both attentional and memory bias toward threatrelated information (5, 6). For instance, persecutory deluded individuals recalled more threat- and depression-related items than normal controls (5) and delusion-prone individuals were better able to recognize angry faces than nondelusion-prone individuals (7). Furthermore, individuals with delusions recalled more threatening propositions from stories heard previously than controls (8). Therefore, these studies showed that patients with delusion demonstrate a cognitive bias towards emotional themes congruent with their delusional beliefs (9). Finally, it has been shown that biases in probabilistic reasoning are involved in both the occurrence and persistence of persecutory delusions (10). Individuals with delusions have a tendency to “jump to conclusions” (11) and display alterations in attributional styles (12), a strong confirmatory reasoning bias (13) as well as a bias against disconfirmatory evidence (14). Interestingly, some of these reasoning biases were shown in both clinical and non-clinical populations (14), suggesting there

Address for Correspondence: Jean-Marie Danion, MD, Hôpitaux Universitaires de Strasbourg, Clinique Psychiatrique, Hôpital Civil, 1 Place de l’Hôpital, 67091 Strasbourg Cedex, France jean-marie.danion@chru-strasbourg.fr

Chronic Persecutory Delusion and Autobiographical Memories in Schizophrenia

are common processes underlying the formation of both suspiciousness and persecutory delusion. Nonetheless, studies focusing on delusion and memory all used episodic, non-ecological material, whereas persecutory delusions are anchored into everyday life situations that are far more complex and emotional than simple items. Therefore, a study of autobiographical events related to delusional beliefs would be of great interest for furthering our understanding of cognitive bias involved in delusions. It has been proposed that the initial stage of the formation of delusions corresponds to the assignment of salience to the external objects and internal representations of oneâ&#x20AC;&#x2122;s experience (15). Delusions are a cognitive effort on the part of the patient to make sense of these experiences that are deemed to be salient (16), associated with high levels of anxiety and interpreted for instance as persecutory if the patient has a tendency to be suspicious. If not challenged, this explanation provides a guiding cognitive scheme for further thoughts and events, which will be recorded as memories of delusional experiences in long-term autobiographical memory. The model recently put forward by Conway (17), which posits reciprocal relationships between autobiographical memories, personal beliefs and the self, provides a theoretical framework to understand the mechanisms underlying the formation and persistence of chronic delusional beliefs. In keeping with this model, repeated recording of these memories would made them less and less distinctive with time and leads to the semantization of their common component, i.e., the feeling of being the subject of persecution. The semantization process in memory leads memories to lose their uniqueness and to become summarized in the form of a more abstract and conceptual representation, that in this case, gradually becomes a persistent delusional belief. The final stage occurs when the delusional beliefs are integrated in the self as a delusional self. Like any other current aspect of the self, this delusional self is likely to be governed by the principle of coherence: personal memories have to be consistent with currents beliefs so that the self is maintained coherent and stable. For instance, the delusional self may favor interpreting personal experiences as persecutory ones and, hence, may contribute to a chronic course of delusional beliefs (18). Moreover, it may be responsible for the aforementioned memory biases: the encoding and retrieval of memories of delusional experiences may be facilitated in so far as their emotional and cognitive content is consonant with the delusional beliefs (confirmatory bias), whereas experiences dissonant with the delusional self 26

may be less easily encoded and retrieved (bias against non congruent knowledge). Finally, the delusional self may also lead to the formation of delusional memories at retrieval. Whereas memories of delusions can be defined as the accurate memories of deluded thoughts and experiences a person had while in a psychotic state, delusional memories are memories of past experiences that either never happened or that happened but which are then meaningfully re-interpreted as remembering in the context of current psychosis (19). This re-interpretation is probably controlled by current persecutory beliefs. In keeping with this general framework that postulates critical relationships between autobiographical memory and persecutory delusion, the aim of this study was to explore autobiographical memories of persecutory experiences in patients with schizophrenia suffering from chronic, stable persecutory delusions. More specifically, the first objective was to assess the cognitive and emotional features of these autobiographical memories. We predicted that, due the process of semantization, memories of persecutory experiences would be more schematized, which means less distinctive, than memories of other events without persecutory content. We also predicted that memories of persecutory experiences would be associated with a higher level of emotions such as anxiety (see 7). The second objective of this study was to assess the influence of these cognitive and emotional features on performance in a recognition task. We reasoned that abnormal recognition performance may result from bias related to interactions between autobiographical memory processing and delusional beliefs and therefore offer means to better understand the mechanisms underlying this bias. For that purpose, we used a diary methodology combined with a recognition task (20, 21), so that it was possible to control the creation of altered events and assess emotional and cognitive mechanisms involved in memory processing. Patients with schizophrenia who displayed persecutory delusions were asked to keep diaries of everyday persecutory events (PE) and everyday non-persecutory events (NPE), which they nonetheless deem to be memorable. They were compared with control participants who encountered daily situations where they thought they were the subject of malevolence on the part of others. In a recognition task two months later, participants were asked to discriminate between true and altered diary entries. In healthy participants, this method showed that the description of an experience acts only as a cue to memory construction (20). Discrimination between true and altered descriptions is achieved by accessing further autobiographical

Fabrice Berna et al.

knowledge that is then used to confirm or invalidate details of the description: a description, whether true or false, is accepted as true if this knowledge is consonant with the current configuration of the self and contains no evidence to contradict the description. Emotional and cognitive processes play here a critical role. A high level of emotions at retrieval tends to enhance the feeling of the event’s veracity and acts as a confirmatory bias. Moreover, the consonance of the general meaning of memories with beliefs may also act as a confirmatory bias. Finally, the propensity for altered events’ descriptions to be recognized as true may be due to poor access to evidence to reject the description. Therefore, we predicted that altered descriptions of PE would be more susceptible to be incorrectly recognized than those of NPE, due to the association of memories of PE with a high level of emotions and to the consonance of these memories with beliefs of malevolence (both acting as a confirmatory bias). We predicted also that PE would be less distinctive than NPE, i.e., that participants would have poorer access to invalidatory knowledge and lack autobiographical details to reject false descriptions of PE. Given that similar mechanisms have been postulated to account for the formation of false beliefs in healthy individuals (22) – an autobiographical knowledge consonant with the current self and pre-existing beliefs and a poor ability to judge the truth or falsity of this knowledge (20) – we expected to find similar result profiles in patients and comparison participants. Nevertheless, because patients with schizophrenia were expected to display pronounced semantization of PE as well as a dysregulation of emotional processes (23), this pattern of results was expected to be more pronounced in patients. Methods Participants

Ten outpatients (7 men, 3 women) recruited by the Psychiatry Department of the French University Hospitals of Strasbourg and Paris took part in the study (Table 1). They all met the DSM-IV-TR (24) criteria for schizophrenia (paranoid, n = 6; residual, n = 4), according to a consensus between the outpatients’ current psychiatrist and two senior psychiatrists in the research team. They were all clinically stabilized, but presented residual persecutory delusional ideas despite receiving appropriate neuroleptic treatment. Patients treated with benzodiazepines, antidepressants or lithium were excluded from the study. All of the patients were being treated with atypical neuroleptics.

The comparison group consisted of 10 subjects (7 men, 3 women) found on the basis of a psychiatric examination not to be suffering from any mental disorders. Neither patients nor comparison subjects had any history of traumatic brain injury, epilepsy, alcohol or substance abuse, or other neurological problems. The main criterion for including participants was their ability to describe daily situations associated with a feeling of malevolence from other people. All participants were systematically invited to take part in a 3-day test period before inclusion, so that we could check whether they really and habitually experienced this kind of situations. After an initial screening, 16 patients and 13 controls took part in the 3-day test period; 6 patients and 3 controls failed to pass this preliminary phase, and 10 participants were finally included in each group. Prior to giving their written informed consent, they received a full description of the study. The study itself was approved by the Faculty’s Ethics Committee. Procedure

The protocol was adapted from Perno-Marino et al. (21). The diary-recording phase spanned a period of two months. Recognition tests took place two months after completion of the last diary entry. There were two sessions of recognition tests, with a one-day interval between them. Clinical symptoms were assessed using the Positive and Negative Syndrome Scale (PANSS; 25) in patients. Both patients and controls completed the Peters et al.’s Delusion Inventory (PDI; 26) to assess their proneness to delusions. Psychometric and neuropsychological evaluations consisted of the French validated version of the National Adult Reading Test (f-NART; 27) for premorbid IQ, Wechsler Memory Scale (WMS-III; 28) for verbal and visual memory and the Trail Making Test (29) for executive functions. These evaluations were performed at the beginning of the studies. However, two patients refused to do them. Diary Records

Every day each participant made four diary entries relating to two categories of events: two events referred to situations where participants had “the feeling other people or the world surrounding them was against them or that someone was being malevolent with them” (persecutory events, PE), and two events referred to other noteworthy events that had occurred during the day, but without any feeling of malevolence (non-persecutory events, NPE). In a recording session, each entry was made on a page of a notebook given to the participants. On the back of the page, the participants were asked to rate each event using 5-point 27

Chronic Persecutory Delusion and Autobiographical Memories in Schizophrenia

scales to assess ten variables (20): three related to emotional aspects (Anxiety, Emotional Intensity, and Surprise associated with an event), six to other aspects known to determine the memorability of an event (Frequency, i.e., how frequently similar types of events had been experienced previously, Distinctiveness, i.e., how distinctive an event was in the context of that whole day’s activities, Change in Ongoing Activity, Personal Importance, Consequentiality, i.e., the implications of an event for ongoing activity, and Fit to Plans, i.e., the extent to which an event furthered current personal plans), and also the Conviction associated with the feeling of others’ malevolence for PE. After having completed each diary entry, participants were asked not to read their events again. Diaries were given to participants comprising 28 entries (i.e., one week diary). After the week-diary was completed, participants were required to give it back to the experimenter. Recognition test

We expected to obtain 244 event descriptions for each participant (122 in each category of events) and to use 160 of them in the recognition task (80 true events with 40 events per category and 80 altered events with 40 events per category). Four patients and two controls failed to manage 80 PE. Consequently, a total of 3,060 events were available for analysis (1,508 in patients and 1,552 in controls). Altered descriptions consisted of recorded events, one aspect of which had been modified by the experimenters. The modification affected either the event’s gist or its context, but in such a way that its meaning and plausibility were preserved. Importantly, both the emotional tone and core belief reflected in the event descriptions were always respected. All modifications were randomly performed and agreed between two experimenters (FB and CH). In each of the two successive recognition tasks, each participant was presented with a list of event descriptions. The descriptions themselves were presented at random on a sheet of paper, and each was studied separately. The participant began by reading each event aloud. Then he or she gave a “yes/no” recognition judgment, with “yes” denoting they deemed the event description to be true and “no” denoting they considered it had been altered. Confidence in the accuracy of this judgment was rated on a 3-point scale. In the case of descriptions judged to be true, participants were then asked to rate Imagery on a 3-point scale and were instructed to give a Remember, Know or Guess response according to whether each aspect of the event recalled was associated with conscious recollection, simply knowing, or guessing, respectively (21). A Remember 28

response was defined as the ability mentally to re-live certain specific aspects such as perceptions, thoughts or feelings that occurred or were experienced at the time of the event. A Know response denoted simply knowing the event had occurred, but without this knowledge being accompanied by any conscious recollection. A Guess response was for events that were neither consciously recollected nor simply known, but guessed. Finally, participants rated the same ten 5-point scales used during the diary period (Table 2). Several precautions were taken to ensure that patients and comparison subjects understood the procedure. All were given a printed set of instructions regarding possible responses (see 19). Statistical analysis Non-parametric statistical analyses were used for both direct group comparisons of clinical and neuropsychological data (Mann-Whitney U-test). For all analyses of memories, we used multilevel statistical analyses. These are particularly relevant for autobiographical memory studies because they allow memories to be treated as the statistical unit while taking into account the intra-subject variance, i.e., the fact that memories are not independent in one individual (30). The multilevel model assigned memories to level 1 and participants to level 2. All the following analyses were performed using the MLwiN software, version 2.10 (31). To study the proportion of correct and incorrect recognitions a logistic regression was used to explain the dependant variable Response type (Yes vs. No response), with Event type (True vs. Altered), Event category (PE vs. NPE) and Group (Patients vs. Comparison subjects) as predictor variables. Secondary analyses were performed only in events recognized by the participants as true (i.e., Yes responses), as a means of studying the state of subjective awareness at retrieval. A multivariate logistic regression was used to analyze the proportion of Remember/Know/Guess responses, using the same aforementioned grouping factors. An analysis restricted to altered descriptions was used to study the proportion of incorrect recognitions due to a modification of the gist or context of the events. A logistic regression was used to explain the dependant variable Response type (Yes vs. No) with Type of modification (Gist vs. Context), Event category and Group as predictor variables. Data from each rating scale were subjected to repeatedmeasures analyses of variance (ANOVAs), with Group, Event type and Event category as between factors and Time (Encoding vs. Retrieval) as a within factor. Data from

Fabrice Berna et al.

Conviction ratings were subjected to a similar repeatedmeasures ANOVA, with Group and Event type as between factors and Time as a within factor. Data from Imagery and Confidence ratings were subjected to separate multivariate logistic regressions with Group, Event type and Event Category as grouping factors. Results Characteristics of the participants (Table 1-2)

Patients and controls did not differ in age, level of education or IQ (all Ps > .15). Patientsâ&#x20AC;&#x2122; memory performance was poorer as shown by their significant lower percentage retention scores in both verbal and visual memory (all Ps

< .015). They also had lower executive performance as shown by their higher median reaction time at the TrailMaking Test (P = .051) but this failed to reach significance. Regarding the PDI Scale, patients rated more PDI items than control participants (U = 13.5, P = .01) and had higher subscores than controls for Distress and Preoccupation (U = 5, P = .001 and U = 7, P = .002, respectively) but not for Conviction (U = 27, P = .15). Analyzes of Yes/No and Remember/Know/Guess responses (Tables 3-4)

The proportion of Yes responses was higher in true than in altered event descriptions (P < 10-5) and in PE than in NPE (P = .002). The frequency of incorrect recognitions

Table 1. Clinical characteristics Controls M

Age (in years)

36.2

8.4

Education (in years)

11.9

3.4

PDI

Patients with schizophrenia

Range

to 50

36.3

7.5

to 19

11.7

3.9

U test N

to 51

0.11

0.91

to 17

0.26

0.79

9.70

3.20

to 14

13.56

2.01

to 16

-2.53

0.01

PDI-distress

20.00

11.72

to 42

46.89

12.89

to 64

-3.23

0.001

PDI-preoccupation

17.50

9.49

to 35

39.22

12.43

to 59

-3.06

0.002

PDI-conviction

25.30

11.55

to 41

40.22

18.34

to 71

-1.43

0.15

PANSS positive

20.50

5.04

to 26

PANSS negative

17.88

6.66

to 30

PANSS general

36.75

10.57

to 53

PANSS total

75.13

19.68

to 108

PANSS

PDI=Peters et al. Delusion Inventory; PANSS=Positive And Negative Symptom Scale.

Table 2. Neuropsychological characteristics Controls M

105.3

7.8

90.6

immediate recall total score*

0.27

0.86

learning curve*

0.10

percent retention scaled score*

-0.07

Patients with schizophrenia Range

Range

to 116.7

102.6

8.6

86.0

-1.00

to 1.33

-0.38

1.01

1.05

-1.67

to 1.33

-0.54

0.60

-1.00

to 0.67

-1.25

0.50

0.48

-0.67

to 1.00

TMT-A (median time)*

32.31

8.81

24.00

to 55.00

TMT B-A (median time)*

52.79

32.08

11.00

to 119.00

U test N

to 113.7

0.49

0.62

-2.00

to 0.67

1.20

0.23

0.94

-2.00

to 1.00

1.29

0.19

1.18

-3.33

to 0.67

2.44

0.015

-0.71

0.91

-1.67

to 0.67

2.49

0.01

37.46

5.17

28.00

to 43.00

-1.95

0.051

80.80

47.50

43.00

to 193.00

-1.51

0.13

Level of intelligence f-NART WMS-III Verbal Memory

WMS-III Visual Memory percent retention score* Trail-Making Test (TMT)

*z-scores, f-NART=french National Adult Reading Test; WMS=Wechsler Memory Scale; TMT-A=TMT part A; TMT B-A =time of part B minus time of part A

Chronic Persecutory Delusion and Autobiographical Memories in Schizophrenia

Table 3. Frequency of Yes vs. No responses in true and altered event descriptions True events

Altered events

Group

Category

Yes

Total

Patients with schizophrenia (n = 10)

94 (26.6%)

83 (23.4%)

54 (15.3%)

123 (34.7%)

354 (100.0%)

NPE

94 (23.5%)

106 (26.5%)

36 (9.0%)

164 (41.0%)

400 (100.0%)

Comparison subjects (n = 10)

85 (22.6%)

103 (27.4%)

38 (10.1%)

150 (39.9%)

376 (100.0%)

NPE

64 (16.0%)

136 (34.0%)

18 (4.5%)

182 (45.5%)

400 (100.0%)

PE=Persecutory Events, NPE=Non Persecutory Event

was higher in PE than in NPE, as shown by a significant interaction between Event type and Event category (P = .003). The frequency of incorrect recognitions was higher in patients than in controls participants but the interaction between Event type and Group factors was not significant (P = .12). Regarding the proportion of Remember/Know/Guess responses, a significant effect of Event type (P < 10-5) was found without any other main effects or interactions. Analyses performed on altered event descriptions showed a higher proportion of incorrect recognitions of events modified in terms of context than of events modified in terms of gist (P < 10-5). Ratings at encoding and retrieval (Table 5)

Neither significant effects of Group nor Event type were found for all variables. Regarding emotional factors, PE were associated in both groups with higher levels of emotions and anxiety than NPE, as indicated by higher Surprise and Anxiety scores in PE than NPE (Ps < .001). A significant effect of Time was found for Surprise with lower scores at retrieval than encoding (all Ps < .001), but not Anxiety. A significant Group by Event category interaction was also found for both Surprise and Anxiety (Ps < .001), the differences between PE and NPE being more pronounced in patients than in comparison participants. Finally, a significant Group by Time interaction was found for 30

Surprise and Anxiety (Ps < .001), both scores decreasing to a lesser extent in patients than in controls between encoding and retrieval. Taken together and according to our hypotheses, PE were more emotional than NPE and this was more pronounced in patients than in controls. The level of emotion remained high in patients at retrieval and did not decrease significantly contrary to controls. Regarding the other cognitive factors, PE were less distinctive than NPE, as indicated by lower Distinctiveness (P < .001). PE were also more frequent and less personally important than NPE, as indicated by higher Frequency and lower Personal Importance scores (Ps < .001); in addition, PE fitted in less with current plans and had fewer consequences on ongoing activities than NPE, as shown by lower Fit to Plan and Change on Ongoing Activities scores (all Ps < .001). These memory characteristics indicated that in both patients and controls, PE corresponded to more repetitive and more schematized events than NPE. Significant interactions between Group and Event category were found for all cognitive variables (Ps < .001) except for Distinctiveness (P > .10). The differences between PE and NPE were more pronounced in patients than in comparison participants for Personal Importance and Fit to Plan, whereas an opposite pattern was found for Change in Ongoing Activities and Frequency scores. Finally, a significant Group by Time interaction was found for all variables (Ps < .001), all scores decreasing to a lesser extent in patients than in controls between encoding and retrieval.

Fabrice Berna et al.

Table 5. Ratings of memories of PE and NPE at encoding and retrieval for patients with schizophrenia and comparison subjects* Non-Persecutory Events Patients’ memoriesa

Persecutory Events

Controls’ memoriesb

Patients’ memoriesc

Controls’memoriesd

Variable

Encoding M ± SD

Retrieval M ± SD

Encoding M ± SD

Retrieval M ± SD

Encoding M ± SD

Retrieval M ± SD

Encoding M ± SD

Retrieval M ± SD

Personal Importance

3.92 ± 1.18

3.19 ± 1.49

3.62 ± 1.10

2.13 ± 1.21

3.45 ± 1.37

3.05 ± 1.57

3.38 ± 1.12

1.98 ± 1.09

Emotional Intensity

3.24 ± 1.26

2.63 ± 1.54

3.12 ± 1.33

1.76 ± 0.89

3.25 ± 1.23

2.72 ± 1.53

3.12 ± 1.24

1.75 ± 0.91

Anxiety

1.64 ± 1.01

1.57 ± 1.11

1.67 ± 0.99

1.36 ± 0.64

2.85 ± 1.52

2.44 ± 1.65

2.25 ± 1.19

1.50 ± 0.87

Surprise

2.96 ± 1.40

2.12 ± 1.43

3.35 ± 1.36

1.85 ± 1.10

3.34 ± 1.38

2.42 ± 1.52

3.26 ± 1.33

1.89 ± 1.16

Consequentiality

2.81 ± 1.47

2.34 ± 1.49

2.78 ± 1.33

1.62 ± 1.00

2.72 ± 1.40

2.36 ± 1.45

2.99 ± 1.22

1.63 ± 0.94

Frequency

3.10 ± 1.34

2.78 ± 1.33

2.76 ± 1.09

2.01 ± 1.07

3.17 ± 1.22

2.91 ± 1.28

3.29 ± 1.11

2.44 ± 1.11

Distinctiveness

3.64 ± 1.24

3.18 ± 1.43

3.49 ± 1.14

2.64 ± 1.32

3.27 ± 1.31

3.10 ± 1.38

3.19 ± 1.13

2.52 ± 1.22

Change in Ongoing Activity

3.13 ± 1.48

2.89 ± 1.49

3.57 ± 1.27

2.68 ± 1.41

3.15 ± 1.43

2.76 ± 1.36

3.18 ± 1.29

2.44 ± 1.27

Fit to plans

4.00 ± 1.30

3.24 ± 1.55

3.60 ± 1.49

3.14 ± 1.61

1.94 ± 1.25

1.76 ± 1.18

1.86 ± 1.16

2.23 ± 1.51

3.74 ± 1.30

3.08 ± 1.57

3.27 ± 1.18

2.32 ± 1.44

Conviction Confidence

2.79 ± 0.50

2.78 ± 0.48

2.76 ± 0.53

2.74 ± 0.53

Imagery

2.51 ± 0.76

2.61 ± 0.68

2.41 ± 0.80

2.64 ± 0.67

n = 488. n = 431. n = 455. n = 454. *recognitions of true and altered events descriptions are presented together given that neither an effect of type nor an interaction between type and other factors were found.

Regarding Image and Confidence, both scores were higher for correct than incorrect recognition (Ps < .001). No other effect and no interaction were found (Ps > .10). Discussion To the best of our knowledge, this is the first study to explore persecutory delusion using a diary methodology. This methodology is unique in that it enables one to collect emotional and cognitive variables at the time of experience and later at remembering, to check the accuracy of memories and to assess the frequency of false recognitions of altered events descriptions. However, high levels of motivation are required of the diarists, during both the recording and remembering phases, explaining why diary studies have only a few participants (20) and why our sample was small. This is a limitation of the study, which was however partly offset by the high number of events recorded and the very strict multilevel statistical analysis. Using this methodology, three main results were obtained. Firstly, as expected, memories of PE were associated with higher levels of anxiety and were less distinctive than memories of NPE. Secondly, while the overall patterns of results were similar in patients and comparison participants, patients’ ratings of PE were disproportionately more emotional and stereotyped than the NPE with regards to controls. Thirdly, the frequency of incorrect recognition

of altered descriptions of PE was higher than those of NPE. Incorrect recognitions consisted mostly of events modified in terms of their context, suggesting no evidence of delusional memories in patients. Regarding the main features of autobiographical memories of persecutory experiences, the results showed higher levels of anxiety and surprise associated with PE than with NPE at the time of both encoding and retrieval and to a higher extent in patients. This confirms the hypothesis that memories of persecutory events are associated with high levels of emotions (7). The results also showed that PE were less distinctive than NPE, and were estimated as more frequent, less personally important, less integrated in current plans, and with fewer consequences on ongoing activities. These features are typical of memories of schematicized experiences (20) following a stereotypical sequence or schema. Some of them were more marked in patients than in comparison participants, confirming the hypothesis that the process of semantization related to chronic delusional beliefs is more pronounced in patients. An abnormal pattern of recognition performance, characterized by a higher frequency of incorrect recognition of altered descriptions of PE than those of NPE, was observed in both groups. This pattern may be taken as evidence of confirmatory bias related to interactions between the particular features of memories of PE and beliefs of malevolence. Because emotions associated with 31

Chronic Persecutory Delusion and Autobiographical Memories in Schizophrenia

memories of PE typically match current delusional beliefs in patients and possibly match interpretative thought patterns in control participants, they might have convinced the participants that the altered descriptions of PE were true. The high frequency of conscious recollection associated with PE may also have convinced them of the veracity of altered descriptions. The match between emotions and current delusional beliefs may be responsible for the enhanced ability to relive the past delusional experiences mentally. The interaction between memory processing and the emotional tone of delusion has been regarded as a critical contributing factor for delusion formation (9) and may be a mechanism underlying a confirmatory bias. Another possible mechanism underlying a confirmatory bias in patients may be related to the very content of memories of PE, which was consonant with the theme of their chronic delusional beliefs. This mechanism may have been reinforced by the semantization of these memories. Previous evidence shows that normal people incorrectly recognize altered events as true when they retrieve the general meaning of the original event but not the specific details (32). In this case, their recognition judgment is based on high conceptual similarity between the original event and its altered record. Therefore, the high level of stereotypy and, hence, of similarity of PE with delusional beliefs may have further convinced patients that the altered descriptions of PE were true. In addition to this mechanism and as a corollary, incorrect recognition may also reflect an inability to reject altered descriptions of PE. Because PE were less distinctive than NPE, autobiographical details may have been lacking for participants to invalidate these descriptions. Finally, since recognition may be influenced by the capacity of central control processes to judge the truth or falsity of past experiences, it is also plausible that incorrect recognition stemmed from a reduced capacity of these metacognitive processes to discriminate between true and altered descriptions of PE. Such a reduction has been described in normal participants as a consequence of stress. Delusion-prone individuals and patients with schizophrenia have an overconfidence in errors, possibly relating to a jumping-to-conclusions bias or a bias against disconfirmatory evidence (14), which is known to predispose to delusional beliefs (33). Taken together, these results are compatible with the hypothesis that memories of persecutory experiences are characterized by enhanced access to confirmatory knowledge (i.e. knowledge supporting the feeling of the eventâ&#x20AC;&#x2122;s veracity), causing participants to accept the altered descriptions as 32

true, and poor access to invalidatory knowledge, causing them not to reject the altered descriptions as false. These mechanisms may be combined to increase the frequency of incorrect recognition of altered PE, compared to NPE. However, it is worth mentioning that our protocol does not make it possible to assess a possible bias against disconfirmatory knowledge, which is postulated by Conwayâ&#x20AC;&#x2122;s model. Nevertheless, in keeping with this general framework, these results provide preliminary evidence that autobiographical memories of PE interact with delusional beliefs in such a way that the self is maintained stable and coherent. More generally, our results point out the role of metacognitive dysfunction in the formation of delusional beliefs, which affects the recognition of specific autobiographical memory elements, as well as the integration of a range of different psychological experiences into a larger representation of oneself and other. Finally, the study may provide new insight for the cognitive therapy of delusions (34), which aims to reduce confirmatory knowledge and increase invalidatory knowledge while working with patients on specific past situations associated with feeling of malevolence. As predicted, the finding of similar results patterns in patients with schizophrenia and comparison participants suggests there are common mechanisms underlying false beliefs in normal people and delusional beliefs in patients with schizophrenia. However, this finding merits further comment. Patients who took part in the study were able to write down in a diary daily events associated with a chronic feeling of persecution, suggesting they were able to step back somewhat from their symptoms and that they were not in a highly productive phase of persecutory delusions. Therefore, it is plausible that the impairments observed in highly motivated patients with residual symptoms would be even more severe in less motivated and/ or more symptomatic patients. On the other hand, the comparison group comprised healthy participants who were able to experience daily situations associated with a feeling of malevolence. Although they had no psychiatric history, their PDI-conviction ratings were similar to those of patients and six out of ten fell into the delusion-prone category. For these reasons, the possibility to extrapolate our results to the population of patients with chronic delusions usually met in daily practice but also to other healthy participants appears to be reduced. On the contrary, it may be also argued that with such participants it was difficult to detect a true difference between groups. Although this argument cannot be ruled out completely, it is worth mentioning that patients had impaired memory performance and rated the PE as disproportionately more

Fabrice Berna et al.

emotional and less distinctive than the NPE, this possibly reflecting substantial difference of the personal significance and clinical consequences of their persecutory beliefs. Patients also had higher scores of PDI, PDI-distress and PDI-preoccupation than controls. This is in keeping with previous studies, which have shown that delusion-prone individuals differed from patients with delusions in terms of their level of distress and preoccupation but not in terms of their conviction at the PDI (26, 35). The present results must therefore be considered preliminary, and further studies, particularly involving more patients, with more active delusions, are needed to replicate these findings. References 1. Dimaggio G, Salvatore G, Popolo R, Lysaker PH. Autobiographical memory and mentalizing impairment in personality disorders and schizophrenia: clinical and research implications. Front Psychol 2012; Nov 26;3:529. 2. Buck KD, Warman DM, Huddy V, Lysaker PH. The relationship of metacognition with jumping to conclusions among persons with schizophrenia spectrum disorders. Psychopathology 2012;45:271-275. 3. Blackwood NJ, Howard RJ, Bentall RP, Murray RM. Cognitive neuropsychiatric models of persecutory delusions. Am J Psychiatry 2001;158:527-539. 4. Freeman D. Suspicious minds: The psychology of persecutory delusions. Clin Psychol Rev 2007;27:425-457. 5. Bentall RP, Kaney S, Bowen-Jones K. Persecutory delusions and recall of threat-related, depression-related, and neutral words. Cognit Ther Res 1995;19:445-457. 6. Green MJ, Williams LM, Davidson DJ. Processing of threat-related affect is delayed in delusion-prone individuals. Br J Clin Psychol 2001;40: 157-165. 7. Larøi F, D’Argembeau A, Van der Linden M. The effects of angry and happy expressions on recognition memory for unfamiliar faces in delusion-prone individuals. J Behav Ther Exp Psychiatry 2006;37:271-282. 8. Kaney S, Bentall RP. Persecutory delusions and the self-serving bias. Evidence from a contingency judgment task. J Nerv Ment Dis 1992;180:773-780. 9. Gibbs AA, David AS. Delusion formation and insight in the context of affective disturbance. Epidemiol Psichiatr Soc 2003;12:167-174. 10. Bell V, Halligan PW, Ellis HD. Explaining delusions: A cognitive perspective. Trends Cogn Sci 2006;10:219-226. 11. Garety PA, Freeman D. Cognitive approaches to delusions: A critical review of theories and evidence. Br J Clin Psychol 1999;38:113-154. 12. Bentall RP, Kinderman P, Kaney S. The self, attributional processes and abnormal beliefs: towards a model of persecutory delusions. Behav Res Ther 1994;32:331-341. 13. Freeman D, Garety PA, McGuire P, Kuipers E. Developing a theoretical understanding of therapy techniques: An illustrative analogue study. Br J Clin Psychol 2005;44:241-254. 14. Woodward TS, Buchy L, Moritz S, Liotti M. A bias against disconfirmatory evidence is associated with delusion proneness in a nonclinical sample. Schizophr Bull 2007;33:1023-1028. 15. Broome MR, Woolley JB, Tabraham P, Johns LC, Bramon E, Murray GK,

et al. What causes the onset of psychosis? Schizophr Res 2005;79:23-34. 16. Maher BA. Delusions. Comprehensive handbook of psychopathology. In Adams HE, Patricia B. Sutker PB, editors. New York: Springer, 2001: pp. 309-340. 17. Conway MA. Memory and the self. J Mem Lang 2005;53:594-628. 18. Bentall RP. The paranoid self. In Kircher T, David A, editors. The self in neuroscience and psychiatry. Cambridge: Cambridge University, 2003: pp. 293-318. 19. Kopelman M, Guinan E, Lewis P. Delusional memory, confabulation, and frontal lobe dysfunction: A case study in De Clérambault’s syndrome. Neurocase 1995;1:71-77. 20. Conway MA, Collins AF, Gathercole SE, Anderson SJ. Recollections of true and false autobiographical memories. J Exp Psychol Gen 1996;125:69-95. 21. Pernot-Marino E, Schuster C, Hédelin G, Berna F, Zimmermann M-A, Danion J-M. True and false autobiographical memories in schizophrenia: Preliminary results of a diary study. Psychiatry Res 2010;179:1-5. 22. Gilbert DT. How mental systems believe. American Psychologist 1991;46:107-119. 23. Myin-Germeys I, Peeters F, Havermans R, Nicolson NA, DeVries MW, Delespaul P, et al. Emotional reactivity to daily life stress in psychosis and affective disorder: An experience sampling study. Acta Psychiatr Scand 2003;107:124-131. 24. American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. 4th ed. Washington, DC: American Psychiatric Association Press, 2000. 25. Kay SR, Flszbein A, Opfer LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 1987;13:261-276. 26. Peters E, Day S, McKenna J, Orbach G. Delusional ideation in religious and psychotic populations. Br J Clin Psychol 1999;38:83-96. 27. Mackinnon A, Mulligan R. [The estimation of premorbid intelligence levels in French speakers]. Encéphale 2005;31:31-43. 28. Wechsler D. Echelle clinique de mémoire de Wechsler MEM-III (WMSIII). Paris: Éditions du Centre de Psychologie Appliquée, 2001. 29. Reitan RM. Validity of the Trail Making Test as an indicator of organic brain damage. Percept Mot Skills 1958;8:271-276. 30. Wright DB. Modelling clustered data in autobiographical memory research: The multilevel approach. Appl Cogn Psychol 1998;12:339-357. 31. Rasbash J, Steele F, Browne W, Goldstein H. A user’s guide to MLwiN v2. 10. Bristol: Centre for Multilevel Modelling, 2009. 32. Barclay CR, Wellman HM. Accuracies and inaccuracies in autobiographical memories. J Mem Lang 1986;25:93-103. 33. Moritz S, Woodward TS. A generalized bias against disconfirmatory evidence in schizophrenia. Psychiatry Res 2006;142:157-165. 34. Nelson H. Cognitive-behavioural therapy with delusions & hallucinations: A practice manual. Cheltenham, UK: Nelson Thornes, 2005. 35. Johns LC, Van Os J. The continuity of psychotic experiences in the general population. Clin Psychol Rev 2001;21:1125-1141.

Appendix. Example of a patient’s diary entry with persecutory content: “This afternoon, I went to my sister’s place to make a phone call for a room reservation. Then I got on the bus to leave, when two girls also enter the bus. It seemed to me they were telling me to ‘get off the bus.’ I got annoyed and got off the bus.”

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Metacognition in Non-psychotic Help-seeking Adolescents: Associations with Prodromal Symptoms, Distress and Psychosocial Deterioration Ravit Scheyer, PhD,1 Noa Reznik, PhD,1 Merav Adres, PhD,1 Alan Apter, MD,2 Larry J. Seidman, PhD,3 and Danny Koren, PhD1,4 1

Psychology Department, University of Haifa, Haifa, Israel Psychological Medicine Clinic, Schneider Medical Center, Petach Tikva, Israel 3 The Massachusetts Mental Health Center, Public Psychiatry Division of the Beth Israel Deaconess Medical Center, Department of Psychiatry, Harvard Medical School, Boston, Massachusetts, U.S.A. 4 Psychiatry Division, Rambam Medical Center, Haifa, Israel 2

Abstract Objective: To explore the notion that difficulties in metacognitive functioning are a core pre-psychotic feature of emerging schizophrenia and its spectrum. Method: Seventy-eight help-seeking, non-psychotic adolescents (age 13-18) were assessed with the Prodromal Questionnaire (PQ), the Structured Interview for Prodromal Syndromes (SIPS), two scales of social and role functioning, and a metacognitive version of two non-social (verbal memory and executive functioning) and two social (facial emotion perception and Theory of Mind) cognition tasks. In addition to the standard administration of the tasks, subjects were also asked to rate their level of confidence in the correctness of each answer, and to choose whether they wanted it to be “counted” toward their overall performance score on the task. Each “volunteered” response received a bonus of 5 cents if correct, but an equal penalty if wrong. Results: Levels of cognitive and metacognitive functioning were not significantly different between subjects at high versus low risk. However, the prediction of psychosocial functioning reached significance when adding the novel metacognitive measures to the conventional measures of cognitive and social-cognitive abilities. Discussion: These results challenge the robustness of the link between neurocognitive deficits and risk for schizophrenia. However, they suggest that metacognition

Address for Correspondence: dkoren@psy.haifa.ac.il

plays an important moderating role in the association between neurocognition and functional outcome before acute onset of the illness.

Over the past two decades massive research has been invested into detection of pre-schizophrenic or early psychotic conditions. This research has been motivated by the assumption that intervention in the early illness stages before the illness becomes chronic and treatment-resistant may ameliorate long-term outcome. Because schizophrenia is practically unpredictable in its early so-called “premorbid” phases (due to its being a “rare epidemiological event”) (1), most research of the past 20 years has focused on the emergence of the psychotic clinical picture (often times referred to as the “prodrome”) in which anomalies of experience and behavioral deviations become more clearly articulated and “characteristic.” Findings from this line of research have shown that certain diagnostic signs (often referred to in the literature as “at-risk mental states”) in clinical and non-clinical populations can predict the transition to full-blown psychosis (2). In addition, findings from these studies have shown that those diagnostic signs can differentiate individuals who are at high risk to develop psychosis from other mentally ill individuals and from normal controls. Finally, data from these studies provides preliminary evidence for the effectiveness of certain interventions in delaying or preventing transition to psychosis, even

Danny Koren, PhD, Psychology Department, University of Haifa, Mount Carmel, Haifa 31905, Israel.

Ravit Scheyer et al.

though this evidence is rather inconclusive regarding the relative benefits of each specific intervention (3). A potential marker of early risk that has received a lot of attention in the above line of research is impaired neurocognition. This interest stems from the well-established finding that neurocognitive impairment is a fundamental and valid feature of schizophrenia and not just a transient, epiphenomenal effect of psychosis (4-6). Although many studies have shown that patients with schizophrenia displayed lower neurocognitive and social-cognitive (7) performance during the early pre-psychotic phase of their illness, there is evidence that substantial numbers of patients who develop schizophrenia exhibit no demonstrable cognitive deficits (8). Thus, it is still not clear whether cognitive impairment is a risk factor for the illness or a consequence of it. This question was recently addressed by Seidman et al. (9) in a large multi-site study (North American Prodrome Longitudinal Study), which examined the association of neuropsychological function with conversion to psychosis and family history of psychosis. Results from this study have shown that individuals at high risk, particularly those who later developed psychosis, exhibit moderate to severe neuropsychological difficulties depending on the type of neuropsychological functioning examined (9). However, these difficulties are lower than in first-episode schizophrenia. Similar patterns of findings have been found in a series of studies that examined social cognition in the prodrome. These studies found that while less severe than among first-episode patients, deficits in social cognition are quite prevalent among individuals at high risk for psychosis (10-12) and are good predictors of conversion to psychosis (13). Finally, a recent meta-analysis showed that both non-social and social neurocognitive impairments are more pronounced among those who convert to psychosis than those who do not (14). While no doubt useful in establishing that neurocognitive and social-cognitive deficits are a risk factor for schizophrenia, these findings leave open several important questions. First, the growing severity of cognitive impairment as full-blown psychosis nears leaves open the question of the degree to which neurocognitive deficits are a trait (i.e., a risk factor) or state (i.e., a consequence) characteristic of the illness. Second, they leave unanswered the question of the degree to which these deficits are specific to risk for psychosis. This is so because they do not include data about distressed individuals who might be at risk for other conditions. Finally, they do not provide information about the possible role that deficits at the metacognitive level

may play in increasing or decreasing the risk for psychosis. This lack of information is particularly important in light of recent studies including studies from our own group, which show that metacognition, assessed with both selfreport (15, 16) and behavioral measures (17-19), serves as a critical determinant of clinical and functional outcome in schizophrenia that is not reducible to traditional measures of neurocognitive functioning. In this study, we attempt to address the above limitations by comparing cognitive and metacognitive performance of adolescents who are at high-risk for psychosis with adolescents who are at high risk for other forms of psychopathology. Metacognition is an umbrella concept for a variety of “meta-level” abilities, such as those that can be described collectively as “knowing about knowing” (20). It is a term used to distinguish what one knows about one’s own cognitive abilities, states of knowledge, actual performance from the cognitive abilities, states of knowledge, and performance per se. It also includes the use of this “meta-level” knowledge to regulate one’s performance. The concept of metacognition has been employed by several cognitive and social-cognitive perspectives that have been applied to schizophrenia to study a variety of “meta-level” deficits such as self-reflectivity, perspective taking (theory of mind), and source-monitoring (21). While all of these models use “metacognition” as their conceptual cornerstones, the exact meaning attached to it within these models is quite different than the one used in this study. The first important difference is that our notion of metacognition refers to a global, overarching “meta-level” of performance, which monitors and controls (if allowed to do so) the correctness of performance in all types of cognitive or social-cognitive tasks including those just considered (i.e., tasks tapping self-reflection, theory of mind, source monitoring and so forth). Second, perhaps more importantly, from a methodological point of view, metacognition as conceptualized here cannot be examined with self-report questionnaires or “forced-response” tests that focus solely on input-bound performance (i.e., the percentage of input items that are answered correctly). Rather, its evaluation depends on incorporation of “free-response” tasks and output-bound measures (i.e., the percentage of answers chosen to be provided that are correct) into standard, baselevel testing procedures. A more elaborated discussion of these differences can be found in Koren et al. (18). Two important functions of metacognitive functioning, as it is used in this study, are monitoring and control. Monitoring is the subjective evaluation of one’s own cognitive functioning, and control is the manner in which 35

Metacognition in Non-psychotic Help-seeking Adolescents

one’s behavior is directed by this evaluation (18). These monitoring and controlling processes are critical enablers of real-world competency that are at least as important as the cognitive abilities they supervise. In other words, knowledge about what one does or does not know can be as important as what one actually knows. Metacognitive abilities have been suggested to vary independently of cognitive skills per se and to have important consequences over and above those skills (22). According to this argument, people with preserved metacognitive abilities will avoid a given situation when they lack the required cognitive skills to properly deal with it. In contrast, those with metacognitive deficits may not recognize that the situation demands higher abilities than can be brought to bear and thus may suffer negative consequences in the process of trying to deal with it. Although there have been recent advances in the study of metacognition in schizophrenia, substantial questions remain about the nature of these processes during the pre-psychotic and early phases of the illness (23). One important question to explore is whether metacognitive deterioration appears in the early stages of the development of schizophrenia or if it is a capacity that atrophies with the onset of this kind of illness either as a result of lack of use or some kind of brain damage. The main goal of the current study was to examine the notion that metacognitive deficits are a potential neurocognitive phenotype in the early, pre-psychotic stages of schizophrenia among a representative sample of help-seeking adolescents. More specifically, the goals of the study were: 1) to explore the relationship between cognitive and metacognitive deficits and commonly used markers of risk (e.g., subclinical psychotic symptoms and psychosocial functioning); and 2) to examine the added value of metacognition to cognitive abilities in predicting markers of risk for schizophrenia. Based on preliminary results as well as theoretical assumptions, we hypothesized that: 1) risk for psychosis, as defined by current markers of risk such as prodromal symptoms and psychosocial decline, will be associated with deficits at both levels of neurocognition (cognition and metacognition) and 2) the prediction of risk would be improved with the addition of metacognitive abilities to the cognitive skills alone. Method Participants

The target population for recruitment to this study was 114 non-psychotic adolescents (and their parents) who sought 36

help for a variety of emotional and behavioral difficulties in two clinical outpatient settings in Israel: the Medical Psychology Outpatient Clinic at Schneider Children’s Hospital in Petach Tikva and the Child and Adolescent Unit at Rambam Medical Center in Haifa. The recruitment process of the study sample consisted of the following steps. First, the intake psychiatrist informed all 114 adolescents and their parents about the study and asked their permission to be contacted directly by the project team. In addition, in order to rule out a potential association between willingness to participate and risk for psychosis, the intake psychiatrists filled out a brief checklist of worrying signs for future psychosis (e.g., decline in social and academic functioning, social withdrawal, changes in mood, difficulties in thinking and perceiving, family history of serious mental illnesses, and assessment of high risk for future psychotic illness). Each item was rated on a five-point scale, ranging from 1 = definitely not to 5 = definitely yes). Cronbach’s alpha for all 6 domains was 0.841. Level of risk (high vs. low) was determined according to the sixth item, whereby participants who were graded 4 or 5 on this item were defined as “high-risk” and all the rest were defined as “low-risk.” Next, those who agreed to participate in the study (n=78; 68.4%) were invited for a face-to-face interview in which their eligibility for the study was evaluated. Participants were included in the study if they met the following criteria: (1) age between 14 and 18 years, (2) fluency in Hebrew, (3) no known past or present psychotic episode as determined by a clinical psychiatric diagnostic assessment, (4) no history of severe head injury or organic brain disorder, and (5) no intellectual disability as defined by IQ lower than 70 (IQ was estimated with the Vocabulary and Block Design subtests from the WAIS-R). Even though some of the participants were taking anti-psychotic medication (see Table 1), we did not use it as an exclusion criterion because in all cases they were prescribed as off-label treatment of non-psychotic symptoms. All participants and their parents signed an informed consent which was reviewed and approved by the IRB committees of the two clinical settings. The final sample included 78 participants (76 from the Schneider’s Children Hospital, and two from the Rambam Medical Center), of whom 30 (38.5%) were rated by the intake psychiatrist as at-high worry and 46 (59.0%) as at-low worry for future psychosis (two additional participants were not assessed with respect to risk status). Level of willingness to participate in the study was slightly higher among help-seekers rated as at-high worry (73%) compared to help-seekers rated as at-low worry (63%). However, this difference was not statistically significant (χ2=1.9, p=0.19).

Ravit Scheyer et al.

Table 1. Socio-demographic, educational, and clinical characteristics of the study sample N

Mean (SD)

Gender (N) male/female

49 / 29

Median

Age (years)

15.8 (1.5)

Attends school (N) yes/no

68/9

Repeated a grade (N) yes/no

3/57

Discipline problems (N) Not at all / average / many

36/13/6

Socioeconomic level (N) low/average/high

3/45/5

Number of siblings

2.5 (1.4)

2.0

Order in relation to siblings

2.3 (1.3)

2.0

Grade Point Average in school

76.2 (16.7)

80.0

Number of close friends

3.6 (3.4)

3.0

Number of days a week meeting friends

3.2 (3.4)

2.0

Mother’s age

46.8 (6.2)

48.0

Father’s age

51.6 (5.7)

51.0

Mother’s years of education

14.4 (2.8)

14.0

Father’s years of education

14.4 (2.9)

14.5

Takes medication in the present (N) yes/no

36/19

Took medication in the past (N) yes/no

27/9

Family history of schizophrenia (N) yes/no

12/42

Depression and Anxiety (N) yes/no

17/61

ADHD (N) yes/no

12/66

OCD (N) yes/no

12/66

Other (N) yes/no

16/62

Similarly, there were no statistically significant differences between the two groups (i.e., at-high versus at-low worry for future psychosis) on any of the basic socio-demographic and/or clinical variables. The basic socio-demographic and clinical characteristics of the final sample appear in Table 1. Measures and procedure

Metacognition, the main variable in the study, was assessed by using the novel metacognitive approach to neuropsychological testing developed by our group (17) in two social and two non-social cognitive domains. The nonsocial domains were verbal memory, assessed with the Rey Auditory Verbal Learning Test (RAVLT; 24), and executive functioning, which was measured by the Wisconsin Card Sorting Test (WCST; 25). These domains were selected because they have been found to be impaired in both patients and individuals at risk of developing schizophrenia (26) and are also significantly correlated with psychosocial functioning (27). The two social domains were Theory of Mind (ToM), assessed with the Hinting Task paradigm

(28), and social perception, which was measured by the Face Emotion Identification Task (FEIT; 29) and the Face Emotion Discrimination Task (FEDT; 29). In accordance with our novel metacognitive approach, participants were also asked in all tasks: 1) to rate their level of confidence in the correctness of their response on a 0 (“just guessing”) to 100 (“completely confident”) scale, and 2) to decide whether they did or did not want to “venture” their response toward their overall performance score on the test. Each “ventured” response received a bonus of 0.2 shekel (approximately 5 cents) if correct, but an equal penalty if wrong. Thus, for each task, in addition to the standard “forced response” quantitative measure of the participant’s ability to perform the task (i.e., the proportion of correct responses out of all the trials), this procedure also yielded a measure of “free response” performance that was dependent on the participant’s metacognitive knowledge. The key metacognitive variables derived were: 1) accuracy score, which was defined as the proportion of correct responses out of the ventured responses; 2) monetary gains, which were calculated as the difference between the correct and incorrect ventured responses; 3) monitoring resolution, defined as the extent to which the confidence judgments distinguished between correct and incorrect responses, was calculated (with a Kruskal–Goodman gamma correlation) across all responses between the level of confidence and the correctness of the sort. This measure adds important information about the degree to which one is aware when one is correct and when one is incorrect, and therefore reflects the subjective ability to evaluate one’s own cognitive functioning; and 4) control sensitivity, defined as the degree to which the control process was dependent on the monitoring process, was calculated (with a gamma correlation) across all responses between the level of confidence and the decision to venture the response. This relationship reflects the manner in which one’s behavior is directed by one’s ability to evaluate one’s own cognitive functioning. Sub-clinical psychosis, the main outcome measure in the study, was assessed in two steps. The first screening instrument used was the Prodromal Questionnaire (PQ; 30), which is a 92-item self-report questionnaire targeting positive, negative, disorganized, and general symptoms (Cronbach’s alpha=0.949). The PQ was followed by a full Structured Interview for Prodromal Syndromes (SIPS; 31) for participants who scored above the cut-off point for a probable Prodromal Syndrome on the PQ (i.e., eight or more positive prodromal symptoms). This instrument is widely used in many high-risk centers around the world. The SIPS was administered by two interviewers, whose 37

Metacognition in Non-psychotic Help-seeking Adolescents

agreement with two gold-standard SIPS diagnoses during the training was in high range (kappa>0.75). The presence and severity of depressive and anxiety symptoms was assessed with the Mood and Anxiety Symptom Questionnaire (MASQ; 32, 33), which is a 61-item self-report questionnaire covering general distress anxiety, general distress depression, anxious arousal, and an-hedonic depression (Cronbach’s alpha = 0.928). Social and academic functioning were assessed with the Global Functioning Social (GFS) and the Global Functioning Role (GFR) Scales (34). These two scales were developed by Cornblatt and colleagues (34) to assess social and role functioning among individuals at high risk for schizophrenia. Both scales generate two scores: level of current functioning and level of functioning over the past year. In this study, each scale was rated by both the adolescents and the parents. Cronbach’s alpha was 0.781 for all items rated by the adolescents and 0.880 for the same items rated by the parents. Data analyses

The hypotheses regarding differences between the groups were tested using a series of t-tests for independent samples. In order to evaluate the unique and added contribution of metacognition to the prediction of clinical and functional outcome beyond that of the cognitive performance alone, three sets of hierarchical multivariate regression models (linear for continuous outcome variables and logistic for dichotomous outcome variables) were used: the first, included only the conventional cognitive variables as predictors; the second, only the novel metacognitive variables; and the third, both the cognitive and metacognitive variables. With a sample size of 78 participants and a total of 5 predictors, the achieved power to detect significant changes in R-square at a significance level of .05 in the present study was 0.72 for an effect size of 0.15 and 0.97 for an effect size of 0.30. All data analyses were conducted using the SPSS software for Windows.

Table 2. Means and standard deviations of cognitive performance in participants classified as prodromal vs. non-prodromal based on the SIPS NonProdromal N=39 Mean(SD)

Prodromal N=19 Mean(SD)

Significance test t(df), p

IQ Estimate

100.67 (12.11)

97.84 (9.79)

0.89(56), .38

Executive Functioning (WCST)

0.71 (0.14)

0.65 (0.14)

1.65(56), .11

Verbal Memory (RAVLT)

0.78 (0.10)

0.77 (0.09)

0.46(56), .65

Faces Task (FEIT + FEDT)

0.74 (0.15)

0.77 (0.11)

0.76(56), .45

ToM (Hinting task)

0.83 (0.15)

0.78 (0.17)

1.21(56), .23

General Quantity

0.75 (0.14)

0.75 (0.06)

0.92(56), .92

no significant differences for the cognitive measures, while mixed results were found for the metacognitive measures. As predicted, adolescents in the prodromal group showed lower levels of “free response” accuracy, albeit not always significantly, and earned less money in all tasks compared to those in the non-prodromal group. However, in contrast to our prediction for some of the tasks, the prodromal group actually showed higher levels of monitoring resolution (the extent to which confidence judgments distinguished between the correct and incorrect answers) and higher levels of control sensitivity (the degree to which the control process was dependent on the monitoring process) than the non-prodromal group (see Figure 1 for a summary of cognitive and metacognitive performance in the prodromal vs. the non-prodromal groups). Figure 1. Cognitive and metacognitive performance in prodromal vs. non-prodromal participants General

ToM

Facial fmotion task

Verbal Memory

Exceutive functioning 0 -0.02 -0.04 -0.06 -0.08

Results

-0.1 -0.12

Relationship between risk for psychosis and cognitive and metacognitive functioning

Table 2 presents the means and SDs of the IQ estimate and the cognitive measure from each task (i.e., quantity score) by level of risk for psychosis, and Table 3 presents the means and SDs of the metacognitive measures from each task (i.e., accuracy score, money units, monitoring resolution, and control sensitivity) by level of risk. As can be seen, there were 38

-0.14 -0.16 -0.18

Prodromal congnition (quantity) Prodromal metacognition (accuracy) Non Prodromal

Ravit Scheyer et al.

Table 3. Means and standard deviations of metacognitive performance in participants classified as prodromal vs. non-prodromal based on the SIPS Non-Prodromal N=39 Mean(SD)

Prodromal N=19 Mean(SD)

Significance test t(df), p

Table 4. R Squared obtained in sequential multiple regression analyses predicting psychosocial functioning, prodromal symptoms, and prodromal syndrome by the general cognitive and metacognitive abilities, and separately by the social cognitive and metacognitive abilities and the non-social cognitive and metacognitive abilities R2 Conventional cognitive predictors alone

Executive Functioning (WCST) Accuracy

0.71 (0.14)

0.69 (0.14)

0.48(56), .63

Money units

27.95 (16.49)

21.47 (17.45)

1.38(56), .17

Monitoring resolution

0.15 (0.33)

0.32 (0.28)

1.97(56), .05

Control sensitivity

0.41 (0.41)

0.49 (0.40)

0.73(56), .47

Accuracy

0.81 (0.11)

0.81 (0.09)

Money units

26.79 (8.90)

Monitoring resolution Control sensitivity

R2 All predictors from both domains a

General cognitive and metacognitive abilities Prodromal Syndrome – SIPSb

.00

0.09(56), .93

Subjects’ Ratings:

Cornblatt’s Social and Academic Scales

24.95 (8.11)

0.72(56), .48

Current social functioning

.02

.12

0.55 (0.25)

0.53 (0.23)

0.26(56), .80

Past social functioning

.12*

.21

0.55 (0.50)

0.70 (0.39)

1.14(56), .26

Current academic functioning

.00

.29**

.37**

0.18(55), .86

Past academic functioning

.00

.49**

.56**

.06

.18

Verbal Memory (RAVLT)

Faces Task (FEIT + FEDT) Emotion Identification (FEIT) Accuracy

R2 Novel metacognitive predictors alone

0.78 (0.08)

0.78 (0.11)

.15

.19

Money units

12.59 (4.15)

13.05 (5.23)

0.37(56), .72

Social cognitive and metacognitive abilities

Monitoring resolution

0.23 (0.14)

0.22 (0.18)

0.28(56), .78

.00

Control sensitivity

0.41 (0.21)

0.39 (0.20)

0.43(56), .67

Prodromal Syndrome – SIPSb Subjects’ Ratings:

Cornblatt’s Social and Academic Scales

Current social functioning

.00

.09

Past social functioning

.01

.10

.02

.24*

.37**

.02

.58**

ToM (Hinting task) Accuracy

0.87 (0.13)

0.80 (0.17)

1.77(56), .08

Money units

7.00 (2.32)

5.37 (3.04)

2.27(56), .03

Monitoring resolution

0.42 (0.64)

0.41 (0.55)

0.06(56), .95

Current academic functioning

Control sensitivity

0.26 (0.55)

0.42 (0.51)

1.10(56), .28

Past academic functioning

General Metacognition

Non-social cognitive and metacognitive abilities

Accuracy

0.77 (0.14)

0.77 (0.05)

0.04(56), .97

Money units

0.53 (0.11)

0.47 (0.11)

1.95(56), .06

Monitoring resolution

0.38 (0.20)

0.38 (0.17)

0.02(56), .99

Control sensitivity

0.46 (0.31)

0.51 (0.31)

0.65(56), .52

Concurrent prediction of risk for psychosis: Does it improve by assessment of metacognition

Table 4 presents the R squared from three sequential logistic regression models in which classification of prodromal syndrome was regressed on: the average across all tasks of the conventional cognitive measure alone (quantity score); the average across all tasks of the novel metacognitive measures alone (accuracy score, money units, monitoring resolution, and control sensitivity); and the cognitive and metacognitive measures together. As can be seen, cognition and metacognition, either alone or together, accounted for a relatively small amount of variance in prodromal status. However, in line with our hypothesis, the unique contribution of metacognition (15%) to the prediction of prodromal status was considerably larger than that

Prodromal Syndrome – SIPSb

.03

.08

Subjects’ Ratings:

Cornblatt’s Social and Academic Scales

Current social functioning

.02

.07

.08

Past social functioning

.21*

.24

Current academic functioning

.10*

.19

.25

Past academic functioning

.08

.23

.36*

Significance levels: *<.05, ** p<.01 a p values relate to significance of the R squared change by adding the metacognitive measure b Logistic regression predicting prodromal syndrome (yes/no)

of cognition (4%). Similar results were found when the classification of prodromal syndrome was regressed on the average across the social tasks alone (ToM and faces task). Table 4 also presents the R squared from a series of hierarchical linear regressions in which psychosocial functioning, prodromal symptoms, and mood and anxiety symptoms as continuous outcome variables were integrated: cognition alone, metacognition alone, 39

Metacognition in Non-psychotic Help-seeking Adolescents

and cognition and metacognition together. As can be seen, in accordance with our prediction, the addition of metacognition significantly improved the amount of variance accounted for in psychosocial functioning beyond that accounted for by cognitive abilities alone. Importantly, in most cases the added contribution of the metacognitive measures did not come at the expense of that of the conventional cognitive measures. On the contrary, it actually enhanced it in the joint model. This was evidenced by the total R squared of the overall model that was greater than the sum of the R squared from both models separately. On the other hand, contrary to our prediction, metacognition did not improve the amount of variance accounted for in prodromal symptoms or mood and anxiety symptoms. Again, similar results were found for only the social tasks, but not for the non-social tasks. Finally, the same set of three linear regression models was repeated for each task alone. When each task was examined separately, metacognition in the scope of executive functioning (WCST) was found to improve the prediction of negative symptoms and symptoms of distress and anxiety beyond that of cognitive abilities alone. In addition, metacognition in the scope of verbal memory (RAVLT) was found to improve the prediction of general distress and general anxiety symptoms. Discussion The main goal of the present study was to explore the notion that metacognitive deficits are a core neurocognitive marker of risk for schizophrenia. To accomplish this goal, we examined the relationship between cognition, metacognition, and other currently accepted markers of risk, such as sub-clinical psychotic symptoms and deterioration in psychosocial functioning. To our knowledge, this is the first study to examine metacognition in the prodromal phase using a behavioral, laboratory-based method for assessment of metacognition. The study results provided mixed and inconclusive support for the notion that impaired metacognition is a marker of risk for schizophrenia. On the one hand, they did demonstrate that metacognition can significantly improve the concurrent prediction of psychosocial functioning in the prodrome beyond what could be explained by conventional cognitive measures alone. On the other hand, no significant associations were found between cognition, metacognition, and sub-clinical psychotic symptoms. Moreover, on some tasks, the direction of the association between metacognition and sub-clinical 40

psychosis was in the opposite direction from our hypothesis. That is, it was higher, though not significantly so, among adolescents with sub-clinical psychotic symptoms (i.e., “prodromal”) than among adolescents without subclinical psychotic symptoms (i.e., “non-prodromal”). The present results are not necessarily in disagreement with those of previous studies, which showed a strong link between the risk for psychosis and neurocognitive deficits (9, 35). However, since level of cognitive and metacognitive performance was similar among helpseeking adolescents who are at high-risk for schizophrenia and those who might be at high-risk for other disorders, they do call into question the specificity of this link to risk for schizophrenia alone. In addition, they suggest that metacognition plays an important moderating role between cognition and psychosocial functioning not only after, but also before, the acute onset of the illness. This moderating effect can take place either in the form of compensation for poor cognitive abilities (i.e., correctly recognizing poor performance) or in the form of not capitalizing on good cognitive abilities (i.e., failing to recognize good performance). Finally, the greater variability of metacognitive functioning among prodromal adolescents suggests that metacognition may be more sensitive to situational factors (e.g., environmental stress, treatment, etc.) and hence may serve as a better predictor of further deterioration or improvement of clinical symptoms. The following sections elaborate on each of the main findings of the study, followed by a discussion of the limitations of the current study and suggested directions for future studies. Relationships between cognitive and metacognitive functioning and other known risk factors for future psychosis

The results were mixed with regard to the question of whether cognitive and metacognitive deficits act as risk factors for developing schizophrenia. On the one hand, cognitive and metacognitive deficits were found to be more impaired in the high-risk / prodromal group than in the low-risk / non-prodromal group on most tasks and measures. However, for some of the tasks, the high-risk / prodromal group showed higher levels of monitoring resolution (the extent to which confidence judgments distinguished between the correct and incorrect answers) and control sensitivity (the degree to which the control process was dependent on the monitoring process) than the low-risk / non-prodromal group. One possible explanation of these surprising results

Ravit Scheyer et al.

is that risk for psychosis is associated with better, rather than worse, metacognitive abilities. However, this possibility does not seem likely in light of the lower accuracy scores and smaller amounts of money earned by the highrisk individuals. A second possible explanation is that these results are a positive artifact of a negative tendency among high-risk individuals to feel less confident in their responses and as a result be more passive in their inclination to venture them than other help-seeking adolescents (note that in certain lacking conditions, chronic lack of confidence and passivity can be an advantage). Support for this possibility can be found in the fact that adolescents at high risk reported lower levels of confidence in the correctness of their answers (independent of their actual correctness) and ventured fewer answers than other help-seeking adolescents. Concurrent prediction of risk by cognitive and metacognitive functioning

As predicted, adding the assessment of metacognition improved the power to explain psychosocial functioning over and above that of cognitive performance alone. These results suggest that metacognition enhances the utility of neuropsychological skills in predicting functional outcome before the onset of schizophrenia. However, in contrast to our hypothesis, metacognition did not increase the amount of variance counted for in prodromal status by cognitive performance alone. In fact, there were no differences between prodromal and non-prodromal adolescents in their metacognitive performance. these results appear to be inconsistent with those of Seidman et al. (9), which showed that neuropsychological impairments do predict later psychosis among people who are considered to be at clinical high risk. However, it should be recalled that the findings of Seidman and his colleagues were based on a comparison of high-risk patients to healthy controls, whereas the present study used a comparison to treatment-seeking individuals. In the absence of a comparison group of healthy controls, the possibility of obtaining similar results in the present study cannot be ruled out. Lastly, results from the study showed that metacognition can significantly improve the concurrent prediction of psychosocial functioning, but not prodromal symptoms. One possible explanation for these somewhat puzzling results is that metacognitive deficits are related only to psychosocial functioning but not to clinical symptoms. Another possible explanation is that deficits at the metacognitive level are more responsive to environmental factors, such as stress

and/or therapy, than deficits at the cognitive level. It should be noted that most of the adolescents in our sample were in some kind of treatment during the time of the study. Indirect support for this possibility can be found in recent studies which showed that metacognitive performance, as opposed to neurocognitive performance, can be improved following therapy, while cognitive deficits were found to be more stable (36, 37). Another possible explanation is that metacognitive deficits are more specific to schizophrenia-spectrum disorders and as such do not correlate with prodromal symptoms for the entire sample of adolescents at risk, but rather only for a small percentage of adolescents who will eventually develop the disorder. Therefore, it was also important to examine whether adolescents defined as prodromal demonstrated lower levels of cognitive and metacognitive performance relative to other helpseeking adolescents. Our results showed no significant differences for cognitive performance but some measures of metacognitive performance were found to be lower in the prodromal group (though these differences did not reach significance), providing partial support for the notion that metacognitive deficits are more specific to schizophrenia. Finally, another more intriguing explanation of these results is that intact metacognitive functioning acts as a protective factor which lowers the risk for developing psychosis. This explanation is based on the fact that only a small portion of adolescents defined as â&#x20AC;&#x153;at riskâ&#x20AC;? will eventually develop the illness. If this is true, then additional follow-up of this group of adolescents should reveal that metacognitive deficits characterize only this small portion, whereas most of the adolescents at risk are characterized by intact metacognition. Confirmation of this interesting hypothesis would enable the use of metacognitive deficits to help identify a smaller group of adolescents at higher risk for developing schizophrenia. For this small group, specific early intervention programs may be useful. These programs may include psychotherapy that targets the capacity for metacognition or thinking about thinking, which may assist persons at high risk to become able to think about themselves and others in a generally more complex and flexible manner (37). This way the prognosis may improve and also lower the severity of the deficits in the psychosocial functioning. However, treatment approaches including metacognitive training are still in their infancy. Future work is needed to study the etiology of deficits in discrete and synthetic metacognition, as well as their overlap 41

Metacognition in Non-psychotic Help-seeking Adolescents

with related constructs such as mentalization and social cognition. Strengths and Limitations of the Study

An important strength of the present study lies in the “non-enriched” nature of our sample. Unlike previous studies that used highly “enriched” samples (i.e., samples with high numbers of adolescents specifically at risk for schizophrenia-spectrum disorders), the selection of participants for the present study was not guided by their initial level of risk for future psychosis. Consequently, the selection process allows a rather secure extrapolation of the study findings to replicable samples of help-seeking adolescents in the community. However, the current study was located within a specific clinical context, thereby reducing its representativeness and generalizability to other populations. Future research should aim to identify adolescents at high risk for developing schizophrenia in a community sample, using school counseling systems or other community services that have access to adolescents suffering from more divergent kinds of problems. The study also has a few important limitations that can guide future research. First, and most importantly, the study lacks follow-up data on the clinical and functional status of the participants over time. This limits the ability to draw conclusions about the predictive value of metacognitive functioning for future clinical, as well as sub-clinical, schizophrenia-spectrum symptoms. Further longitudinal research is needed in order to establish metacognitive deficits as a potent marker of vulnerability for psychosis. Second, the study lacks comparison data on cognitive and metacognitive functioning among age-matched non-help-seeking adolescents. This limits the ability to draw conclusions about the degree to which the two groups of patients differ from the normal control base. Thus, future studies should include a comparison group of non-help-seeking adolescents. Finally, another limitation of the current study is its relatively small sample size. Consequently, the results regarding the regression coefficients should be interpreted with caution. Conclusions and Future Directions In conclusion, the present results provide preliminary concurrent support for the notion that metacognition plays an important moderating role between cognition and psychosocial functioning before the acute onset of the illness. Therefore, it appears that psychosocial function42

ing depends not only on how much one knows, but also on how much one can trust this knowledge. In addition to the theoretical implications of the present data, they provide preliminary support for enriching current early detection models of neurocognitive vulnerability to psychosis with assessment of metacognitive deficits. Finally, the study provides the rationale for further prospective investigation of the longitudinal relationship between metacognitive deficits, prodromal symptoms, and the risk for schizophrenia-spectrum conditions. Acknowledgements The authors wish to thank Tout Naaman, Osnat Ziv, Shahar Iger, Assaf Gayer, and Yuval Tal for their help with subject recruitment and data collection and coding. In addition, the authors wish to thank Netta Tzemach for her help with editing this manuscript. Finally, the authors wish to thank the three anonymous reviewers of this paper for their excellent comments and suggestions.

References 1. Warner R. Problems with early and very early intervention in psychosis. Br J Psychiatry 2005; 48: 104-107. 2. McGlashan TH, Johannessen JO. Early detection and intervention with schizophrenia: Rationale. Schizophr Bull 1996; 22: 201-222. 3. Stafford MR, Jackson H, Mayo-Wilson E, Morrison AP, Kendall T. Early interventions to prevent psychosis: Systematic review and meta-analysis. BMJ 2013;346: 185. 4. Seidman LJ, Buka SL, Goldstein JM, Tsuang MT. Intellectual decline in schizophrenia: Evidence from a prospective birth cohort 28 year follow-up study. J Clin Exp Neuropsychol 2006; 28: 225-242. 5. Seidman LJ, Cherkerzian S, Goldstein JM, Agnew-Blais J, Tsuang MT, Buka SL. Neuropsychological performance and family history in children at age 7 who develop adult schizophrenia or bipolar psychosis in the New England family studies. Psychol Med 2013; 43: 119-131. 6. Niendam T, Bearden C, Rosso I, Sanchez LE, Hadley T, Nuechterlein KH, Cannon TD. A prospective study of childhood neurocognitive functioning in schizophrenic patients and their siblings. Am J Psychiatry 2003; 160: 2060-2062. 7. Couture SM, Penn DL, Roberts DL. The functional significance of social cognition in schizophrenia: A review. Schizophr Bull 2006; 32: S44-S63. 8. Palmer BW, Heaton RK, Paulsen JS, Kuck J, Braff D, Harris MJ, Zisook S, Jeste DV. Is it possible to be schizophrenic yet neuropsychologically normal? Neuropsychology 1997; 11: 437-446. 9. Seidman LJ, Guiliano AJ, Meyer EC, Addington J, Cadenhead KS, Cannon TD, McGlashan TH, Perkins DO, Tsuang MT, Walker EF, Woods SW, Bearden CE, Christensen BK, Hawkins K, Heaton R, Keefe RSE, Heissen R, Cornblatt BA. Neuropsychology of the prodrome to psychosis in the NAPLS consortium. Arch Gen Psychiatry 2010; 67: 578-588. 10. Amminger GP, Schafer MR, Papageorgiou K, Klier CM, Schlogelhofer M, Mossaheb N, Werneck-Rohrer S, Nelson B, McGorry PD. Emotion recognition in individuals at clinical high-risk for schizophrenia. Schizophr Bull 2012; 38: 1030-1039. 11. Thompson A, Papas A, Bartholomeusz C, Allott K, Amminger GP, Nelson B, Wood S, Yung A. Social cognition in clinical “at risk” for psychosis and first episode psychosis populations. Schizophr Res 2012; 141: 204-209. 12. Roddy S, Tiedt L, Kelleher I, Clarke MC, Murphy J, Rawdon C, Roche RAP, Calkins ME, Richard JA, Kohler CG, Cannon M. Facial emotion recognition in adolescents with psychotic-like experiences: A schoolbased sample from the general population. Psychol Med 2012; 42: 2157-2166.

Ravit Scheyer et al.

13. Kim HS, Shin NY, Jang JH, Kim E, Shim G, Park HY, Hong KS, Kwon JS. Social cognition and neurocognition as predictors of conversion to psychosis in individuals at ultra-high risk. Schizophr Res 2011; 130: 170-175. 14. Fett AKJ, Viechtbauer W, Dominguez M, Penn DL, van Os J, Krabbendam L. The relationship between neurocognition and social cognition with functional outcomes in schizophrenia: A meta-analysis. Neurosci Biobehav Rev 2011; 35: 573-588. 15. Lysaker PH, Dimaggio G, Buck KD, Callaway SS, Salvatore G, Carcione A, Nicolò G, Stanghellini G. Poor insight in schizophrenia: Links between different forms of metacognition with awareness of symptoms, treatment need, and consequences of illness. Compr Psychiatry 2011; 52: 253-260. 16. Lysaker PH, Shea AM, Buck KD, Dimaggio G, Nicolo G, Procacci M, Salvatore G, Rand KL. Metacognition as a mediator of the effects of impairments in neurocognition on social function in schizophrenia spectrum disorders. Acta Psychiatr Scand 2010; 122: 405-413. 17. Koren D, Seidman LJ, Poyurovsky M, Goldsmith M, Viksman P, Zichel S, Klein E. The neuropsychological basis of insight in first-episode schizophrenia: A pilot metacognitive study. Schizophr Res 2004; 70: 195-202. 18. Koren D, Poyurovsky M, Seidman LJ, Goldsmith M, Wegner S, Klein E. The neuropsychological basis of competence to consent in firstepisode schizophrenia: A pilot metacognitive study. Biol Psychiatry 2005; 57: 609-616. 19. Koren D, Seidman LJ, Goldsmith M, Harvey PD. Real-world cognitive and metacognitive dysfunction in schizophrenia: A new approach for measuring (and remediating) more “right stuff.” Schizophr Bull 2006; 32: 310-326. 20. Shimamura AP, Metcalfe J, eds. Metacognition: Knowing about knowing. Cambridge, Mass.: MIT, 1994. 21. Lysaker PH, Vohs JL, Ballard R, Fogley R, Salvatore G, Popolo R, Dimaggio G. Metacognition, self-reflection and recovery in schizophrenia. Future Neurol 2013; 8: 103-115. 22. Koriat A, Goldsmith M. Monitoring and control processes in the strategic regulation of memory accuracy. Psychol Rev 1996; 103: 490-517. 23. Keshavan MS, Rabinowitz J, DeSmedt G, et al. Correlates of insight in first episode psychosis. Schizophr Res 2004; 70:187-194. 24. Lezak MD. Neuropsychological assessment. Oxford: University Press, 1995. 25. Berg EA. A simple objective technique for measuring flexibility in thinking. J Gen Psychol 1948; 39: 15-22.

26. Morey RA, Inan ST, Mitchell TV, Perkins DO, Lieberman JA, Belger A. Imaging frontostriatal function in ultra-high-risk, early, and chronic schizophrenia during executive processing. Arch Gen Psychiatry 2005; 62: 254-262. 27. Green MF. What are the functional consequences of neurocognitive deficits in schizophrenia? Am J Psychiatry 1996; 153: 321-330. 28. Corcoran R, Mercer G, Frith DD. Schizophrenia, symptomatology and social inference: Investigating “theory of mind” in people with schizophrenia. Schizophr Res 1995; 17: 5-13. 29. Kerr SL, Neale JM. Emotion perception in schizophrenia: Specific deficit or further evidence of generalized poor performance? J Abnorm Psychol 1993; 102: 312-318. 30. Loewy RL, Bearden CE, Johnson JK, Raine A, Cannon TD. The prodromal questionnaire (PQ): Preliminary validation of a self-report screening measure for prodromal and psychotic syndromes. Schizophr Res 2005; 79: 117-125. 31. Miller TJ, McGlashan TH, Rosen JL, Cadenhead K, Cannon T, Ventura J, McFarlane W, Perkins DO, Pearlson GD, Woods SW. Prodromal assessment with the structured interview for prodromal syndromes and the scale of prodromal symptoms: Predictive validity, interrater reliability, and training to reliability. Schizophr Bull 2003; 29: 703-715. 32. Watson D, Clark L. The Mood and Anxiety Symptoms Questionnaire. Iowa City: University of Iowa Department of Psychology, 1991. 33. Watson D, Weber K, Assenheimer JS, Clark LA, Strauss ME, McCormick RA. Testing a tripartite model: I. Evaluating the convergent and discriminant validity of anxiety and depression symptom scales. J Abnorm Psychol 1995; 104: 3-14. 34. Cornblatt BA, Auther AM, Neidam T, Smith CW, Zinberg J, Bearden CE, Cannon TD. Preliminary findings for two new measures of social and role functioning in the prodromal phase of schizophrenia. Schizophr Bull 2007; 33: 688-702. 35. Giuliano AJ, Li H, Mesholam-Gately RI, Sorenson SM, Woodberry KA, Seidman LJ. Neurocognition in the psychosis risk syndrome: A quantitative and qualitative review. Curr Pharm Des 2012; 18: 399-415. 36. Kalmus E. Insight and metacognition in schizophrenia: A comparative study of patients-siblings paired samples. Thesis dissertation. University of Haifa, 2004. 37. Lysaker PH, Glynn SM, Wilkness SM, Silverstein SM. Psychotherapy and recovery from schizophrenia: A review of potential application and need for future study. Psychol Serv 2010; 7: 75-91.

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Metacognition in Schizophrenia and Schizotypy: Relation to Symptoms of Schizophrenia, Traits of Schizotypy and Social Quality of Life Shai Joseph Rabin, MA,1 Ilanit Hasson-Ohayon, PhD,1 Moran Avidan, MA,1 Silvio Rozencwaig, MD,2 Hadar Shalev, MD,3 and Shlomo Kravetz, PhD1 1

Department of Psychology, Bar-Ilan University, Ramat Gan, Israel Day Care and Rehabilitation Unit, Beer Yaakov Mental Health Center, Beer Yaakov, Israel 3 Psychiatric Department, Soroka University Medical Center, Beer Sheva, Israel 2

Abstract Objective: The current study examined a mediation model in which symptoms of schizophrenia and schizotypy traits mediate the positive relations between metacognition and Social Quality of Life (SQoL) among persons with schizophrenia and persons without mental illness. Method: 39 persons diagnosed with schizophrenia and 60 persons without a severe mental illness diagnosis participated in this study. Instruments included the Metacognition Assessment Scale-Abbreviated (MAS-A), the SQoL scale of the QLI-MH, the PANSS scale and the O-LIFE self-report questionnaire that assesses schizotypy traits. Results: Persons with schizophrenia exhibit lower SQoL and metacognitive abilities than persons without mental illness. For persons with schizophrenia, negative symptoms mediate the positive relation between the ability to understand other persons’ minds and SQoL. However, although for persons without mental illness, understanding other minds was found to correlate negatively with introvertive anhedonia and SQoL, a mediation model was not confirmed. Discussion: Understanding of others’ minds seems relevant to the SQoL for both samples. In addition, negative symptoms of schizophrenia and introvertive anedonia traits are also related to SQoL among persons with schizophrenia and among persons without mental

illness respectively. The lack of support for a mediation model for persons without mental illness is consistent with the theories that claim schizotypy is not a mirror image of schizophrenia and, therefore, may not necessarily lead to schizophrenia. Limitations of this study and suggestions for future research are discussed.

Introduction Deficits in social functioning have frequently been found to be associated with schizophrenia. Consequently, they are included in the list of the diagnostic signs of the illness (1, 2). These deficits involve difficulties in forming and preserving social interactions (3) and are thought to contribute significantly to the poor quality of life of persons with schizophrenia (4). The positive and negative symptoms of schizophrenia have also been found to impact negatively on the quality of life of persons with schizophrenia (5). Recent theories of schizophrenia attribute both positive symptoms such as delusions and hallucinations and negative symptoms of schizophrenia such as anhedonia and apathy, as well as the reduction in social functioning and quality of life associated with these symptoms, to metacognitive limitations that are considered a major aspect of schizophrenia (6-9). The present study attempted to assess the process by which metacognition and symptoms contribute to social quality of life (SQoL) of persons with schizophrenia. In

Address for Correspondence: Ilanit Hasson-Ohayon, PhD, Department of Psychology, Bar-Ilan University, Ramat Gan, Israel ilanit.hasson-ohayon@biu.ac.il

Shai Joseph Rabin Et al.

particular, it combined the Indiana Psychiatric Illness Interview (IPII; 10) with the Metacognition Assessment Scale-Abbreviated (MAS-A; 11) to compare the role of self-reflectivity in the above specified relation to the role played by the understanding of others’ minds. In addition, on the basis of the assumption that measures of schizotypy assess schizophrenia proneness (12), it tested whether the findings for symptoms of schizophrenia in a sample of persons with a diagnosis of schizophrenia would be found for schizotypic traits in a sample of research participants without a mental illness diagnosis. Theoretical and empirical research on the phenomena of metacognition have gone through an interesting process of change and development in which the term metacognition has come to encompass a variety of phenomena. At first the term referred to the ability of individuals to understand their own cognitive processes with regard to their desired actions, appraisals and self-awareness of these processes (13). Later, the concept of metacognition shifted from understanding oneself to include the understanding of the others’ minds (11). Metacognitive abilities enable persons to monitor and reflect upon their and other persons’ cognitive processes such as thoughts, desires, actions and anticipations regarding the world (14-17). These dual abilities of thinking about the self and thinking about others have been denoted by such terms as reflective awareness (18), metarepresentation (19, 20), mentalization (16), mind reading, and theory of mind (21). In this study we adopt the broad perspective on metacognition that are presented in this issue (22, 23). According to this perspective, metacognition can be defined as a spectrum of activities which includes discrete acts, such as noticing errors, memories or specific beliefs about other beliefs and more synthetic kinds of activities involved in integrating and bringing together any number of perceptions into complex ideas about the self and others as unique agents in the world. Semerari et al. (15) developed a narrative methodology, the MAS, to measure metacognition from an analysis of psychotherapy transcripts. Originally, this scale was divided into three sub functions: Understanding one’s own mind, the ability to comprehend one’s own mental states; Understanding others’ minds, the ability to comprehend other individuals’ mental sates; and Mastery, the strategies a subject uses to regulate his/her own mental states. This division of metacognitive abilities measured by the scale into sub functions, facilitated the determination of the exact abilities that might be deficient due to such conditions as personality disorders. Thus, the scale could be used to

assess whether psychotherapy had a positive influence on that particular metacognitive function of persons with a personality disorder (24, 25). In order to assess metacognitive abilities for a broad range of narratives, the MAS has been abbreviated and adapted for the study of IPII transcripts. This abbreviation and adaptation has been labeled the MAS-A (11). Studies of the metacognitive processes of persons with schizophrenia, using the combined IPII and MAS methodology and other measures of metacognition (8, 26), have shown that persons with schizophrenia exhibit impaired self-reflectivity and understanding of others’ minds. This impairment has been found to increase progressively as the symptoms of the illness of schizophrenia become more prominent (6, 11, 18, 27-31). The metacognitive difficulties of this population involve low self-awareness (32, 33), difficulties in detecting the mental states of others (27), low mentalization which reflects deficits in the ability to attribute mental states to others (28), and deficient social cognition that underlies those cognitive processes and capacities of persons to understand the behavior of others and to react adequately in social situations (34). Self-reflectivity and understanding of others’ minds are regarded as key elements required for the establishment and maintenance of social relationships (35). They allow a person to relate to the emotional states of her/himself and to the emotional states of the other and, thus, to predict future behavior and mental states (14, 36). These elements of metacognition also involve the detection and processing of social information (37) and, thereby, the knowledge of how to behave appropriately in a variety of social contexts. Thus, a deficit in metacognitive ability could play a major role in the difficulties in interpersonal relationships that affect the social quality of life (SQoL) of persons with schizophrenia (38-43). A range of metacognitive deficits are also exhibited by persons with high levels of schizotypic traits. Schizotypic traits include predispositions for unusual experiences, cognitive disorganization, impulsive nonconformity, and introvertive anhedonia (44). These traits are regarded as being similar to, but less extreme than, the classical negative and positive symptoms of schizophrenia (45, 46). As pointed out above (12), according to a number of theoreticians, in a variety of circumstances, these traits can be precursors of schizophrenia (46, 47). Research has shown that self-awareness is negatively correlated with the level of the schizotypic traits (48). In addition, persons who exhibit high levels of these traits have been 45

Metacognition in Schizophrenia and Schizotypy

found to have difficulty in engaging in “Theory of Mind” processes, which entail the understanding of others’ thoughts and emotions (48-50). However, while these studies showed a high level of schizotypic traits to be related to metacognitive deficits, they have not produced consistent support for the impact of metacognitive deficits on the SQoL of persons without mental illness. Some studies indicate that persons who exhibit high levels of schizotypic traits show the same level of social functioning as those who exhibit low levels of these traits (51-54) while other studies indicate that the persons who exhibit high levels of schizotypic traits show significant deficiencies in forming close relationship as well as low emotional intelligence, difficulties in interpersonal perception and low social skills (55-61). Theory and research have suggested that deficits in the metarepresentation of the self and the other may lead to and account for the development of schizotypic traits and the symptoms of schizophrenia (6, 62, 63). This assumption attributes both positive symptoms and negative symptoms, and the parallel traits of schizotypy to deficient metarepresentation. Thus, delusions of alien control or auditory hallucinations have been attributed to difficulties in metarepresenting and monitoring willed intentions (6). In addition, it has been suggested that impairment of the neurodevelopment of metacognitive abilities may contribute to the overall severity of schizophrenia, including such negative symptoms as deficient cognitive processing and social functioning (64). Clusters of both positive and negative symptoms may be the consequence of metacognitive deficits that could lead to lower SQoL. The current study tested the hypothesis that persons with a diagnosis of schizophrenia will report statistically significant lower levels of SQoL than persons without mental illness. In addition, to the best of our knowledge, this is the first study that uses the MAS-A with a nonclinical population. Therefore, it was expected, in keeping with studies that used other scales of metacognition, that persons with schizophrenia will show lower metacognitive abilities in comparison to persons without mental illness when their narratives are analyzed according to the MAS-A. In addition, this study examined the possibility that the positive and negative symptoms of schizophrenia and the two traits of schizotypy, exceptional experiences and introvertive anhedonia, that correspond respectively to the positive and negative symptoms of schizophrenia, will mediate the relation between metacognitive abilities 46

and SQoL in populations of persons with schizophrenia and persons without mental illness. Thus, we expected that the symptoms of schizophrenia will mediate the relation between metacognition and SQoL of persons with schizophrenia and that schizotypic traits will mediate the relation between metacognition and SQoL of persons without mental illness. This putative similar pattern in both populations of the relations between metacognition, symptoms of schizophrenia/schizotypy traits and SQoL is based on theoretical and empirical research. This research suggests that deficits in the metacognition of persons without mental illness that exhibit high levels of schizotypy is a mirror of the metacognitive deficits of persons with schizophrenia (65). According to Baron and Kenny’s (66) approach, in the present study, the following four findings are required to establish a mediation model: 1) symptoms of schizophrenia and traits of schizotypy will be found to be negatively correlated with SQoL; 2) these symptoms and traits will be negatively correlated with metacognition; 3) metacognition will be positively correlated with SQoL; 4) when the symptoms of schizophrenia and the schizotypic traits are controlled for, the relation between metacognition and SQoL will be significantly decreased. These hypotheses will be examined for this study’s sample of persons with a diagnosis of schizophrenia and for this study’s sample of persons without a psychiatric diagnosis. Method Participants

Two groups of research participants took part in the present study. One group included 39 persons with a diagnosis of schizophrenia. To be included in this group, a person had to meet the following criteria: 1) not to have another psychiatric diagnosis and 2) to have had the diagnosis of schizophrenia for at least one year. These participants were either being treated by the psychiatric unit of Soroka University Medical Center or the Beer Yaakov Mental Health Center. The second group consisted of 60 persons who meet the following criteria: 1) no psychiatric diagnosis and 2) no reported neurocognitive disorder. The latter criteria were assessed by self-report. Significant differences in age, gender and years of education were observed between the two groups. The mean and standard deviation for age of the persons with a diagnosis of schizophrenia was 39.36 and 12.62 years, respectively with a range from 19–67 years; 57.9 % of this group were male and 18.4 % had more than 12 years of education. The

Shai Joseph Rabin Et al.

mean and standard deviation for the age of the participants without a psychiatric diagnosis were 22.23 and 2.00 years, respectively with a range from 19–26 years; 76.3 % of this group were female. All of the participants in this group had more than 12 years of education. Procedure

The present study was conducted after the ethical committee of the two participating medical centers provided approval. Metacognition, symptoms and SQoL were assessed individually during one meeting that lasted about an hour. Each of two researchers conducted interviews with approximately half of the research participants. The self-report measures were filled out with the help of the researchers only when the participants had difficulties reading the questionnaire. Whereas the schizotypic traits and SQoL of the participants without a psychiatric diagnosis were assessed by self-report questionnaires, psychiatric symptoms and SQoL of participants with schizophrenia were assessed by the research participants’ psychiatrists. Instruments

The Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE, 44) was used to measure the schizotypic traits. This inventory is a self-report measure of schizotypy in healthy populations. It consists of 159 yes-no items that have been shown to generate four schizotypy factors, “unusual experiences’’ (UE); “cognitive disorganization’’ (CD); “impulsive nonconformity’’ (IN); and “introvertive anhedonia’’ (IA) (67). It also provides a global schizotypy score. For the purpose of the current study, the two subscales, “unusual experiences’’ and “introvertive anhedonia’’ that respectively parallel the positive and negative symptoms of schizophrenia were used. Cronbach’s alpha for the each of the scales for the current study was 0.90. The Positive and Negative Syndrome Scale (PANNS; 68). This 30 item questionnaire that was filled out by the treating psychiatrist was used to assess the symptoms of schizophrenia. For the purposes of the current study, only the Positive Symptoms and Negative Symptoms subscales were used. Persons are rated on these items on the basis of a structured clinical interview. For each item, a person can receive a rating of severity from between 1 to 7, when 1 represents the absence of the symptom and 7 represents an extremely severe symptom. Cronbach’s alpha for the total scale score, the positive symptoms and the negative symptoms were 0.81 and 0.92 respectively. The Indiana Psychiatric Illness Interview (IPII; 10) and the Metacognition Assessment Scale- Abbreviated (MAS-A;

11). The IPII and the MAS-A were used jointly to assess metacognition. The IPII is a semi-structured interview that was developed to assess mental illness narratives. The interview that typically lasts from between 30 to 60 minutes was conducted by a research assistant who recorded the interviewee’s responses. The interview is divided into the following five sections. First, rapport is established between the interviewer and the research participants and the participants are asked to tell the story of their lives in as much detail as possible. Second, the participants are asked if they think they have a mental illness and how they understand it. Third, their response to this question is followed by questions concerning what has affected their interpersonal and psychological life activities. Fourth, they are asked if their condition “controls” their life and if they “control” their illness. They are also asked whether their illness is affected by others and how much others have been affected by their illness. Fifth, they are asked what they expect to stay the same and what will be different in the future with regards to their interpersonal and psychological functioning. In order to use the IPII with the healthy participants, the IPII was modified so that instead of asking about a psychiatric illness, participants were asked about an important life challenge.In all other respects, the interview was the same. The responses to the IPII were analyzed according to the MAS-A. The MAS-A is a rating scale for assessing the metacognitive abilities manifested by an individual’s verbal responses. It was originally designed to assess changes in the ability of persons with severe personality disorders to think about their own thinking on the basis of psychotherapy transcripts. Later, the MAS was adapted for the assessment of IPII transcripts (11). The MAS-A conceptualizes metacognition as the set of abilities that allows persons to understand mental phenomena and to use that understanding to tackle tasks that are sources of distress. The MAS consists of four scales. The first, self-reflectivity, refers to the ability to form increasingly complex and integrated representations of one’s self. The second scale, understanding of others’ minds, refers to the ability to form increasingly complex and integrated representations of other persons. The third scale, decentration, refers to the ability to see the world in which one finds oneself as understandable from a number of different perspectives. The final scale, mastery, refers to the ability to use knowledge of mental states to solve psychological problems. In the current study, only the two subscales, self-reflectivity and understanding of others’ minds were 47

Metacognition in Schizophrenia and Schizotypy

used. Inter-rater reliabilities for the “Self-Reflectivity” subscale and for the “Understanding the Other’s Mind” subscale were found 0.87 and 0.95 respectively. Interviews were carried out by two interviewers. Prior to the present study, a pilot study was undertaken with 15 research participants from each group. Each of the interviewers rated the transcripts of the interviews independently. Inter-rater reliabilities were high, for self-reflectivity, 87 and for understanding of the other, .95. Therefore the task of rating the remaining transcripts was divided equally between the two interviewers. Social Quality of Life (SQoL) was assessed by the Social Quality of Life subscale of the Hebrew translation and adaptation (69) of the Wisconsin Quality of life Index for Mental Health (QLI-MH) that was developed by Becker et al. (70). This index provides measures of quality of life in seven life domains. In each of these life domains, this index provides questionnaires for measuring quality of life from the perspective of the person with the psychiatric diagnosis, from the perspective of that person’s clinician, and from the perspective of a relative of the person with the diagnosis. In this study, only the measure of quality of life in the social domain was used from the clinician’s perspective. For the sample without a psychiatric diagnosis, SQoL was assessed from the perspective of the participants. Cronbach’s alpha in the current study was 0.68 for both the patient version and the psychiatrist version.

Table 1. Mean, t-tests and standard deviation of SQoL and the metacognition for participants with schizophrenia and for persons without a psychiatric diagnosis (total N=99, N of schizophrenia sample= 39, N of non clinical sample=60).

M (sd) Self reflectivity

Participants with schizophrenia

4.73 (2.54)

Participants without a psychiatric diagnosis

6.76 (0.90)

Understanding of others’ minds

Participants with schizophrenia

3.43 (2.51)

Participants without a psychiatric diagnosis

5.72 (0.97)

SQoL

According to the psychiatrist’s report

4.38 (1.18)

Reported by participants without a psychiatric diagnosis

5.52 (0.69)

t-test (std. error difference)

5.68**

7.34**

6.03 (0.18)**

*p<0.05 **p<0.01 (1 tailed)

compared to persons without a diagnosis of schizophrenia. In addition, individuals without a diagnosis of schizophrenia reported higher SQoL than persons with a psychiatric diagnosis as reported by their psychiatrists. Thus, persons with schizophrenia showed lower metacognitive ability and lower SQoL in comparison with persons without a mental illness. Correlations between variables and regression analysis: Group of persons with schizophrenia

Results Descriptive statistics of the research variables and group comparisons

Descriptive statistics of the study variables for the two groups are presented in Table 1. As can be seen from this table, the ratings of the narratives of the participants with a diagnosis of schizophrenia indicated a lower capacity for metacognitive thinking on the MAS-A subscales of self-reflectivity and understanding others’ minds when

Table 2 presents the correlations between the study’s variables for the persons with schizophrenia. As is evident from Table 2, positive and negative subscales of the PANSS, were negatively correlated with the psychiatrists’ assessments of SQoL. However, the psychiatrist’s assessment of the SQoL of the participants with schizophrenia correlated positively and significantly only with these participants’ metacognitive ability to understand the

Table 2. Correlations between quality of life and the metacognition variables for persons with schizophrenia (N=39).

Positive Symptoms

Positive symptoms

Negative symptoms

Self reflectivity

Understanding of others’ minds

SQoL as assessed by the psychiatrist

0.70**

0.16-

0.52-**

0.59-**

Negative symptoms

0.27-*

0.50-*

0.63-**

Self reflectivity

0.60**

0.19

Understanding of others’ minds

0.48**

SQoL as assessed by the psychiatrist

*p<0.05 **p<0.01 (1 tailed)

Shai Joseph Rabin Et al.

Table 3. Hierarchical regression analysis for the research participants with schizophrenia (N=39). Beta

Std. Error

R2 Change

0.23

10.95**

0.46

0.23

9.81**

Step 1 Understanding of others’ minds

0.48**

0.07

0.23

Understanding of others’ minds

0.17

0.07

0.08

Level of negative symptoms

0.38-*

0.02

0.05-

Level of positive symptoms

0.23

0.03

0.04-

0.48

Step 2

0.62

*p<0.05 **p<0.01 (1 tailed)

others’ minds. Furthermore, only the negative symptoms were negatively correlated with both metacognitive abilities, while the positive symptoms were significantly correlated only with understanding of the others’ mind. Thus, symptoms of schizophrenia are significantly related to SQoL while only the negative symptoms were significantly related to both metacognitive abilities (understanding the self and understanding others’ minds), and positive symptoms were related only to understanding of others’ minds. Finally, SQoL was found only to be related to the understanding of others’ minds. Therefore, to test the hypothesized mediation model according to which psychiatric symptoms mediates the relation between metacognitive capacity as assessed by the MAS-A and SQoL, a hierarchical regression was carried out for the persons with a diagnosis of schizophrenia with SQoL as assessed by the psychiatrist as the dependent variable. On the basis of the results of the correlation analysis presented in Table 2, understanding of others’

minds, but not self-reflectivity, was entered into this analysis at step 1 and negative and positive symptoms, were entered into the analysis at step 2. Table 3 presents the results of this analysis. Df As can be seen from Table 3, at step 1, understanding of others’ minds was found to contribute signifi36 cantly and positively to the SQoL (β=0.48; P <. 01) of the participant with schizophrenia. However, when levels of negative and positive symptoms were added to the analysis, at step 2, this statistically significant contribution of understanding of others’ minds to SQoL was reduced and was no longer statistically significant (β=0.17; P>. 05). In addition, level of 34 negative symptoms of schizophrenia at step 2 was shown to contribute significantly and negatively to SQoL (β = -0.38; P <.05). These results are consistent with the hypothesis that the negative symptoms of schizophrenia mediate the positive relation between metacognition and SQoL for persons with schizophrenia. Accordingly, the capacity for understanding of others’ minds of the participants with schizophrenia appears to reduce the negative symptoms of schizophrenia and, thus, to increase these participants’ SQoL. Correlations between variables and regression analysis: Group of persons without mental illness

Correlations between study variables for the group without a mental illness diagnosis are presented in Table 4. As can be seen from this table, only introvertive anhedonia had a negative and significant correlations with SQoL. Unusual experiences did not show such correlation with SQoL. In addition, both metacognitive abilities were significant correlated with introvertive anhedonia and unusual experiences. Finally, of the metacognitive subscales, only understanding of others’ minds was correlated with SQoL. The above pattern of results suggests that introvertive anhedonia, but not unusual experiences, mediates the positive relation between understanding the others’ minds,

Table 4. Correlations between quality of life and the metacognition variables for persons without a psychiatric diagnosis (N=60). Unusual Experiences

Introvertive Anhedonia

Self reflectivity

Understanding of others’ minds

SQoL as reported by the participant

Unusual Experiences

0.17

0.22-*

0.33-**

0.16-

Introvertive Anhedonia

0.35-**

0.63-**

0.37-**

Self reflectivity

0.41**

0.17

Understanding of others’ minds

0.42**

SQoL as reported by the participant

*p<0.05 **p<0.01 (1 tailed)

Metacognition in Schizophrenia and Schizotypy

Table 5. Regression analysis for the participants without a psychiatric diagnosis (N=60). Beta

Std. Error

R2 Change

0.42

0.18

12.45**

0.20

0.02

6.99**

Step 1 Understanding of others’ minds

0.42**

0.08

0.30

Understanding of others’ minds

0.30

0.11

0.21

Introvertive Anhedonia

0.18-

0.03

0.04-

Step 2 0.44

*p<0.05 **p<0.01 (1 tailed)

and SQoL. To test this hypothesized mediation model, a hierarchical regression was carried out for persons without a diagnosis of schizophrenia. Table 5 present the results of this regression analysis. In keeping with the results of the correlation analysis, at step 1, understanding of others’ minds contributed significantly and positively to the self-reported SQoL (β=0.42; P <. 01) of persons without a psychiatric diagnosis. When level of introvertive anhedonia was added to the analysis at step 2, the statistically significant and positive contribution of understanding of others’ minds to SQoL was reduced and was no longer statistically significant (β=0.30; P>.05). However, the negative relation between introvertive anhedonia and SQoL was also not statistically significant (β = -0.18 P >. 05). Thus, the results of the hierarchical regression are not consistent with a full mediation model. Discussion This study replicated the general finding that persons with a diagnosis of schizophrenia express significantly less SQoL than persons without such a diagnosis (4). This is also the first study to use the MAS-A to assess metacognition in a non-clinical sample. This study’s findings are also consistent with previous literature that shows that the metacognitive abilities of persons with schizophrenia are lower than the metacognitive abilities of persons without schizophrenia (71, 72). Subsequently, this study examined the extent to which and the manner in which the positive and negative symptoms of schizophrenia and two parallel traits of schizotypy mediate the relation between metacognition and SQoL. With regard to both samples, correlations were found between self-reflectivity and the negative symptoms 50

of schizophrenia, and the parallel schizotypal trait of negative anhedonia. However, in both samples, self-reflectivity was not found to be related to SQoL. While testing the mediation model, with regard to Df the sample of persons with schizophrenia, negative symptoms of schizophrenia were found to mediate 58 the relationship between knowing the others’ minds and SQoL (assessed by the psychiatrist). Thus, knowing the others’ minds seems to increase the SQoL of persons with schizophrenia because it decreases 57 the negative symptoms of the illness. Similar results were uncovered with regard to schizotpy. Regression analysis showed that the metacognitive capacity of understanding others’ minds was negatively related to self-reported SQoL. However, in step 2, while entering the introvertive anhedonia to the regression, neither the schizotypal trait, nor knowing the others’ minds, predicted SQoL. This study’s finding that negative symptoms, but not positive symptoms, mediate the relation between metacognitive processes and SQoL is consistent with much research that shows negative symptoms to be a major source of the reduction in SQoL exhibited by persons with schizophrenia (73). It is also consistent with studies showing that negative symptoms of schizophrenia, and not positive symptoms are related to metacognition (74). In the current study, among participants with schizophrenia, the negative symptoms were found to be related to both metacognitive abilities, while the positive symptoms were found to be related only to knowing the others’ minds aspect of metacognition. In the nonclinical sample both schizotypal traits that are parallel to the negative and positive symptoms of schizophrenia, were found to be related to both metacognitive abilities. As mentioned above, regression analyses were conducted only for the metacognitive ability of knowing the others’ minds, since self-reflectivity was not found to be related to SQoL. These analyses supported a mediation model among participants with schizophrenia, and not among the non-clinical sample of the study. With regard to schizophrenia, an impaired metacognitive ability of understanding others’ minds, could lead to such negative symptoms of schizophrenia as extreme anhedonia, apathy, and flattening of affect. These symptoms may reduce SQoL directly by interfering with the interpersonal behavior that is required for the initiation and maintenance of gratifying social relationships or indirectly by interfering with understanding of the other, thereby, producing mutual misunderstanding and frustration.

Shai Joseph Rabin Et al.

As pointed out above, although similar patterns of correlations were found among both samples, full mediation of the schizotypal traits, between understanding of others’ minds and SQoL, was not found among the population without the illness. The results of the regression analyses among the non-clinical sample suggest that the schizotypal trait that parallel the negative symptoms of schizophrenia, and the metacognitive ability of understanding others’ minds, neutralize the effects of each other on SQoL. Although both correlate with SQoL, when entered together in the regression, none of them predicts SQoL. This multicollinearity suggests that both variables, the schizotypal trait of introvertive anhedonia and the metacognitive ability of understanding others’ minds, explain congruent parts of SQoL among person without mental illness. Therefore, the current study provides only partial support for the contention that schizotypy is a kind of a specific schizophrenia proneness or, in terms used by Rossi and Daneluzzo (75, p. 67), a “liability” to schizophrenia. Actually, Rossi and Daneluzzo (75), consider schizotypy to be a non-specific “psychosis-proneness” and claim that their comparison of schizotypal dimensions in persons without a psychiatric diagnosis to these dimensions in persons with either a diagnosis of schizophrenia, bipolar disorder, unipolar disorder, or an obsessive-compulsive disorder provides support for the above position. Findings of a study conducted by Stirling et al. (62) on the relations between hallucination proneness, schizotypy, and metacognition can be interpreted as support for conceptualizing schizotypy as a “normally distributed multidimensional trait which reflects the extent to which an individual may be considered psychosis-prone” and “as a necessary (but not sufficient) condition for the subsequent development of schizophrenia” (62, p. 1406). These findings also suggest one way in which the phenomenology of schizotypy may overlap the phenomenology of schizophrenia and, given the right circumstances, lead to the symptoms and impaired function that are often a consequence of schizophrenia. Furthermore, Stirling et al. (62) contend that schizotypy may give rise to “heightened sensitivity towards, greater awareness of, and increased concern about both positive and negative aspects of thinking” (p. 1407). Such sensitivity, awareness, and concern could have both beneficial and harmful consequences. This study’s results and the explanations that have been provided to explain them may have significant theoretical, empirical and clinical implications. First, this study has produced evidence of the relations between

negative symptoms and metacognition among persons with schizophrenia consistent with previous studies that used the MAS-A (11, 31). Therefore, it supports the major role that negative symptoms play in relation to both SQoL and metacognition. In addition, this study, to the best of our knowledge, is the first to use the MAS-A to show that these relations between metacognition and the schizotypy traits in normal populations parallel the relations between metacognition and negative symptoms of schizophrenia. Accordingly, this study’s results provide grounds for attributing an important role to self-reflectivity and understanding others’ minds for persons without mental illness, suggesting that this relation between metacognition and schizotypy traits may involve other processes than those that are involved in the manner in which the negative symptoms of schizophrenia mediate the relation between metacognition and SQoL. However, the significance of this study’s results is limited by the small size and homogeneity of its samples, as well as by its cross-sectional design. Therefore, studies with larger and more heterogeneous samples, using a longitudinal approach should be carried out in order to support and validate the current study’s results. References 1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: American Psychiatric Association, 1994. 2. World Health Organization. Classification of Mental and Behavioral Disorders: Diagnostic Criteria for Research, 10th revision. Geneva, Switzerland: World Health Organization, 1992. 3. Becker T, Leese M, Krumm S, Ruggeri M, Vázquez-Barquero JL. Needs and quality of life among patients with schizophrenia in five European centres: What is the impact of global functioning scores? Soc Psychiatry Psychiatr Epidemiol 2005;40:628-634. 4. Ulas H, Akdede BB, Ozbay D, Alptekin K. Effect of thought disorders on quality of life in patients with schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 2008;32:332-335. 5. Eack SM, Newhill CE. Psychiatric symptoms and quality of life in schizophrenia: A meta-analysis. Schizophr Bull 2007;33:1225–1237. 6. Frith CD. The cognitive neuropsychology of schizophrenia. Hove, U.K.: Lawrence Erlbaum, 1992. 7. Greig TC, Bryson GJ, Bell MD. Theory of mind performance in schizophrenia: Diagnostic, symptom, and neuropsychological correlates. J Nerv Ment Dis 2004;192:12-18. 8. Lysaker PH, Ringer JM, Buck KD, Grant M, Olesek K, Leudtke BL, Dimaggio G. Metacognitive and social cognition deficits in patients with significant psychiatric and medical adversity: A comparison between participants with schizophrenia and a sample of participants who are HIV-positive. J Nerv Ment Dis 2012;200:130-134. 9. Jalbrzikowsk M ,Carter C, Senturk D, Chow C, Hopkins JM, Green MF, Galván A, Cannon TD, Bearden CE. Social cognition in 22q11.2 microdeletion syndrome: Relevance to psychosis? Schizophr Res 2012;142:99-107. 10. Lysaker PH, Clements CA, Plascak-Hallberg CD, Knipscheer SJ, Wright

Metacognition in Schizophrenia and Schizotypy

DE. Insight and personal narratives of illness in schizophrenia. Psychiatry 2002;65:197-206. 11. Lysaker PH, Carcione A, Dimaggio G, Johannesen JK, Nicolo G, Procacci M, Semerari A. Metacognition amidst narratives of self and illness in schizophrenia: Associations with insight, neurocognition, symptom and function. Acta Psychiatr Scand 2005;112:64-71. 12. Kwapil TR, Barrantes-Vidal N, Silvia PJ. The dimensional structure of the Wisconsin schizotypy scales: Factor identification and construct validity. Schizophr Bull 2008;34:444-457. 13. Flavell JH. Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. Am Psychol 1979;34:906-1110. 14. Baron-Cohen S. Theory of mind and autism: A fifteen year review. In: Baron-Cohen S, Tager-Flusberg H, Cohen D, editors. Understanding other minds. Oxford: Oxford University, 2000: pp. 3-20. 15. Semerari A, Carcione A, Dimaggio G, Falcone M, Nicolo G, Procaci M, Alleva G. How to evaluate metacognition function in psychotherapy? The metacognition assessment scale - its applications. Clin Psychol Psychother 2003;10:238-261. 16. Lee KH. Empathy defective in schizophrenia. In: Farrow FD, Woodruff PWR, editors. Empathy in mental illness. Cambridge: Cambridge University, 2007: pp. 17-33. 17. Dunlosky J, Metcalfe J. Metacognition. Thousand Oaks, Cal.: Sage, 2009. 18. Langdon R, Michie PT, Ward PB, McConaghy N, Catts SV, Coltheart M. Defective self and/or other mentalising in schizophrenia: A cognitive neuropsychological approach. Cogn Neuropsychiatry 1997;2:167-193. 19. Leslie AM. Pretense and representation: The origins of “theory of mind.” Psychol Rev 1987;94:412-426. 20. Perner J. Understanding the representational mind. Cambridge, Mass.: MIT, 1991. 21. Premack D, Woodruff G. Does the chimpanzee have a theory of mind? J Behav Brain Sci 1978;1:515-526. 22. Lysaker PH, Hasson-Ohayon I. Metacognition in schizophrenia: Introduction to the special issue. Isr J Psychiatry Relat Sci 2014;51:4-7. 23. Lysaker PH, Leonhardt, BL, Pijnenborg M, van Donkersgoed R, de Jong S, Dimaggio G. Metacognition in schizophrenia spectrum disorders: Methods of assessment and associations with neurocognition, symptoms, cognitive style and function. Isr J Psychiatry Relat Sci 2014;51:54-62. 24. Semerari A, Carcione A, Dimaggio G, Nicolò G, Pedone R, Procacci M. Metarepresentative functions in borderline personality disorder. J Pers Disord 2005;19:690-710. 25. Dimaggio G, Carcione A, Nicolo G. Metacognition disorders: Research and therapeutic implications. Psychother Bull 2007;42:32-37. 26. Bell M, Bryson G, Lysaker P. Positive and negative affect recognition in schizophrenia: A comparison with substance abuse and normal control subjects. Psychiatry Res 1997;73:73-82. 27. Frith CD, Corcoran R. Exploring theory of mind in people with schizophrenia. Psychol Med 1996;26:521-530. 28. Corcoran R, Cahill C, Frith CD. The appreciation of visual jokes in people with schizophrenia: A study of “mentalizing” ability. Schizophr Res 1997;24:319-327. 29. Doody GA, Gotz M, Johnstone EC, Frith CD, Owens DG. Theory of mind and psychoses. Psychol Med 1998;28:397-405. 30. Garety PA, Freeman D. Cognitive approaches to delusions: A critical review of theories and evidence. Br J Clin Psychol 1999;38:47-62. 31. Nicolò G, Dimaggio G, Popolo R, Carcione A, Procacci M, Hamm J, Buck KD, Pompili E, Buccione I, Lagrotteria B, Lysaker PH. Associations of metacognition with symptoms, insight, and neurocognition in clinically stable outpatients with schizophrenia. J Nerv Ment Dis 2012;200:644-647. 32. Frith CD. When conscious recollection disrupts memory: Commentary. Arch Gen Psychiatry 1999;56:645-646. 33. Frith CD. Schizophrenia and theory of mind. Psychol Med 2004;34:385-389. 34. Pinkham AE, Penn DL, Perkins DO, Lieberman J. Implications for the

neural basis of social cognition for the study of schizophrenia. Am J Psychiatry 2003;165:815-824. 35. Flavell JH, Miller PH, Miller SA. Cognitive development (3rd ed.). Englewood Cliffs, N.J.: Prentice-Hall, 1993: p. 174. 36. Wellman HM, Lagattuta KH. Developing understanding of mind. In: Baron-Cohen S, Tager-Flusberg H, Cohen DJ, editors. Understanding other minds. Oxford: Oxford University, 2000: pp. 21- 49. 37. Corrigan PW, Penn DL. Introduction: Framing models of social cognition and schizophrenia. In: Corrigan PW, Penn DL, editors. Social cognition and schizophrenia. Washington, D.C.: American Psychological Association, 2001: pp. 3-39. 38. Roncone R, Falloon RH, Mazza M, DeRisio A, Pollice R, Neczione S, Morosini Pl, Casacchia M. Is theory of mind in schizophrenia more strongly associated with clinical and social functioning than with neurocognitive deficits? Psychopathology 2002;35:280-288. 39. Bora E, Eryavuz A, Kayahan B, Sungu G, Veznedaroglu B. Social functioning, theory of mind and neurocognition in outpatients with schizophrenia; mental state decoding may be a better predictor of social functioning than mental state reasoning. Psychiatry Res 2006;145:95-103. 40. Brüne M, Abdel-Hamid M, Lehmkamper C, Sonntag C. Mental state attribution, neurocognitive functioning, and psychopathology: what predicts poor social competence in schizophrenia best? Schizophr Res 2007;92:151-159. 41. Bell M, Tsang H, GreigTC, Bryson GJ. Neurocognition, social cognition, perceived social discomfort, and vocational outcomes in schizophrenia. Schizophr Bull 2009;35:738-747. 42. Lysaker PH, Buck KD, Salvatore G, Popolo R, Dimaggio G. Lack of awareness of illness in schizophrenia: Conceptualizations, correlates and treatment approaches. Expert Rev Neurother 2009;9:1035-1043. 43. Lysaker PH, Dimaggio G, Carcione A, Procacci M, Buck KD, Davis LW, Nicolò G. Metacognition and schizophrenia: The capacity for self-reflectivity as a predictor for prospective assessments of work performance over six months. Schizophr Res 2010;122:124-130. 44. Mason O, Claridge G. The Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE): Further description and extended norms. Schizophr Res 2006;82:203-211. 45. Baker CA, Morrison AP. Cognitive processes in auditory hallucinations: Attributional biases and meta-cognition. Psychol Med 1998;28:11991208. 46. Raine A, Lencz T, Mednick SA. Schizotypal personality. Cambridge: Cambridge University, 1995. 47. Meehl PE. Toward an integrated theory of schizotaxia, schizotypy, and schizophrenia. J Pers Disord 1990;4:1-99. 48. Dinn WM, Harris CL, Aycicegi A, Greene P, Andover MS. Positive and negative schizotypy in student sample: Neurocognitive and clinical correlates. Schizophr Res 2002;56:171-185. 49. Langdon R, Coltheard M. Mentalising, schizotypy and schizophrenia. Cognition 1999;71:43-71. 50. Pickup G. Theory of mind and its relation to schizotypy. Cogn Neuropsychiatry 2006;11:117-192. 51. Toomey RM, Schuldberg D. Recognition and judgment of facial stimuli in schizotypal subjects. J Commun Disord 1995;28:193-203. 52. Waldeck TL, Miller LS. Social skills deficits in schizotypal personality disorder. Psychiatry Res 2000;93:237-246. 53. Van’t Wout M, Aleman A, Kessels RP, Laroi F, Kahn RS. Emotional processing in a non-clinical psychosis-prone sample. Schizophr Res 2004;68:271-281. 54. Jashan CS, Sergi MJ. Theory of mind, neurocognition and functional status in schizotypy. Schizophr Res 2007;89:278-286. 55. Barkus EJ, Stirling J, Hopkins RS, Lewis S. The presence of neurological soft sign along the psychosis proneness continuum. Schizophr Bull 2006;32:573-577. 56. Raine A. Schizotypal personality: Neuro developmental and psychosocial

Shai Joseph Rabin Et al.

trajectories. Annu Rev Psychol 2006;2:291-326. 57. Aguirre F, Sergi MJ, Levy CA. Emotional intelligence and social functioning in persons with schizotypy. Schizophr Res 2008;104:255-264. 58. Platek SM, Fonteyn LCM, Myers TE, Izzetoglu M, Ayaz H, Li C, Chance B. Functional near infrared spectroscopy reveals differences in self-other processing as a function of schizotypal personality traits. Schizophr Res 2005;73:125-127. 59. Mikhailova ES, Vladimirova TV, Iznak AF, Tsusulkovskaya EJ, Sushko NV. Abnormal recognition of facial expressions of emotions in depressed patients with major depression disorder and schizotypal personality disorder. Biol Psychiatry 1996;40:697-705. 60. Malcolm S, Paul Keenan J. My right I: Deception detection and hemispheric differences in self-awareness. J Soc Behav Pers 2003;31:767-772. 61. Langdon R, Coltheart M. Recognition of metaphor and irony in young adults: The impact of schizotypal personality traits. Psychiatry Res 2004;125:9-20. 62. Stirling J, Barkus E, Lewis S. Hallucination proneness, schizotypy and meta-cognition. Behav Res Ther 2007;45:1401-1408. 63. Lysaker PH, Davis LW, Lightfoot J, Hunter N, Stasburger A. Association of neurocognition, anxiety, positive and negative symptoms with coping reference in schizophrenia spectrum disorders. Schizophr Res 2005;80:163-171. 64. Ozguven HD, Oner,O, Baskak B Oktem F, Olmez S, Munir K. Theory of mind in schizophrenia and Aspergerâ&#x20AC;&#x2122;s syndrome: Relationship with negative symptoms. Klin Psikofarmakoloji Bulteni B 2010;20:5-13 (Turkish). 65. Lenzenweger MF. Schizotypy and schizophrenia: The view from experimental psychopathology. New York, N.Y.: Guilford, 2010. 66. Baron RM, Kenny DA. The moderator-mediator variable distinction

in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51:1173-1182. 67. Claridge G.(Ed.) Schizotypy: Implications for illness and health. New York, N.Y.: Oxford University, 1997. 68. Bell MD, Lysaker PH, Goulet JG, Milstein RM, Lindenmayer JP. Five factor model of schizophrenia. Psychiatry Res 1994, 52:295-303. 69. Dasberg I. Quality of life among people with mental illness: Comparison between therapist and patient perceptions and perceived control over the illness. MA thesis, Bar-Ilan University: Ramat Gan, 1998 (Hebrew). 70. Becker M, Diamond R, Sainfort F. A new patient focused index for measuring qulity of life in persons with severe and persistent mental illness. Qual Life Res 1993;2:239-251. 71. Lysaker JT, Lysaker PH. Being interrupted: The self and schizophrenia. J Speculative Philosophy 2005;19:1-22. 72. Lysaker PH, Lysaker JT. Psychotherapy and schizophrenia: An analysis of requirements of an individual psychotherapy for persons with profoundly disorganized selves. J Constr Psychol 2006;19:171-189. 73. Rabinowitz J, Levine SZ, Garibaldi G, Bugarski-Kirola D, Berardo CG, Kapur S. Negative symptoms have greater impact on functioning than positive symptoms in schizophrenia: Analysis of CATIE data. Schizophr Res 2012;137:147-150. 74. NicolĂ˛ G, Dimaggio G, Popolo R, Carcione A, Procacci M, Hamm J, Buck KD, Pompili E, Buccione I, Lagrotteria B, Lysaker PH. Association of metacognition with symptoms, insight and neurocognition in clinically stable outpatients with schizophrenia. http://www.ncbi.nlm.nih.gov/ pubmed/22759945 J Nerv Ment Dis 2012;200:7:644-647. 75. Rossi A, Daneluzzo E. Schizotypal dimensions in normals and schizophrenic patients: A comparison with other clinical samples. Schizophr Res 2002;54:67-75.

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Metacognition in Schizophrenia Spectrum Disorders: Methods of Assessment and Associations with Neurocognition, Symptoms, Cognitive Style and Function Paul H. Lysaker, PhD,1 Bethany L. Leonhardt, MA,2 Marieke Pijnenborg, PhD,3 Rozanne van Donkersgoed, MA,3 Steven de Jong, MA,3 and Giancarlo Dimaggio, MD4 1

Roudebush VA Medical Center and Department of Psychiatry, Indiana University School of Medicine, Indianapolis, Indiana, U.S.A. School of Psychological Science University of Indianapolis, Indianapolis, Indiana, U.S.A. 3 University of Groningen, Groningen, The Netherlands 4 Centro di Terapia Metacognitiva Interpersonale and Clinical Psychology Specialization Program, La Sapienza University, Rome, Italy 2

Abstract Deficits in metacognitive capacity in schizophrenia can be conceptualized as existing along a spectrum from more discrete to more synthetic activities. While each represents an equally important focus of study, synthetic metacognitive activities may be more difficult to measure given they are more a matter of assessing complexity of thought rather than concrete accuracy; and therefore have received less attention. This review summarizes research on synthetic metacognition using a paradigm in which metacognitive capacity is rated within personal narratives. Results across the work reviewed here provides evidence that these deficits are detectable in patients with schizophrenia and that deficits are related to, but not reducible to, symptom severity and poorer neurocognitive function. Independent of symptoms and neurocognition, deficits in synthetic metacognition are related to a range of mental activities including reasoning style, learning potential and insight. These deficits may also play a role in long term outcome via their impact on the ability to function in work settings and to think about and sustain social connections.

Schizophrenia is among the most complex and devastating forms of mental illness. Most models have stressed

that the dysfunction found in this condition results from an array of symptoms, neurocognitive deficits, and associated social factors such as trauma, poverty and stigma (e.g., 1-5). More recent work, however, has stressed that deficits in core psychological processes also play a role in outcome. In particular, one set of processes which may mediate or moderate the impact of social and biological factors upon daily life in schizophrenia is metacognition. Though originally used within the educational literature to refer to the ability to think about one’s own thinking when learning (6), metacognition has come to take on a range of meanings including broader processes in which information is integrated in more or less effective ways into complex and evolving representations of the self and others (e.g., 7). As noted earlier in this issue (8), metacognition may reflect a spectrum of activities which includes discrete acts, such as noticing errors, memories or specific beliefs about other beliefs and more synthetic kinds of activities involved in integrating and bringing together any number of perceptions into complex ideas about the self and others as unique agents in the world. These discrete and synthetic acts are thought to bi-directionally inform one another as persons evolve ideas of themselves and others in the flow of daily life (9, 10). For example, noticing that one is having a specific emotion or making an error in a specific situation may affect one’s larger sense of oneself just as one’s larger sense of oneself may affect how one notices errors and emotions in that moment.

Address for Correspondence: Paul Lysaker, PhD, Roudebush VA Med Center (116H), 1481 West 10th St., Roudebush VA Medical Center, Indianapolis, Indiana 46202, U.S.A. plysaker@iupui.edu

Paul H. Lysaker et al.

In this paper we will focus on research concerned with the more synthetic elements of metacognition, namely the complexities of self representations within the personal narrative of schizophrenia. This form of metacognition may have unique links with function as it is part of the basis for an evolving, nuanced personal understanding of immediate and long standing challenges of schizophrenia and trials of regular life outside of psychosis. Our goals are first to discuss one set of methods for quantitatively assessing synthetic aspects of metacognition within personal narratives in schizophrenia. We will then summarize research using this method to study the links between metacognition in schizophrenia with (i) deficits in neurocognitive ability and symptoms, (ii) cognitive styles and related elements and iii) functional outcomes such as vocational and social proficiency. A nuanced understanding of these issues could have important implications not merely for models of function, but also for addressing underlying barriers to recovery. A summary of the results of all studies presented along with participant characteristics is presented in Table 1. Assessing Synthetic Aspects of Metacognition in Schizophrenia

Many methods have been developed to assess more discrete metacognitive and/or social cognitive abilities. For instance, in a laboratory task participants might be asked to detect whether they have made an error, determine whether movements they were watching on a screen were mimicking their own movements, or recognize whether they know something or not (11-15). These type of assessments, however, are not necessarily useful for assessing synthetic abilities. While they offer a means of observing the accuracy of perceptions, at issue for synthetic functions are the complexity and adaptiveness with which material is integrated into a large representation of oneself and others. For instance, it may be more important that one’s idea of oneself is flexible rather than whether it is ultimately correct. Additionally, these tasks are often cued in affectively neutral contexts, whereas synthetic metacognitive acts often occur in emotion-laden contexts in which there may be no explicit cues to make a certain judgement. To address these limitations we have proposed a method to rate synthetic metacognitive abilities from a spontaneously generated speech sample in which persons discuss their lives and personal understanding of the trials they have faced. That speech sample is obtained through a semi-structured interview called the Indiana Psychiatry Illness Interview (IPII), which elicits a narrative of self and illness and typically lasts between 30 and 60 minutes

(16). The IPII differs from other psychiatric interviews in that the interviewer is instructed not to introduce content. For instance, no questions are asked about the presence of symptoms. The interviewer may ask for clarification when confused and query non-directively with the intent of understanding how persons make sense of their lives but not to collect specific historical facts. Thus, there are a number of opportunities in their stories where participants may demonstrate ability to think about their own mental states, the mental states of others, or how certain challenges are best faced. To quantify metacognitive capacity within IPII narratives, we have used a modified version of the Metacognition Assessment Scale – Abbreviated (MAS-A; 17). The MAS-A contains four scales: “Self-reflectivity,” or the comprehension of one’s own mental states, “Understanding of others’ minds,” or the comprehension of other individuals’ mental states, “Decentration,” which is the ability to see the world as existing with others having independent motives, and “Mastery,” which is the ability to use one’s mental states to respond to social and psychological dilemmas. It is assumed that the metacognitive capacities assessed by each scale are semi-independent and can vary along a continuum. Higher scores on each of the four MAS-A subscales reflect abilities to perform increasingly complex acts within the domain captured by that scale. For instance, higher scores on Self-reflectivity would suggest a capacity to form more complex representations of oneself while higher scores on Mastery would suggest the capacity to use more complex forms of metacognitive knowledge about oneself and others to respond to psychological and social challenges. Acceptable levels of interrater reliability and internal consistency have been reported along with evidence of stability of MAS-A assessments across a six month interval (17-19). Evidence that these procedures capture difficulties specific to psychosis includes findings that participants with schizophrenia have lower scores on all of the MAS-A subscales compared to others who also have significant medical and social adversity but not psychosis (20). Concerning its validity, MAS-A scores have been linked with independent assessments of awareness of illness and cognitive insight (17, 21). Associations of Metacognition with Symptoms and Neurocognition

In order to understand the links between metacognition and outcome, one set of studies has examined whether having more severe symptoms and neurocognitive deficits is linked with poorer synthetic metacognitive 55

Metacognition in Schizophrenia Spectrum Disorders: Methods of Assessment

Table 1. Summary of studies of the association performance on the Metacognition Assessment Scale Source

Other Instruments

Location

Demographics

Key Findings

Buck KD, Warman DM, Huddy V, Lysaker PH. The relationship of metacognition with jumping to conclusions among persons with schizophrenia spectrum disorders. Psychopathology 2012; 45:271-275

Beads Task, HVLT, WCST

USA

Veteran Sample 37 Men 3 Women 24 Schizophrenia 16 Schizoaffective Post-acute phase of illness

Lower level of mastery correlated with greater propensity to jump to conclusions

Davis LW, Eicher AC, Lysaker PH. Metacognition as a predictor of therapeutic alliance over 26 weeks of psychotherapy in schizophrenia. Schizophr Res 2011; 129: 85-90

HVLT, PANSS, WAIS, WAIS-C

USA

Veteran/Community Sample 53 Men 10 Women 41 Schizophrenia 22 Schizoaffective Post-acute phase of illness

Higher levels of mastery had better scores of therapeutic alliance

Fridberg DJ, Brenner A, Lysaker PH. Intrusion errors in schizophrenia: Association with self-monitoring, symptoms and executive function. Psychiatry Res 2010; 179, 6-11

CPT-II, HVLT, PANSS, WAIS, WCST

USA

Veteran/Community Sample 68 Men 11 Women 51 Schizophrenia 28 Schizoaffective Post-acute phase of illness

More intrusion errors on a verbal memory test was correlated with lower self-reflectivity, disorganized symptoms, and poorer executive function

Hamm JA, Renard SB, Fogley RL, Leonhardt BL, Dimaggio G, Buck KD, Lysaker PH. Metacognition and Social Cognition in Schizophrenia: Stability and Relationship to Concurrent and Prospective Symptom Assessments. J Clin Psychol 2012; 68:1303-1312

BLERT, PANSS, WCST

USA

Veteran Sample 44 Men 5 Women Schizophrenia Spectrum Post-acute phase of illness

MAS-A total scores are stable over time and correlated with current and prospective positive, negative, and affective symptoms. MAS-A were related to prospective negative symptom scores after controlling for initial levels of negative symptoms.

Luedtke BL, Kukla M, Renard S, Dimaggio G, Buck KD, Lysaker PH. Metacognitive functioning and social cognition as predictors of accuracy of self-appraisals of vocational function in schizophrenia. Schizophr Res 2012;137:260-261

BLERT, PANSS, WBI

USA

Veteran/Community Sample 41 Adults Schizophrenia Spectrum Post-acute phase of illness

Low self-reflectivity correlated with less accurate appraisals of one’s own work performance

Lysaker PH, Carcione A, Dimaggio G, Johannesen JK, Nicolò G, Procacci M, Semerari A. Metacognition amidst narratives of self and illness in schizophrenia: Associations with insight, neurocognition, symptom and function. Acta Psychiatr Scand 2005; 112: 64-71

DSS, HVLT, PANSS, QOL, SUMD, VRT, VS, WCST

USA

Veteran Sample 61 Men 40 Schizophrenia 21 Schizoaffective Post-acute phase of illness

Higher scores on domains of MAS-A related to: increased memory, processing speed, premorbid intelligence; lower scores on MAS-A linked with greater emotional withdrawal

Lysaker PH, Dimaggio G, Buck KD, Callaway SS, Salvatore G, Carcione A, Nicolò G, Stanghellini G. Poor insight in schizophrenia: Links between different forms of metacognition with awareness of symptoms, treatment need and consequences of illness. Comprehensive Psychiatry 2011;52:253-260

CPT-II, Hinting Task, HVLT, MAS, SUMD, WAIS, WCST

USA

Veteran/Community Sample 56 Men 9 Women 41 Schizophrenia 24 Schizoaffective Post-acute phase of illness

Self-reflectivity related to awareness of symptoms of psychosis; Awareness of the mind of the other was related awareness of treatment need and performance on the hinting task and mastery were related to awareness of consequence of illness

Lysaker PH, Dimaggio G, Buck KD, Carcione A, Nicolò G. Metacognition within narratives of schizophrenia: Associations with multiple domains of neurocognition. Schizophr Res 2007; 93:278-287

BLERT, PANSS, WAIS-III, WCST, WMS

USA

Veteran/Community Sample 61 Men 8 Women 43 Schizophrenia 26 Schizoaffective Post-acute phase of illness

Higher self-reflectivity was related to better performance on neurocognitive assessments and fewer negative and disorganized symptoms; higher Self-reflectivity coupled with higher levels of Decentration was related to better visual memory

Lysaker PH, Dimaggio G, Carcione A, Procacci M, Buck KD, Davis LW, Nicolo G. Metacognition and Schizophrenia: The capacity for self- reflectivity as a predictor for prospective assessments of work performance over six months. Schizophr Res 2010;122:124-130

WBI, WCST

USA

Veteran/Community Sample 47 Men 9 Women All Schizophrenia or Schizoaffective Post-acute phase of illness

High self-reflectivity correlated with better work performance over six months in vocational rehabilitation

Paul H. Lysaker et al.

Source

Other Instruments

Location

Demographics

Key Findings

Lysaker PH, Dimaggio G, Daroyanni P, Buck KD, LaRocco VA, Carcione A, Nicolò G. Assessing metacognition in schizophrenia with the Metacognition Assessment Scale: Associations with the Social Cognition and Object Relations Scale. Psychology and Psychotherapy 2010; 83: 303-315

HVLT, PANSS, SCORS, TAT, WCST

USA

Veteran/Community Sample 35 Men 2 Women 21 Schizophrenia 16 Schizoaffective Post-acute phase of illness

higher levels of mastery linked to a greater understanding of complex psychological forces that affect relationships

Lysaker PH, Erickson MA, Buck B, Buck KD, Olesek K, Grant MLA, Salvatore G, Popolo R, Dimaggio G. Metacognition and social function in schizophrenia: Associations over a period of five months. Cognitive Neuropsychiatry 2011; 16: 241-255

HVLT, MAQ, MSEI, SUMD, WAIS, WCST, WCQ

USA

Veteran/Community Sample 83 Men 15 Women 65 Schizophrenia 33 Schizoaffective Post-acute phase of illness

Mastery predicted concurrent social function; mastery at baseline affected mastery five months later as well as social function five months later.

Lysaker PH, Erickson MA, Ringer J, Buck KD, Semerari A, Carcione A, Dimaggio G. Metacognition in schizophrenia: The relationship of mastery to coping, insight, self-esteem, social anxiety and various facets of neurocognition. Br J Clin Psychology 2011; 50:412-424

HVLT, MAQ, MSEI, SUMD, WAIS, WCST, WCQ, WMS

USA

Veteran/Community Sample 83 Men 15 Women 65 Schizophrenia 33 Schizoaffective Post-acute phase of illness

Higher mastery correlated with coping through thinking and talking about events; intermediate mastery related to higher levels of resignation when facing stressors and more social phobia

Lysaker PH, McCormick BP, Snethen G, Buck KD, Hamm PANSS, JA, Grant MLA, Nicolò G, Dimaggio G. Metacognition UPSA, and social function in schizophrenia: Associations of WCST mastery with functional skills competence. Schizophr Res 2011; 131:214-218

USA

Veteran/Community Sample 36 Men 4 Women 19 Schizophrenia 21 Schizoaffective Post-acute phase of illness

Mastery was related to poorer performance on the comprehension/ planning subscale of UPSA independent of neurocognition

Lysaker PH, Ringer JM, Buck KD, Grant MLA, Olesek K, Luedtke B, Dimaggio D. Metacognitive and social cognition deficits in patients with significant psychiatric and medical adversity: A comparison between participants with schizophrenia and a sample of participants who are HIV-positive. J Nerv Ment Dis 2012; 200:130-134

BLERT, Hinting Task, HVLT

USA

Veteran/Community Sample 65 Men 23 Schizophrenia 17 Schizoaffective disorder 25 HIV and no psychosis

Participants with schizophrenia had lower levels of Selfreflectivity, awareness of the Other, Decentration, Mastery and performance on the Hinting test than a comparison group diagnosed with HIV but not psychosis.

Lysaker PH, Shea AM, Buck KD, Dimaggio, G, Nicolò G, Procacci M, Salvatore G, Rand KL. Metacognition as a mediator of the effects of impairments in neurocognition on social function in schizophrenia spectrum disorders. Acta Psychiatr Scand 2010; 122: 405-413

HVLT, PANSS, QOL, WAIS, WCST

USA

Veteran/Community Sample 87 Men 15 Women 68 Schizophrenia 34 Schizoaffective Post-acute phase of illness

Mastery mediates the impact of neurocognition upon the quality and quantity of social relationships after controlling for symptoms

Lysaker PH, Warman DM, Dimaggio G, Procacci M, LaRocco V, Clark LK, Dike C, Nicolò G . Metacognition in prolonged schizophrenia: Associations with multiple assessments of executive function. J Nerv Ment Dis 2008; 196: 384-389

BCIS, DKEFS, PANSS

USA

Veteran/Community Sample 49 Men 29 Schizophrenia 20 Schizoaffective Post-acute phase of illness

Self-reflectivity was correlated with mental flexibility; decentration, awareness of other, and mastery were correlated with response inhibition

Nicolò G, Dimaggio G, Popolo R, Procacci M, Hamm J, Buck KD, Pompili E, Buccione I, Lagrotteria B, Lysaker PH. Associations of metacognition with symptoms, insight and neurocognition in schizophrenia in an Italian replication study. J Nerv Ment Dis 2012; 200: 644-647

PANSS, Rey’s 15 Word Test, SUMD, WAIS, WCST, WMS

Italy

Community 27 Men 18 Women 32 Schizophrenia 13 Schizoaffective Post-acute phase of illness

Higher scores on domains of MAS-A related to: increased memory, processing speed, premorbid intelligence; lower scores on MAS-A linked with greater emotional withdrawal

Tas C, Brown EC, Esen-Danaci A, Lysaker PH, Brüne, M. Intrinsic motivation and metacognition as predictors of learning potential in patients with remitted schizophrenia. J Psychiatric Res 2012; 46: 1086-1092

IMI, PANSS, WCST, WMS

Turkey

Hospital Sample 16 Men 14 Women 30 Schizophrenia In symptom remission

Greater metacognitive capacity related to higher levels of intrinsic motivation; mastery found to be the best independent predictor of learning potential

function. For instance, symptom ratings and performance on neurocognitive testing was correlated with MAS-A scores among 61 men with schizophrenia in a non-acute phase of illness enrolled in rehabilitation (17). The results revealed that participants with greater capacity for Self-reflectivity had better verbal and visual memory,

processing speed and premorbid intelligence. Greater capacities for Knowing the other’s mind and Mastery were also related to better verbal memory. Concerning symptoms, greater levels of emotional withdrawal were linked with greater deficits in Self-reflectivity, Awareness of the other and Mastery. This pattern of results was later 57

Metacognition in Schizophrenia Spectrum Disorders: Methods of Assessment

replicated in an Italian sample also in a non-acute phase of illness not enrolled in rehabilitation (22). A second study compared assessments of neurocognition and symptoms among a new sample of 68 adults in a non-acute phase of schizophrenia with three different metacognitive profiles: (i) those who had achieved basic Self-reflectivity and Decentration (n=11), (ii) those with basic Self-reflectivity but without Decentration (n = 25), and (iii) those without both basic Self-reflectivity and Decentration (n = 25; 18). Analyses of variance revealed no between-group differences in demographic information. Basic Self-reflectivity was significantly related to better performance on a number of neurocognitive assessments and fewer negative and disorganized symptoms. Achieving Decentration was related to better visual memory. To explore the links between synthetic aspects of metacognition and executive function, MAS-A scores were correlated with selected subtests of the Delis Kaplan Executive Function System (DKEFS, 23) including tests of inhibition and set shifting and mental flexibility (21). The sample consisted of 49 participants drawn from the two studies described above if they had completed the DKEFS as part of another study. Results revealed that Self-reflectivity was more closely linked to mental flexibility than the other domains MAS-A subscales. The ability to inhibit a response was more closely linked to Decentration, Awareness of the otherâ&#x20AC;&#x2122;s mind and Mastery. This was interpreted as consistent with the possibility that those who are less able to define complex matter in multiple ways, such as thoughts and feelings, may have difficulties in sustaining awareness of internal complexity. Similarly, without an ability to inhibit thoughts about events in the world, some may find it difficult to call to mind the perspectives of others and to detect a range of possible reactions others are having in rapidly evolving situations. Finally, exploring the links between metacognition and symptoms over time in a group of 49 adults with schizophrenia in a stable phase of illness (19), it was found that the total score of the MAS-A was correlated with concurrent and prospective assessments of positive, negative and disorganization symptoms on the Positive and Negative Symptom Scale (24). In a multiple regression analysis, the MAS-A total score was found to predict prospective ratings of negative symptoms even after covarying for baseline negative symptoms scores. Results were interpreted as suggesting that metacognition may be related to negative symptoms in the moment and a risk factor for their emergence in the future. 58

Associations of Metacognition with Reasoning, Awareness and Learning

Exploring the relationship of metacognition to outcome, other studies have examined whether synthetic aspects of metacognition are related to other cognitive phenomenon including reasoning style, learning potential and the general ability to make sense of psychiatric challenges again regardless of clinical status. One of these studies (25) compared assessments of Mastery using the MAS-A with assessments of symptoms, neurocognition, and performance in a probabilistic reasoning task among 40 adults with a schizophrenia spectrum disorder. Partial correlations controlling for memory, executive function and symptoms revealed that lower levels of Mastery were associated with a greater propensity to jump to conclusions or to require less information before rendering a judgment requiring an appraisal of probability. Results were interpreted as suggesting that not being able to use psychological knowledge to solve problems may incline some to give up in the face of uncertainty and accept initial impressions rather than reason more deeply about the issue. Following these ideas, the relationship of Mastery to coping preference or the tendency to engage in problem solving versus giving up or ignoring the problem was examined (26). Participants were 98 adults with a schizophrenia spectrum disorder in a non-acute phase of illness classified into three groups on the basis of ratings of their MAS-A Mastery score: low Mastery (unable to plausibly represent psychological challenges; n = 33), intermediate Mastery (able to plausibly represent psychological problems but cope primarily through avoidance; n = 52), and high Mastery (able to cope with plausible problems through cognitive means; n = 13). Participants completed assessments of coping preference, anxiety, positive and negative symptoms and neurocognition. Analyses of variance revealed that the high Mastery group had a significantly greater preference for coping with stressors by thinking and talking about them than the intermediate or low Mastery group. The intermediate Mastery group reported higher levels of resignation when facing stressors and more social phobia than the other two groups. Group differences in a coping preference persisted when neurocognition and symptoms were controlled for statistically. Exploring the relationship of synthetic with more discrete forms of metacognition, another study investigated whether Self-reflectivity was associated with intrusion errors in a verbal memory task (27). Intrusion errors

Paul H. Lysaker et al.

refer to the free recall of incorrect material during tests of memory and as such represent a false memory that is not detected and corrected. Participants were 68 adults with a schizophrenia spectrum disorder administered the IPII and assessments of symptoms, executive function and verbal memory which recorded the presence of intrusion errors. After controlling for overall verbal memory performance, participants who made intrusion errors were found to have significantly poorer Self-reflectivity, more disorganized symptoms and poorer executive function. The relationship of Self-reflectivity with the presence of intrusion errors persisted after controlling for symptoms and executive functioning. Results were interpreted as suggesting difficulties in self reflection may lead to difficulties distinguishing actual from spurious memories, independent of deficits in executive functioning. Self-reflectivity has also been linked to the ability to accurately appraise the quality of oneâ&#x20AC;&#x2122;s own behavior in a study of 41 adults with a schizophrenia spectrum disorder enrolled in a cognitive therapy and vocational rehabilitation program (28). To estimate accuracy of self-appraisal of work quality, the mean absolute difference of the biweekly supervisor and participantâ&#x20AC;&#x2122;s self-assessment of work quality was calculated in this study for each week a participant worked over a period of six months. Participants were then divided into low, intermediate and high Self-reflectivity based on their MAS-A score. After controlling for symptoms, participants with low Self-reflectivity were found to have less accurate self appraisals of their work performance. Results were interpreted as suggesting that with lower levels of Self-reflectivity persons are less able to perform more discrete metacognitive tasks such as error detection in a work setting. To investigate the relationship of metacognition with awareness of the symptoms of schizophrenia, treatment need and consequences of illness, the IPII and assessments of social cognition, insight and neurocognition were gathered among 65 adults with a schizophrenia spectrum disorder in a non-acute phase of illness (29). After controlling for neurocognition, regressions revealed that Self-reflectivity was most closely linked to awareness of symptoms of psychosis while Mastery was most closely linked to awareness of treatment and consequence of illness. Interestingly, Mastery and social cognition were found to independently contribute to the prediction of awareness of consequences of illness. Results were interpreted as suggesting that with poorer levels of metacognition, many with schizophrenia may struggle

to construct complex, flexible and adaptive accounts of their psychiatric challenges and its implications. Finally, another study has examined whether assessments of metacognition with the MAS-A was related to learning potential in cognitive remediation experimental training (30). Participants were 32 adults with schizophrenia in a state of symptom remission. Intrinsic motivation and metacognition were assessed prior to the training. Results revealed that greater metacognitive capacity was significantly related to higher levels of intrinsic motivation. Patients with higher intrinsic motivation and preserved metacognition also improved more in the learning paradigm. Of the MAS-A scales, Mastery was found to be the best independent predictor of learning potential. Results are interpreted as suggesting that metacognitive ability may also increase the potential for learning. It is suggested that individuals with lower levels of metacognition may tend to rely on extrinsic motivation and hence struggle in paradigms which rely on intrinsic forms of motivation. Metacognition and Functional Outcomes

Turning to the issue of functional outcome, several studies have also suggested that metacognition may play a role in the ability to successfully function in both social and work settings. Concerning work function, one study has examined whether self-reflectivity predicted work performance measured every other week for six months (31). Participants were 56 adults with schizophrenia enrolled in a vocational rehabilitation program and were divided into three groups on the basis of their Self-reflectivity score on the MAS-A obtained prior to going to work: high (n=13), intermediate (n=21), and low Self-reflectivity (n= 22). A repeated measures analysis of variance comparing assessments of work performance revealed that the high Self-reflectivity group had significantly better work performance overall the entire six months than either of the other two groups. That difference persisted after controlling for executive function as assessed prior to starting work. To explore how metacognition might be linked to social function independent of neurocognition, assessments of neurocognition, quality and quantity of social relationships, neurocognition and symptoms were administered to 102 adults with a schizophrenia spectrum disorder in a post acute phase of illness (32). A Principal Components Analysis was used to reduce the assessments of neurocognition into a single index and then structural equation modeling techniques were used to test the model that 59

Metacognition in Schizophrenia Spectrum Disorders: Methods of Assessment

the capacity for metacognitive mastery mediates the impact of neurocognition upon the quality and quantity of social relationships after controlling for symptoms. Results revealed that an acceptable level of goodness of fit was observed between the model and data. As a followup to this study, a second path analysis was conducted to determine whether the cross sectional relationships observed above persisted over time (33). Specifically, two assessments of quality and quantity of social function and metacognition conducted five months apart for 72 of the original 102 participants were examined. In this study a model was specified in which: i) Mastery predicted concurrent social function, and ii) Mastery at baseline affected Mastery five months later; which similarly affected social function five months later. As in the first path analysis, acceptable levels of fit were found for the proposed model. Results of both studies were interpreted as suggesting metacognitive capacity may partially determine the extent to which neurocognitive deficits complicate efforts to relate to others. Regarding functional competence in terms of community function, another study (34) examined the relationship between metacognition and performance on an assessment of functional skills, the UCSD PerformanceBased Skills Assessment Battery (UPSA) (35), a battery that assesses competence with everyday living skills, such as paying oneâ&#x20AC;&#x2122;s bills, communication, effectively using transportation, etc. Participants were 45 adults in a post acute phase of schizophrenia who completed the IPII, assessments of neurocognition, symptoms and the UPSA. Correlational analyses revealed that scores on Mastery were correlated with the comprehension/ planning subscale of the UPSA after controlling for symptoms and executive function. Results were interpreted as suggesting that decrements in Mastery may make some withdraw from more complex daily tasks such as organizing complex plans for activity, resulting in the atrophy of those functional abilities. Concerning the relationship of metacognition with schema used to make sense of social interactions, another study compared MAS-A scores with assessments of social schema, symptoms, neurocognition were administered to a sample of 37 adults with schizophrenia in a nonacute phase of illness (36). Social schema was measured using the Social Cognition Object Relations Scale (37) which assesses awareness of interpersonal relationships as a result of complex psychological forces, as well as the recognition that people in relationships have independent needs. Correlational analyses controlling for symptoms 60

and neurocognition revealed that higher levels of Mastery were linked to a greater understanding of the complex psychological forces that affect relationships and the existence of independent needs of individuals in relationships. These findings were interpreted as suggesting that metacognitive abilities may provide the basis for the development of schemata which allow the comprehension of social relationships. Finally, to examine the link between metacognition and a specific relationship linked to outcome, namely therapeutic alliance with mental health providers, a study (38) has examined the link between Mastery and scores on the Working Alliance Inventory - Short Form (WAI-S, 39). Participants were 63 adults in a non-acute phase of schizophrenia enrolled in cognitive behavior therapy. Participants completed the IPII prior to therapy and then rated their experience of the alliance with their therapist on a monthly basis for six months. Participants were divided into three groups according to their baseline Mastery score: high (n=8), intermediate (n=38), and minimal (n= 17). Comparisons of the WAI-S scores revealed that those with high and intermediate levels of Mastery had better scores of therapeutic alliance than those with low Mastery. These findings were interpreted as suggesting that individuals with lesser capacity for Mastery may struggle both to form a psychological problem which might serve as the basis for treatment and have difficulties repairing ruptures that occur in the therapeutic relationship. It was speculated that this mirrors challenges which occur in other relationships. Conclusions This paper has reviewed research on the links between synthetic forms of metacognition with symptoms, neurocognition, other forms of cognition and psychosocial function. Results provide consistent evidence that these deficits are detectable in patient with schizophrenia and that deficits are related to, but not reducible to, symptom severity and poorer neurocognitive function. Independent of symptoms and neurocognition, deficits in synthetic metacognition, particularly in the areas of Self-reflectivity and Mastery, appear related to a range of cognitive phenomena such as reasoning style, learning potential and insight, and may represent a unique impediment to the recovery of social and vocational function for persons in a non-acute phase of illness. There were important issues not discussed. In particular, this paper has not dealt with the issue of etiology.

Paul H. Lysaker et al.

Indeed it is unclear whether these are phenomena that predate the illness and/or whether they can result from a number of different causal influences, including atrophy, loss of cognitive functioning, attachment style or exposure and response to trauma. It was also not discussed how the construct of metacognition converges and diverges with related constructs including social cognition, mentalization and emotional intelligence. There seem to be some areas of overlap, for instance, with all of these constructs being concerned with the detection of emotion and intention, though the construct of metacognition may diverge in its focus on both the development of complex representations of self and others and the use of that to respond to difficulties. We also explored one method for assessing metacognitive activities and it should be noted that there is a range of work focused on autobiographical memory (e.g., 40-42) as well as other methods for rating synthetic metacognitive function (42) that should be considered in future work. Additionally, there are limitations to the work presented here. Most of the studies discussed, though not all (22, 30), were carried out in one laboratory. Participants also tended to be males in a later stage of illness enrolled in treatment. Replication is needed with broader samples including women, persons in an early stage of treatment and others who reject treatment. Long term longitudinal work is also needed to better understand the relationships suggested above as well as work comparing the metacognitive function of persons with schizophrenia to others with different forms of psychopathology. Finally, with replication there may be clinical implications that are important to consider. For one, it may be that promoting recovery in schizophrenia may need more than the provision of support, education and skill remediation. Consistent with Hasson-Ohayon (43), it may be that some with schizophrenia need assistance, possibly from a form of psychotherapy, to integrate information and form their own personal and adaptive accounts of themselves and their challenges, ultimately allowing them to take charge of their lives and find a way to have achieve a fully acceptable quality of life. One viable possibility is to use the MAS-A to both develop treatment targets and track progress in evolving treatments. References 1. Braff DL, Freedman R, Schork NJ, Gottesman II. Deconstructing schizophrenia: an overview of the use of endophenotypes in order to understand a complex disorder. Schizophr Bull 2007;33:21-32. 2. Light GA, Braff DL. Mismatch negativity deficits are associated with poor functioning in schizophrenia patients. Arch Gen Psychiatry

2005;62:127-136. 3. van Os J, Kenis G, Rutten BP. The environment and schizophrenia. Nature 2010;468:203-212. 4. Brüne M, Dimaggio G, Lysaker PH. Metacognition and social functioning in schizophrenia: Evidence, mechanisms of influence and treatment implications. Curr Psychiatr Rev 2011;7:239-247. 5. Lysaker PH, Glynn SM, Wilkness SM, Silverstein SM. Psychotherapy and recovery from schizophrenia: A review of potential application and need for future study. Psychol Serv 2010;7:75-91. 6. Flavell JH. Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. Am Psychol 1979;34:906-911. 7. Semerari A, Carcione A, Dimaggio G, Falcone M, Nicolo G, Procaci M, Alleva G. How to evaluate metacognitive function in psychotherapy? The metacognition assessment scale its applications. Clin Psychol Psychother 2003;10:238-261. 8. Lysaker PH, Hasson-Ohayon. Metacognition in schizophrenia: Introduction to the special issue. Is J Psychiatry Rel Sci 2014;51:4-7. 9. Dimaggio G, Vanheule, V, Lysaker PH, Carcione A, Nicolo G. Impaired self-reflection in psychiatric disorders among adults: A proposal for the existence of a network of semi independent functions. Conscious Cogn 2009;18:653-664. 10. Lysaker PH, Vohs J, Ballard R, Fogley R, Salvatore R, Salvatore G, Popolo R, Dimaggio G. Metacognition, self reflection and recovery in schizophrenia: Review of the literature. Future Neurology, in press. 11. Bacon E, Danion JM, Kauffmann-Muller F, Bruant A. Consciousness in schizophrenia: A metacognitive approach to semantic memory. Conscious Cogn 2001;10:473-484. 12. Bacon E, Izaute M. Metacognition in schizophrenia: processes underlying patients’ reflections on their own episodic memory. Biol Psychiatry 2009;66:1031-1037. 13. Fourneret P, de Vignemont F, Franck N, Slachevsky A, Dubois B, Jeannerod M. Perception of self-generated action in schizophrenia. Cogn Neuropsychiatry 2002;7:139-156. 14. Koren D, Seidman LJ, Poyurovsky M et al. The neuropsychological basis of insight in first-episode schizophrenia: A pilot metacognitive study. Schizophr Res 2004;70:195-202. 15. Koren D, Seidman LJ, Goldsmith M, Harvey PD. Real-world cognitive – and metacognitive – dysfunction in schizophrenia: a new approach for measuring (and remediating) more “right stuff.” Schizophr Bull 2006;32:310-326. 16. Lysaker PH, Clements CA, Plascak-Hallberg CD, Knipscheer SJ, Wright, DE. Insight and personal narratives of illness in schizophrenia. Psychiatry 2002;6:197-206. 17. Lysaker PH, Carcione A, Dimaggio G, Johannesen JK, Nicolò G, Procacci M, Semerari A. Metacognition amidst narratives of self and illness in schizophrenia: Associations with insight, neurocognition, symptom and function. Acta Psychiatr Scand 2005;112:64-71. 18. Lysaker PH, Dimaggio G, Buck KD, Carcione A, Nicolò G. Metacognition within narratives of schizophrenia: Associations with multiple domains of neurocognition. Schizophr Res 2007;93:278-287. 19. Hamm JA, Renard SB, Fogley RL, Leonhardt BL, Dimaggio G, Buck KD, Lysaker PH. Metacognition and social cognition in schizophrenia: Stability and relationship to concurrent and prospective symptom assessments. J Clin Psychol 2012; 68:1303-1312. 20. Lysaker PH, Ringer JM, Buck KD, Grant MLA, Olesek K, Luedtke B, Dimaggio D. Metacognitive and social cognition deficits in patients with significant psychiatric and medical adversity: A comparison between participants with schizophrenia and a sample of participants who are HIV-positive. J Nerv Ment Dis 2012;200:130-134. 21. Lysaker PH, Warman DM, Dimaggio G, Procacci M, LaRocco V, Clark LK, Dike C, Nicolò G. Metacognition in prolonged schizophrenia: Associations with multiple assessments of executive function. J Nerv Ment Dis 2008;19:384-389. 22. Nicolò G, Dimaggio G, Popolo R, Carcione A, Procacci M, Hamm J, Buck

Metacognition in Schizophrenia Spectrum Disorders: Methods of Assessment

KD, Pompili E, Buccione I, Lagrotteria B, Lysaker PH. Associations of metacognition with symptoms, insight, and neurocognition in clinically stable outpatients with schizophrenia. J Nerv Ment Dis 2012; 200:644-647. 23. Delis DC, Kaplan E, Kramer JH. Delis Kaplan Executive Function System: Technical Manual. San Antonio, Texas: The Psychological Corporation, 2001. 24. Kay SR, Fizszbein A, Opler LA. The positive and negative syndrome sale for schizophrenia. Schizophr Bull 1987;13:261-276. 25. Buck KD, Warman DM, Huddy V, Lysaker PH. The relationship of metacognition with jumping to conclusions among persons with schizophrenia spectrum disorders. Psychopathology 2012;45:271-275. 26. Lysaker PH, Erickson MA, Ringer J, Buck KD, Semerari A, Caricione A, Dimaggio G. Metacognition in schizophrenia: The relationship of mastery to coping, insight, self-esteem, social anxiety and various facets of neurocognition. Br J Clin Psychol 2011;50:412-424. 27. Fridberg DJ, Brenner A, Lysaker PH. Intrusion errors in schizophrenia: Association with self-monitoring, symptoms and executive function. Psychiatry Res 2010;179:6-11. 28. Luedtke BL, Kukla M, Renard S, Dimaggio G, Buck KD, Lysaker PH. Metacognitive functioning and social cognition as predictors of accuracy of self-appraisals of vocational function in schizophrenia. Schizophr Res doi: 10.1016/j.schres.2012.02.006. 29. Lysaker PH, Dimaggio G, Buck KD, Callaway SS, Salvatore G, Carcione A, Nicolò G, Stanghellini G. Poor insight in schizophrenia: Links between different forms of metacognition with awareness of symptoms, treatment need and consequences of illness. Compr Psychiatry 2011;52:253-260. 30. Tas C, Brown EC, Esen-Danaci A, Lysaker PH, Brüne, M. Intrinsic motivation and metacognition as predictors of learning potential in patients with remitted schizophrenia. J Psychiatric Res 2012;46:1086-1092. 31. Lysaker PH, Dimaggio G, Carcione A, Procacci M, Buck KD, Davis LW, Nicolo G. Metacognition and schizophrenia: The capacity for self-reflectivity as a predictor for prospective assessments of work performance over six months. Schizophr Res 2010;122:124-130. 32. Lysaker PH, Shea AM, Buck KD, Dimaggio, G, Nicolò G, Procacci M, Salvatore G, Rand KL. Metacognition as a mediator of the effects of impairments in neurocognition on social function in schizophrenia spectrum disorders. Acta Psychiatr Scand 2010;122:405-413.

33. Lysaker PH, Erickson MA, Buck B, Buck KD, Olesek K, Grant MLA, Salvatore G, Popolo R, Dimaggio G. Metacognition and social function in schizophrenia: Associations over a period of five months. Cogn Neuropsychiatry 2011;16:241-255. 34. Lysaker PH, McCormick BP, Snethen G, Buck K, Hamm J, Grant M, Nicolò G, Dimaggio G. Metacognition and social function in schizophrenia: Associations of mastery with functional skills competence. Schizophr Res 2011;131:214-218. 35. Patterson TL, Goldman S, McKibbin CL, Hughs T, Jeste DV. UCSD performance based skills assessment: Development of a new measure of everyday functioning for severely mentally ill adults. Schizophr Bull 2001;27:235-245. 36. Lysaker PH, Dimaggio G, Daroyanni P, Buck KD, LaRocco VA, Carcione A, Nicolò G. Assessing metacognition in schizophrenia with the Metacognition Assessment Scale: Associations with the Social Cognition and Object Relations Scale. Psychol Psychother 2010;83:303-315. 37. Westen D, Lohr N, Silk K, Gold L, Kerber K. Object relations and social cognition in borderlines, major depressives, and normals: A Thematic Apperception Test analysis. Psychol Assessment 1990;2:355-364. 38. Davis LW, Eicher AC, Lysaker PH. Metacognition as a predictor of therapeutic alliance over 26 weeks of psychotherapy in schizophrenia. Schizophr Res 2011;129:85-90. 39. Milne D, Claydon T, Blackburn IM, James,I, Sheikh A. Rationale for a new measure of competence in therapy. Behav Cogn Psychotherapy 2001;29:21-33. 40. Bennouna-Greene M, Berna F, Conway MA, Rathbone CJ, Vidailhet P, Danion JM. Self-images and related autobiographical memories in schizophrenia. Conscious Cogn 2011;21:247-257. 41. Raffard S, D’Argembeau A, Bayard S, Boulenger JP, Van der Linden M. Scene construction in schizophrenia. Neuropsychology 2010;24:608-615. 42. Mitchell LJ, Gumley AI, Reilly ES, Macbeth A, Lysaker PH, Carcione A, Dimaggio G. Metacognition in forensic patients with schizophrenia and a past history of interpersonal violence: An exploratory study. Psychosis, in press. 43. Hasson-Ohayon I. Integrating cognitive behavioral-based therapy with an intersubjective approach: Addressing metacognitive deficits among people with schizophrenia. J Psychotherapy Integration, in press.

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Martin Brüne

Metacognition in Schizophrenia: A Concept Coming of Age Commentary to the Special Issue of the Israel Journal of Psychiatry, edited by Paul H. Lysaker and Ilanit Hasson-Ohayon Martin Brüne, MD, PhD LWL University Hospital Bochum, Division of Cognitive Neuropsychiatry and Psychiatric Preventive Medicine, Germany

Abstract The concept of metacognition in the context of schizophrenia has recently received increasing attention in clinical psychology and psychiatry. Impaired metacognitive abilities in patients with schizophrenia have been associated with symptomatology, cognition, social functioning and quality of life. In this series of articles published as a Special Issue of the Israel Journal of Psychiatry, edited by Paul H. Lysaker and Ilanit Hasson-Ohayon, several new possible ramifications of metacognition with regard to persistence of symptoms, treatment issues and early recognition of schizophrenia are presented, in addition to corroborations of previous findings relating to the aetiology of impaired metacognition and outcome measures. The present invited commentary focuses on potential implications for a more individualized therapeutic approach to schizophrenia, early detection of at-risk states, and ways to link metacognition with neuroscientific approaches.

Introduction The term “schizophrenia,” first mentioned by the Swiss psychiatrist Eugen Bleuler in a scientific meeting in 1908, refers to a clinically heterogeneous group of mental illnesses that are characterized by the presence of delusions, hallucinations, conceptual disorganization and catatonic features. Bleuler considered ambivalence, autism (not to be confused with childhood autism, which was concep-

tualized much later by Leo Kanner and Hans Asperger), affective flattening and the loosening of associations (the four “A”s) as central to these disorders (1). None of these symptoms is, however, mandatory or specific for a diagnosis of schizophrenia. In spite of remarkable progress in the field of psychopharmacology that has enabled clinicians to effectively treat the so-called positive symptoms, namely delusional thinking and perceptual abnormalities, therapeutic challenges remain with regard to behavioral symptoms. In fact, negative symptoms such as avolition, apathy, social withdrawal, as well as cognitive symptoms are at least equally relevant to predict functional outcome and patients’ subjective quality of life than positive symptoms (2). After all, it seems that the efficacy of schizophrenia therapies in general is not only a matter of “classic” symptom reduction, but also closely linked to the amelioration of cognitive and social impairment (3). This is therapeutically even more important in light of evidence for a functional deterioration over the course of the illness (4). That is, chronicity associated with residual negative symptoms affects at least two thirds of patients diagnosed with schizophrenia, whereby negative symptoms respond poorly to antipsychotic medication (5). Moreover, impaired social functioning is already prominent in prodromal stages of schizophrenia, i.e., before the onset of manifest positive symptoms (6). All this calls for non-pharmacological approaches to ameliorate functional impairment among patients with schizophrenia, whereby metacognitive skills have proven a promising target for psychological intervention. In this issue, six articles and an introductory contribution highlight the diagnostic and therapeutic potential of metacognitive approaches to schizophrenia. As becomes clear from the outset, the concept of metacognition has

Address for Correspondence: Professor Dr. Martin Brüne, LWL University Hospital Bochum, Division of Cognitive Neuropsychiatry and Psychiatric Preventive Medicine, Ruhr-University Bochum, Alexandrinenstr. 1, 44791 Bochum, Germany. martin.bruene@rub.de

Metacognition in Schizophrenia: A Concept Coming of Age

so many different facets such that it seems warranted to spend a few lines on definitional aspects. Defining metacognition Historically speaking, it was Bleuler’s frame of reference (1), later elaborated on by Parnas and Bovet (7) and Stanghellini and Ballerini (8), suggesting that autism in schizophrenia concerns a disturbance of intentionality, a disturbance in self-reflectivity, and impairment in intersubjectivity. This approach seems to strongly overlap with the idea of dysfunctional metacognitive capacities in schizophrenia. A particular strength of viewing the symptomatology in light of metacognitive disturbances is that it supports a more individualized approach to the understanding of psychopathological features. Seeing it this way, Lysaker and Hasson-Ohayon (9) rightly emphasize that patients with schizophrenia are “not merely passive subjects of social and biological factors.” Instead, the way people with the condition interpret and try to make sense of their internal and external perceptions is, to a substantial degree, shaped by their individual biographical histories, which may include traumatizing experiences, rejection and stigma, and certainly by the unique quality of early relationships with their primary caregivers (10). All this impacts their thoughts and emotions about themselves and their social environment, as well as the way they think what others think, feel, desire, intend or pretend. Moreover, thinking about one’s own and others’ emotional states and attitudes greatly influences patients’ capacity to cope with symptoms (e.g., persistent suspiciousness) or deal with situations unrelated to the disorder (e.g., finding a street in a street directory). Thus, the core features of metacognition, as they are understood in the majority of papers of this Special Issue, comprise an array of mechanisms that enable an individual to make sense of own and others’ mental states in the context of human interaction, including matters of social rank or social exclusion (11). For the sake of clarity, it needs to be pointed out, however, that divergent definitions of metacognition exist, ranging from the original use of the term in the context of learning, where it described the ability to choose the best learning strategy. From a developmental psychological perspective, metacognition was subsequently framed with regard to its representational nature, highlighting the emergence of the ability to think about thinking (12). In the clinical context, the term metacognition has also 64

been used to describe the ability to cope with stressful thoughts and feelings (13). This aspect of metacognition is somewhat distinct from seeing metacognition as a cognitive skill to monitor one’s own thoughts and to reflect upon one’s personal history, as Parker et al. (14) point out, because metacognition (in the former sense) entails beliefs and attitudes about thinking, such as when one worries about his or her own worrying (14). Finally, metacognition has also been introduced in the attachment literature emphasizing its function to understand that others’ behavior is guided by intentional mental states, an idea that seems more compatible with the broader frame of reference than has been put forth elsewhere (15). Such a definitional plurality of a concept is associated with potential risks and benefits. One possible problem of the conceptual heterogeneity of metacognition could be that the cross-talk between different research groups is hampered by focusing too much on definitional sophistry. Indeed, metacognition seems to broadly overlap with terms such as “social cognition,” “mentalization,” “theory of mind,” especially, if “social cognition” is seen as a multidimensional construct that includes a broader set of mental operations such as emotion recognition, social perception and knowledge, and attributional style (16). With regard to metacognition, Saxe and Offen (17) put forth a useful distinction between “attributive” and “strategic” metacognition, according to which attributive metacognition concerns the ability to attribute beliefs and desires to oneself and others, whereas strategic metacognition concerns the ability to monitor and control ongoing mental operations. Along similar lines, Lysaker and Hasson-Ohayon (9) suggest in this Special Issue that one could distinguish more discrete from more synthetic activities within the spectrum of metacognitive capacities, whereby discrete processes concern, for example, “immediate awareness or accuracy of judgments about one’s own experiences,” whereas synthetic metacognitive operations “involve the organization of and reflect upon increasingly complex and coherent representations of self and others.” Provocatively speaking, this could reflect the fact that metacognition is a chameleon term in psychotherapeutic jargon for mental operations that more or less involve conscious reflection about one’s own and others’ mental operations as well as one’s integrity as a person, whereas social cognition is the umbrella term for similar processes utilized by a more neuroscientifically oriented camp of researchers. Indeed, it seems that metacognition as a

Martin Brüne

concept is more strongly promoted by psychotherapeutically oriented clinicians. And here’s a particular strength of conceptual plurality of metacognition. The syndromes we refer to as “schizophrenia,” it seems, are all too often one-sidedly viewed from a deficit perspective, implying structural brain abnormalities, genetic peculiarities, etc. A deficit perspective, however, leaves little room for symptom reduction by means of psychological intervention. It also disregards the potential of many patients with schizophrenia for coping with adversity, and change of perspective. In contrast to a deficit-oriented view, metacognitive approaches to schizophrenia may help identify psychologically treatable problems and offer patients means to address these in ways that take into account patients’ individual life histories. Diagnostic and therapeutic aspects of metacognition The articles published in this Special Issue nicely illustrate how metacognitive approaches to the understanding and individualized treatment of schizophrenia can be put into practice across a variety of sub-syndromes. Parker and colleagues (14), for instance, focus on changes in metacognitive worry as a predictor of outcome in cognitive therapy. It was hypothesized that change of metacognitive worry was associated with a more positive outcome in patients receiving cognitive therapy. In partial support of this assumption, changes in metacognitive worry were linked with a reduction in hallucinatory experiences, but not with delusional ideation. Brent et al. (10) present ideas about how poor mentalization skills (constituting an important facet of metacognition) could emerge in relation to aberrant attachment in individuals who later develop psychotic disorders. Although partly speculative, this approach offers a possible explanation for a link between childhood adversity (such as abuse or neglect) occurring in the context of early relationships to caregivers, and the functional under-development of metacognitive skills that eventually may lead to “hypermentalizing” in the form of delusional beliefs about intentions and attitudes of significant others towards the self. Interestingly, this theoretical contribution combines a developmental perspective with considerations of neurobiological aspects of schizophrenia, suggesting that parental neglect may lead to an overactivation of the HPA stress system, which in turn may induce a dysregulation of mesolimbic dopaminergic pathways in vulnerable individuals.

Berna and co-workers (18) describe how delusional beliefs can be understood as patients’ attempts to make sense of their experiences, and how closely delusions are linked to anxiety. They used a unique approach using patients’ diary entries, thereby showing that memories of persecutory events were linked with higher levels of negative emotions than memories of non-persecutory events. Memories of persecutory events were also less detailed, but more stereotyped. Importantly, these findings can contribute to the understanding of how delusional beliefs are maintained and integrated into patients’ autobiographical memories. A particularly novel area of research is presented by Scheyer et al. (19) as well as Rabin et al. (20) who examine metacognitive abilities in at-risk states of psychosis, as well as in clinically healthy individuals who differ in regard of the level of schizotypal symptoms. Scheyer et al. (19) define metacognition as a global “meta-level” which helps to monitor and control the correctness of performance in cognitive tasks, that is, the subjective evaluation of one’s cognitive capacity. While this kind of metacognitive skill had moderating properties between cognition and psychosocial functioning in the prodromal stage of psychosis, prodromal subjects displayed superior monitoring skills (i.e., awareness of when one is correct and when one is incorrect) and control sensitivity than non-prodromal subjects. Rabin et al. (20), in contrast, examined metacognition along the definitional lines outlined above. They utilized a now well-established scale for the assessment of metacognitive skills, the Metacognitive Assessment Scale (MAS) following a semi-structured interview that was designed to study metacognition, the Indiana Psychiatric Illness Interview (IPII; details of application of the IPII and the MAS are provided by Lysaker et al. [21], this Special Issue). In particular, Rabin and co-workers (20) addressed the relationship between metacognition and social quality of life. Consistent with previous work, they found that this relationship was mediated by negative, but not positive psychotic symptoms in a sample of patients with manifest schizophrenia. Even more interestingly, a weaker effect in a similar direction was found among non-clinical subjects with schizotypal traits. Specifically, the non-psychotic equivalent trait to negative symptoms, “introvertive anhedonia,” correlated with both metacognitive skills and social quality of life, although the statistical mediation model was non-significant in this group. Despite partially inconclusive results, these latter two studies are of great clinical interest, because 65

Metacognition in Schizophrenia: A Concept Coming of Age

they transfer an established assessment method of metacognitive abilities into the study of at-risk mental states. Given the importance of early detection of at-risk states of schizophrenia and the endeavors of research groups around the world to find (non-pharmacological) ways to reduce transition rates into full-blown psychosis or at least to delay transition and ease the severity of psychosis, I personally believe that the significance of a metacognitive approach to tackle this highly relevant clinical problem can hardly be overestimated. The Special Issue concludes with an overview of the association of metacognition in schizophrenia (spectrum disorders) with symptomatology, cognition and function. Here, Lysaker and colleagues elegantly summarize the evidence for a relationship between self-reflectivity and a variety of neurocognitive measures, symptoms, awareness of illness (insight) and functioning. Of note, metacognitive skills have the potential to predict the development of negative symptoms, which is an invaluable resource for early intervention. If confirmed in different clinical samples and laboratories, one could envisage that metacognitive therapy could help prevent or reduce the development of negative symptoms, which would constitute a leap forward in dealing with the most treatmentresistant symptoms among patients with schizophrenia. Concluding remarks This Special Issue of the Israel Journal of Psychiatry, edited by Paul H. Lysaker and Ilanit Hasson-Ohayon (9), demonstrates the usefulness of studying metacognition in severe mental disorders such as schizophrenia. Examining metacognition in schizophrenia may help diagnosing specific features of this group of disorders that easily escape clinicians’ attention, if not explicitly assessed. For example, failure to detect patients’ difficulties in utilizing their metacognitive skills – be they discrete or synthetic, attributive or strategic – may leave clinicians with an incomplete picture of how to account for lack of awareness of illness, poor adherence to treatment, impaired social functioning, etc. This can be detrimental in light of evidence that these problems do not respond well to antipsychotic medication. In addition, studying metacognition in schizophrenia may assist in tailoring therapy at an individual basis, certainly to the advantage of patients who present with highly diverse life histories, beyond a possibly shared neurobiological background. As already expressed above, this approach seems to be desirable to be implemented in the early recognition of at-risk states of psychosis. 66

What I personally would like to see considered in future research is a stronger foundation of the concept of metacognition in neuroscience. For example, Brent et al. (10) have pointed out the possibility that impaired metacognitive skills in schizophrenia could, in part, be related to an overactivation of the HPA stress system due to aberrant attachment patterns that some patients develop during early infancy and childhood. Along similar lines, one could study the role of neuropeptides in the context of metacognition, given the interaction of oxytocin with the HPA axis, its intimate functional association with the attachment system, and its possible effects on anxiety (22). The link of metacognitive abilities with the default mode network of the brain could be equally interesting, in light of suggestions that mental activity during “resting states” is metacognitive in nature. Since patients with schizophrenia seem to show abnormal resting state patterns (23), it could be interesting to examine whether or not metacognitive training or therapy has the potential to normalize resting state activation patterns in this and other clinical groups. A further question for future research could be to disentangle the schizophrenia spectrum and look more specifically at sub-syndromes, as well as studying metacognition in bipolar disorder, and in special samples such as patients with a forensic background. All this could contribute to establish metacognition as a useful concept in clinical research and to emancipate it, to some degree, from approaches that strongly overlap in content (like social cognition), but do not focus that strongly on those aspects that distinguish metacognition, namely the ability to “reflect upon increasingly complex and coherent representations of self and others” and to utilize this knowledge for problem solving and coping with symptoms (9). In any event, to me it seems that metacognition is an emerging concept that has received increasing acceptance in the clinical literature. Metacognition is coming of age. References 1. Bleuler E. Dementia praecox oder Gruppe der Schizophrenien. Leipzig, Wien: F. Deuticke, 1911. 2. Rabinowitz J, Berardo CG, Bugarski-Kirola D, Marder S. Association of prominent positive and prominent negative symptoms and functional health, well-being, healthcare-related quality of life and family burden: A CATIE analysis. Schizophr Res 2013, Jul 27. doi:pii: S09209964(13)00358-7. 3. Couture SM, Granholm EL, Fish SC. A path model investigation of neurocognition, theory of mind, social competence, negative symptoms and real-world functioning in schizophrenia. Schizophr Res 2011;125:152-160.

Martin Brüne

4. Malaspina D. Schizophrenia: A neurodevelopmental or a neurodegenerative disorder. J Clin Psychiatry 2006;67:e07. 5. Kane JM, Malhotra A. The future of pharmacotherapy for schizophrenia. World Psychiatry 2003;2:81-86. 6. Niendam TA, Jalbrzikowski M, Bearden CE. Exploring predictors of outcome in the psychosis prodrome: implications for early identification and intervention. Neuropsychol Rev 2009;19:280-293. 7. Parnas J, Bovet P. Autism in schizophrenia revisited. Compr Psychiatry 1991; 32:1-15. 8. Stanghellini G, Ballerini M. What is it like to be a person with schizophrenia in the social world? A first-person perspective study on schizophrenic dissociality – Part 1: State of the art. Psychopathology 2008; 44:172-182. 9. Lysaker PH, Hasson-Ohayon I. Metacognition in schizophrenia: Introduction to the special issue. Isr J Psychiatry Relat Sci 2014;51:4-7. 10. Brent BK, Holt DJ, Keshavan MS, Seidman LJ, Fonagy P. MentalizationBased Treatment for psychosis: Linking an attachment-based model to the psychotherapy for impaired mental state understanding in people with psychotic disorders. Isr J Psychiatry Relat Sci 2014;51:17-24. 11. Liotti G, Gilbert P. Mentalizing, motivation, and social mentalities: Theoretical considerations and implications for psychotherapy. Psychol Psychother Theor Res Pract 2011;84:9-25. 12. Flavell JH. Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. Am Psychol 1979,34,906-911. 13. Wells A. Emotional disorders and metacognition: Innovative cognitive therapy. New York: John Wiley and Sons, 2000. 14. Parker S, Wells A, Morrison AP. Metacognitive change as a predictor of outcome in cognitive therapy for psychosis. Isr J Psychiatry Relat Sci 2014;51:8-16. 15. Brüne M, Dimaggio G, Lysaker PH. Metacognition and social functioning in schizophrenia: Evidence, mechanisms of influence and treatment

implications. Curr Psychiatry Rev 2011;7:239-247. 16. Green MF, Olivier B, Crawley JN, Penn DL, Silverstein S. Social cognition in schizophrenia: Recommendations from the measurement and treatment research to improve cognition in schizophrenia new approaches conference. Schizophr Bull 2005;31:882-887. 17. Saxe R, Offen S. Seeing ourselves. What vision can teach us about metacognition. In Dimaggio G, Lysaker PH, editors. Metacognition and severe adult mental disorders: From basic research to treatment. London: Routledge, 2010: pp. 13-29. 18. Berna F, Huron C, Kazès M, Offerlin-Meyer I, Willard D, Verry P, Hédelin G, Krebs M-O, Danion J-M. Chronic persecutory delusion and autobiographical memories in patients with schizophrenia: A diary study. Isr J Psychiatry 2014;51:25-33. 19. Scheyer R, et al. Metacognition in non-psychotic help-seeking adolescents: Associations with prodromal symptoms, distress, and psychosocial deterioration. Isr J Psychiatry Relat Sci 2014;51:34-43. 20. Rabin SJ, Hasson-Ohayon I, Avidan M, Rozencwaig S, Shalev H, Kravetz S. Metacognition in schizophrenia and schizotypy: Relation to symptoms of schizophrenia, traits of schizotypy, and social quality of life. Isr J Psychiatry Relat Sci 2014;51:44-53. 21. Lysaker PH, Leonhardt BL, Pijnenborg M, van Donkersgoed R, de Jong S, Dimaggio G. Metacognition in schizophrenia spectrum disorders: Methods of assessment and associations with neurocognition, symptoms, cognitive style and function. Isr J Psychiatry Relat Sci 2014;51:54-62. 22. Fischer-Shofty M, Brüne M, Ebert A, Shefet D, Levkovitz Y, ShamayTsoory SG. Improving social perception in schizophrenia: The role of oxytocin. Schizophr Res 2013;146:357-362. 23. Garrity AG, Pearlson GD, McKiernan K, Lloyd D, Kiehl KA, Calhoun VD. Aberrant “default mode” functional connectivity in schizophrenia. Am J Psychiatry 2007;164:450-457. Erratum in: Am J Psychiatry 2007;164:1123.

Isr J Psychiatry Relat Sci - Vol. 51 - No 1 (2014)

Reliability and Cross-Validation of the Night Eating Questionnaire (NEQ): Hebrew Version Yael Latzer, DSc,1,2 Orna Tzischinsky, DSc, 3 Michal Hason Rozenstein, PhD,1,2 and Kelly Allison, PhD4 1

Haifa University, Haifa, Israel ED Clinic, Rambam Medical Center, Israel 3 Behavioral Science Department, Emek Yezreel Academic College, Emek Yezreel, Israel 4 University of Pennsylvania School of Medicine, Department of Psychiatry, Philadelphia, Penn., U.S.A. 2

Abstract Background: Several developments in diagnosing night eating syndrome (NES) occurred during the last decade. New proposed diagnostic criteria are now available, and a short Night Eating Questionnaire (NEQ) was published. The study aims were to validate the NEQ for assessing NES, to identify the optimal NEQ cut-point for NES diagnosis, and to validate and test the internal reliability of the translated Hebrew version of the NEQ. Method: 141 participants completed the questionnaire and a diagnostic interview, divided into four groups: NES (n=59), Partial diagnosis PD-NES (n=42), other eating disorders (n=8) and controls (32). Validity was measured by calculating reliability, factor structure, and comparing the interview diagnosis to the NEQ score using different cut scores. Results: Cronbachâ&#x20AC;&#x2122;s alpha was 0.78, and principal components analysis yielded a five factor structure. A cut score of 21 provided the best balance of false and true positive diagnosis. Conclusions: We concluded that the NEQ may be an acceptable screening instrument for assessing NES symptomatology.

This study was conducted in the Department of Social Welfare and Health Sciences, University of Haifa; and in the Eating Disorder Clinic of Rambam Medical Center, Haifa, Israel.

Introduction Night eating syndrome (NES) was first described in 1955 by Stunkard et al. (1). The criteria for NES have been revised and modified several times since the original description by different authors, mainly in the field of eating disorders (1-6). The core features consistently used to characterize NES since 1955 have been evening hyperphagia (EH), morning anorexia, insomnia and/or other sleep disturbances, nocturnal ingestions (NI), and mood disturbances (3), with evening hyperphagia and nocturnal ingestions representing the core features. However, none of the modifications seem to have been formally validated, and only one standardized assessment instrument is available. NES was not an official eating disorder diagnosis in the DSM-IV (7), but had been included as a form of other Feeding and Eating Disorders in the DSM-5 (www.dsm5.org). Variance in the previous definitions of NES has existed across the different features. Evening hyperphagia has been defined differently with regard to the amount of food eaten and the time-frame involved more than 25 percent of daily caloric food intake after the evening meal (8, 9). Nocturnal ingestion (nocturnal eating) has been defined as waking up at night to eat. It is often associated with feeling a strong urge to eat during the night and/or holding the belief that eating is necessary in order to initiate falling asleep or to continue sleeping ( 9, 10). Morning anorexia has been defined with regard to whether breakfast is eaten and/or if hunger is present in the morning (4, 8, 11-13). Finally, sleep disturbances have been described mainly as initial or maintenance insomnia (4, 6, 11, 12), and mood disturbances have included evening tension and/or overall depressed mood, sometimes with mood worsening in the evening (7, 9, 11).

Address for Correspondence: Michal Hason Rozenstein, PhD, Rambam Medical Center, Haaliya 9602, Haifa 31096, Israel â&#x20AC;&#x2020; michalhason@gmail.com, m_hason@rambam.health.gov.il

Yael Latzer et al.

An important differential diagnosis for NES is sleep related eating disorder (SRED) (13, 14). SRED is described as a parasomnia that consists of arising from sleep and eating, with reduced consciousness and partial awareness, reduced level of recall (partial or total amnesia) for night eating, and ingestion of unusual or even non-food items. This has been addressed with the introduction of a proposed set of research criteria for NES (15). NES Diagnostics The need to recognize NES with official diagnostic criteria has become increasingly important. Thus, participants in the First International NES Symposium (April 2008) brought together a set of provisional diagnostic criteria for NES (15). The diagnostic criteria describe a significantly increased caloric intake in the evening and/or night-time, characterized by consuming at least 25% of food intake, and /or experiencing at least two episodes of nocturnal eating per week (criterion A). Awareness and recall of evening and nocturnal eating episodes are present, as a means of differentiating NES from Sleep Related Eating Disorder (criterion B). In addition the daily pattern manifested by at least three of the following features: morning anorexia, a strong urge to eat between dinner and sleep onset and/or during the night, insomnia, a belief that one must eat in order to sleep, and depressed and/or mood worsens in the evening (criterion C). The disorder is associated with significant distress (criterion D), is maintained for at least three months, and is not secondary to medical or psychiatric disorder (criterion E). These proposed criteria currently represent the most updated reference used to diagnose this syndrome. NES Assessment Stunkard and colleagues (6) started developing a questionnaire to screen for the presence of night eating in the 1990s. The first version was a 9-item self-report Night Eating Questionnaire (NEQ) that was not published or validated. Its items assessed: morning anorexia (2 items), evening hyperphagia (1 item), initial insomnia (1 item), mid-phase insomnia (1 item), nocturnal ingestions (1 item), and mood (3 items). Van der Wal and colleagues (16) evaluated the diagnostic utility of various combinations of questions taken from this early version of the NEQ for the recognition of NES. They found that the screening questionnaire had acceptable sensitivity across the various definitions of NES but poor specificity. The

authors suggested that NEQ lacked a cut-point that would result in both high sensitivity and specificity and, thus, a better screening tool for NES assessment was needed. The latest version of the NEQ was validated by Allison et al. in 2008 (17). The NEQ is comprised of 14 items assessing morning hunger and timing of first food consumption (2 items), food cravings and control over eating behavior both before bedtime (2 items) and during night-time awakenings (2 items), percentage of food consumed after dinner (1 item), initial insomnia (1 item), frequency of nocturnal awakenings and ingestion of food (3 items), mood disturbance (2 items), and awareness of nocturnal eating episodes (1 item). This latter item is not included in the total NEQ score as an indicator of severity of NES in the English version, but rather used as a means of differentially diagnosing NES as compared to SRED. Therefore scores range from 0 to 52. The original English version of the NEQ (17) was shown to have a Cronbach’s alpha of 0.70, and three factors were identified: Nocturnal Ingestions, Evening Hyperphagia and Morning Anorexia. Clinical cut-points were also evaluated, with a cut of 25 yielding a modest positive predictive value of 41% and a higher cut of 30 yielding a positive predictive value of 73%. The NEQ was translated into Spanish and validated in Spain recently, yielding a Cronbach’s alpha of 0.79 and a similar four-factor solution as the English version (18). Similarly, a Portuguese translation of the NEQ in Brazil yielded a Cronbach’s alpha of 0.78 (19). Finally, the NEQ was translated into French in Canada and adapted for children, using a self-report NEQ and a parent-reported NEQ (20). Cronbach’s alphas were 0.54 for the child version, and 0.55 for the parent-report version. The only published semi-structured interview for NES is the Night Eating Syndrome History and Inventory (NESHI) (21). It contains questions assessing the schedule and the amount of food eaten in 24 hours in a typical day, history of NES symptoms, sleep patterns, mood and stressors, weight, diet history, and previous treatment strategies. The NEQ items are embedded in the NESHI and can yield a score (17). Its items track onto the research diagnostic criteria so that a diagnosis can be determined. Given the increasing international use of the NEQ, we sought to translate and evaluate this instrument for use in Hebrew. Our first aim was to validate the NEQ as a brief questionnaire for assessing NES and to identify the optimal NEQ cut-point for NES diagnosis according to the newly proposed diagnostic criteria among an Israeli cohort. Our second aim was to validate and test 69

Reliability and Cross-Validation of the Night Eating Questionnaire (NEQ): Hebrew Version

the internal reliability of the translated Hebrew version of the NEQ. In order to evaluate these aims, we compared different classifications of individuals with NES by their NEQ scores, in comparison to an interview-based diagnosis constructed from the 2010 criteria (15) in a community sample. Method Participants, Instruments and Procedure

One hundred and forty-one adults between the ages of 18-65 were recruited for the study. All participants were recruited via internet advertisement, as well as from individuals seeking nutritional treatment for weight loss. We separated the participants into two groups: a clinical group and a control group. The total clinical group consisted of 109 participants who fit one of the two core features of NES according to the 2010 diagnostic criteria. These two features were defined as consumption of at least 25% of intake after the evening meal (evening hyperphagia) and/or nocturnal awakenings with ingestions at least twice per week (nocturnal ingestions) (15). The clinical group was divided into three subgroups: 1) The NES group (n=59; 44 females, 15 males) met full criteria for NES according to the 2010 criteria; 2) the Partial Diagnosis of NES (PD-NES) group (n=42; 25 females, 17 males) met partial criteria for NES, missing the NES diagnosis by meeting only two instead of three clinical features required in criterion C; and 3) Others (n= 8; 7 females, 1 male), who answered “yes” to one of the two main questions, and fulfilled the diagnostic criteria for a different type of eating disorder, rather than one of the two NES subgroups (either bulimia nervosa or binge eating disorder). The control group consisted of 32 participants (25 females, 7 males) who did not suffer from NES or any other eating or psychiatric disorders. The study was approved by the Helsinki Committee of Rambam Medical Center and the ethical committee IRB at the University of Haifa. Prior to undergoing the clinical interview, participants were informed about the purpose of the study and they provided written informed consent. The participants then completed the self-report NEQ. Subsequently, participants were interviewed individually using the NESHI (21), administered by trained clinicians: a clinical social worker, a clinical dietitian, and a clinical psychologist. For the current study, the NESHI was translated into Hebrew, and back-translated into English. 70

The NEQ was also translated into Hebrew and backtranslated into English. The NEQ total score was calculated by reverse coding items 1, 4, and 14, and summing all items (15). The rationale for excluding item 13 (assessing awareness of nocturnal ingestions) in previous studies was that it was used only to differentiate NES cases from SRED cases, but did not contribute to the severity of a particular case of NES. We chose to include the awareness item in the total score in the current study. Participants reporting no awakenings at night and not meeting the criteria for evening hyperphagia were excluded after the interview (due to a sleeping problem rather than eating problem). Thus, the awareness item became relevant for those who do eat at night, so we chose to include it. However, we examined the impact of this decision; our analyses were run with and without item 13 in the total score, with no significant changes in outcomes. Statistical Analyses

Descriptive statistics were used to characterize the sample and a histogram was created to show the distribution of NEQ scores for the total sample. Internal consistency was measured using the Cronbach’s alpha procedure. A principal components analysis with Varimax rotation was conducted to extract the factor structure for the Hebrew version of the NEQ. If items loaded above .4 on more than one factor, the higher of the two scores was used if there was a separation of greater than .10. Cross loadings will be discussed. In order to assess the validity of NEQ, we made a cross validation and compared the outcome of the interview based on the 2010 diagnostic criteria to the NEQ score. First, using Analysis of Variance (ANOVA) we tested whether there were differences in the NEQ scores across the clinical and control groups. Then, using crosstabs (observed vs. expected), we created tables of classification results for three cut-points (i.e., 18, 21, 22). Thus, the scores tested ranged between -0.5 SD and +0.5 SD. We chose to test three different cut-offs that were chosen based on the mean of the entire sample, to see whether they differed from each other in term of true and false positive diagnosis. We also chose them in order to present different scores that can be used for different purposes, such as screening health service or for scientific purposes, which are different in their aims. These cut-scores were used according to the frequency testing the scores that were most frequent (mean +- 1 Std) (see Figure 1). This comparison yields four possible outcomes: 1. True Positive, which results when the NEQ cut-point correctly

Yael Latzer et al.

Results

Figure 1. NEQ descriptive distribution (n=141).

No significant differences were found for age or BMI between groups. In addition no significant differences were found in gender distribution within each group. However, in the Partial Diagnosis of NES (PD-NES) group, there were more males than would be expected compared to the other groups, yet this finding was not statistically significant [Chi-square = 5, p < 0.17]. See Table 1 for groups’ demographic information. As previous research did not include the awareness item in the total score, it is important to note that the total mean score for all participants using only the 13 items was 21.3, SD = 9.03. The 14 item mean total score was 22.8, SD = 10.3. Means across diagnostic groups for the 14-item scale are presented in Table 1.

mean = 22.83 std. dev. = 10.285

Frequency

0 0.00

10.00

20.00

30.00

40.00

50.00

60.00

total score (14 items)

identifies people with NES; 2. True Negative, which results when the NEQ identifies people without NES as not having NES; 3. False negative, which results when the NEQ does not identify people with NES as having the syndrome; and 4. False positive, which results when the NEQ mistakenly identifies people without NES as having the syndrome. Theoretically a low cut-score would yield more true positives and false positives, as opposed to a high cut-score, which would yield more false negatives and true negatives. The choice of an optimal cut-score depends on the purpose of diagnosis. However, theoretically, it is important to know which cut-score represents the syndrome at the most accurate rate. For each cut-score examined, we compared the classification of groups based on the four possible outcomes. We also used Chi-Square tests to compare the groups according to their NEQ scores, and calculated the eta (i.e., the % of the explained variance) to test which cut-score best explained the differences between the groups.

Factor Structure Cronbach’s alpha of the 14-item Hebrew version of the NEQ was 0.78. Principal components analysis with Varimax rotation revealed a 5 factor structure that explained 72% of the scale’s variance. The first factor, Nocturnal Ingestions, explained 25% of the variance and included items 8-12 and 14 (initial insomnia, frequency of awakenings, cravings at night, belief in needing to eat to sleep, frequency of nocturnal eating, and control over nocturnal eating). The second factor, Morning/Evening Eating, explained 14% of the variance and included items 1, 2, 5, and 6 (morning hunger, time of first meal, percentage of food consumed after dinner, and mood). However, item 1 loaded below the .40 level and is considered a weak contributor to the scale. The third factor, Evening Cravings, explained 12% of the variance and included two items, 3 and 4 (cravings and control for eating after dinner). The fourth factor, Awareness, explained 11% of the variance and contained only the awareness item, #13. The fifth factor, Evening Mood, explained 10% of the variance and included items 7 and 8, the initial insomnia item (cross-loaded with Nocturnal Ingestions), and the diurnal variation of mood.

Table 1. Mean (SD) of demographic and NEQ scores described by group Age

BMI

Mean±SD

Range

Mean±SD

Range

NEQ score

Control

41.64±11.44

26-64

25.14±5.17

17.37-39.45

11.47±7.09

32 (21.9% male)

NES

39.94±13.26

19-74

29.88±8.11

18.75-46.78

27.81±7.38

59 (25.4% male)

PD-NES

41.74±14.47

18-65

28.69±6.93

18.41-45.84

20.76±6.38

42 (40.5% male)

Other

49±10.41

36.5-63

30.41±4.31

24.07-35.44

21.38±13.25

8 (12.5% male)

Reliability and Cross-Validation of the Night Eating Questionnaire (NEQ): Hebrew Version

Examination of Cut Scores ANOVA revealed a significant effect by diagnosis [F(3,140) = 33.54, p < .001]. Bonferroni post hoc comparisons revealed significant differences between all group pairs, except between the PD-NES and “Other” group, probably due to the high variance in the “Other” group. This means that the NES group differed from the PD-NES group and the controls, and that the PD-NES group differed from the control, as well. Eta squared was 0.423. As the “Others” group included only eight participants, we excluded them from further analysis. The second stage of the validation analysis differentiated among the three remaining groups by NEQ cut points (see Table 2 for classification outcomes). Eta was used to investigate the strength of the association between diagnostic group and NEQ cut-scores. Discussion The need to recognize and define NES leads to the creation of the newly proposed diagnostic criteria (15). The NEQ is the most widely validated scale for assessing NES (19-21) and can be administrated easily as a self-report questionnaire (17). The purpose of this research was to cross-validate and test the internal consistency of the Hebrew version Table 2. Classification outcomes according to the NEQ True Positive

True Negative

False Negative

False Positive

Cut-score = 18 Controls NES

87.5% 93.2%

PD-NES

12.5% 6.8%

28.6%

71.4%

Cut-score = 21 Controls NES

90.6% 88.1%

PD-NES

9.4% 11.9%

59.5%

40.5%

Cut-score = 25 Controls NES PD-NES

93.8% 67.8%

6.2% 32.2%

78.6%

21.4%

Cut-score 18: χ2(3) = 61.72, p < .001. Eta for group dependent =0.298, Eta for NEQ-cut-score 18 = 0.662 Cut-score 21: χ2(3) = 57.13, p < .001. Eta for group dependent =0.104, Eta for NEQ-cut-off 21 = 0.637. Cut-score 25: χ2(3) = 40.7, p < .001. Eta for group dependent =0.052, Eta for NEQ-cut-score 25 = 0.537

of the NEQ according to the newly proposed criteria, and identify the optimal NEQ cut-point when screening for the presence of NES. The main results of this yielded high internal consistency at an alpha of 0.78 and a five factor structure. The cut score of 21 provided the best balance of false and true positive diagnosis. In this first translation and reliability study of the NEQ in Hebrew, the results showed relatively high reliability using the 14 items of the NEQ, which was similar to alphas found in the Spanish (18) and Portuguese (19) translations. A five-item factor structure was extrapolated. The first factor, Nocturnal Ingestions factor, was quite similar to the same factor in the English version (17) except for the addition of the initial insomnia item. This insomnia item cross-loaded on the fifth factor (Evening Mood) which was also where it loaded in the English version. The second factor, Morning/Evening Eating is similar to the Morning Anorexia factor in the English version, while the third factor, Evening Cravings, is similar to the Evening Hyperphagia factor in the English version, with the notable omission of item 5 (percent of calorie intake after dinner) in the Hebrew version. In the English version, item 5 cross loaded on these two factors. In the Hebrew version, it appears on only factor 2. The fourth factor, Awareness, is unique to the Hebrew version. It is interesting that it contains only one item, supporting its use as a tool to differentiate NES from SRED, but not as a symptom that goes hand in hand with other aspects of NES. The final factor, Evening Mood, is similar to the Mood/Sleep factor of the English version. As for validity, three different cut-scores were tested. The first cut-score of 18 revealed a high level of true positives for NES, the price being a high false positive rate, particularly for the PD-NES group which present similar symptoms of NES but were missing only one of the NES diagnosis criteria. Therefore a cut-score of 21 may be a more conservative approach to identify NES in community samples using this Hebrew version. When using the NEQ for clinical purposes we would suggest using 18 as a cutscore, because it would misclassify only 6.8% of NES patients. Clinicians may find this cut-point useful since it would also help them identify individuals who are at risk for NES and refer them for further evaluation. On the opposite end of the range of cut-scores, a score of 25 may be useful for research purposes of screening participants for pure NES. Because this is the first study in Israel with the NEQ, we are careful to not endorse a single cut-off point, but instead call for further research

Yael Latzer et al.

on the topic. We do, however, suggest using the NEQ as a standard screening self report inventory for dietitians and physicians, who should probably use the lowest cut-off point – 18 in order to refer patients for further diagnosis. The study results also indicated that the PD-NES group reported lower mean NEQ scores than those of the NES group, and higher than those of the controls, which is in the direction intended. The resulting cut scores from this study are lower than those reported by Allison et al. (22), despite using all 14 items in the total score in this current study. This difference suggests that caution should be used when attempting to generalize cut-scores for measures across different cultures and with different translations. However, it is in line with Allison et al.’s suggestion that different cut-scores may be warranted when using the same measure depending on the purposes and setting. One of the study’s limitations is the small sample size which did not allow for computation of the different cut-points among those who have nocturnal ingestion, evening hyperphagia, or both. In addition, we are unable to identify which is the best cut-point among people who suffer from NES and who seek treatment for eating disorders, as they were not included in the sample. These limitations call for further research. In conclusion, the NEQ is a short assessment instrument that can be efficiently administered in a variety of populations and settings and can be used for different purposes using a different cut of point. Although the NEQ was found to be an efficient instrument for identifying NES, for clinical evaluation, we suggest using the Night Eating Syndrome History and Inventory version (NESHI) interview for further in-depth diagnosis for those who reach cut-score of 21 or greater. The NESHI reviews the typical timing and content of intake across 24-hours. Similarly, the NEQ total score, which was confirmed to be a valid indicator of a higher-order construct of night eating in this study, may be used for an index of severity, or the factor scales may be examined separately if particular aspects of NES are of interest. Further research should focus on determining its validity and utility for different populations, including eating disorder samples, and subgroups of those with NES. Reference 1. Stunkard AJ, Grace WJ, Wolff HG. The night-eating syndrome: A pattern of food intake among certain obese patients. Am J Med 1955;19:78-86. 2. Allison KC, Latzer Y, Tzischinsky O, Vinai P. What is night eating syndrome? New directions towards unified definition. Int J Child Adolesc

Health 2008;2:1-10. 3. de Zwaan M, Roerig DB, Crosby RD, Karaz S, Mitchell JE. Night-time eating: A descriptive study. Int J Eat Disord 2006;39:224-232. 4. Birketvedt GS, Florholmen J, Sundsfjord J, Osterud B, Dinges D, Bilker W, et al. Behavioral and neuroendocrine characteristics of the nighteating syndrome. JAMA 1999 ;282:657-663. 5. Napolitano MA, Head S, Babyak MA, Blumenthal JA. Binge eating disorder and night eating syndrome: Psychological and behavioral characteristics. Int J Eat Disord 2001;30:193-203. 6. Stunkard A, Berkowitz R, Wadden T, Tanrikut C, Reiss E, Young L. Binge eating disorder and the night-eating syndrome. Int J Obes Relat Metab Disord 1996;20:1-6. 7. American Psychiatric Association Committee on Nomenclature, and Statistics. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), Fourth Edition. Washington, DC: American Psychiatric Association, 1994. 8. Ceru-Bjork C, Andersson I, Rossner S. Night eating and nocturnal eating - two different or similar syndromes among obese patients? Int J Obes Relat Metab Disord 2001;25:365-372. 9. Rand CS, Macgregor AM, Stunkard AJ. The night eating syndrome in the general population and among postoperative obesity surgery patients. Int J Eat Disord 1997;22:65-69. 10. Tzischinsky O, Latzer Y. Nocturnal eating prevalence, features and night sleep among Binge Eating Disorders and Bulimia Nervosa patients in Israel. Eur Eat Disord Rev 2004;12:101-109. 11. Kuldau JM, Rand CSW. The night eating syndrome and bulimia in the morbidly obese. Int J Eat Disord 1986;5:143-148. 12. Gluck ME, Geliebter A, Satov T. Night eating syndrome is associated with depression, low self-esteem, reduced daytime hunger, and less weight loss in obese outpatients. Obes Res 2001;9:264-267. 13. Schenck CH, Hurwitz TD, Bundlie SR, Mahowald MW. Sleep-related eating disorders: Polysomnographic correlates of a heterogeneous syndrome distinct from daytime eating disorders. Sleep 199;14:419-431. 14. Schenck CH, Hurwitz TD, O’Connor KA, Mahowald MW. Additional categories of sleep-related eating disorders and the current status of treatment. Sleep 1993;16:457-466. 15. Allison KC, Lundgren JD, O’Reardon JP, Geliebter A, Gluck ME, Vinai P, et al. Proposed diagnostic criteria for night eating syndrome. Int J Eat Disord 2010;43:241-247. 16. Van der Wal JS, Waller SM, Klurfeld DM, McBurney MI, Dhurandhar NV. Night eating syndrome: Evaluation of two screening instruments. Eat Behav 2005;6:63-73. 17. Allison KC, Lundgren JD, O’Reardon JP, Martino NS, Sarwer DB, Wadden TA, et al. The Night Eating Questionnaire (NEQ): Psychometric properties of a measure of severity of the Night Eating Syndrome. Eat Behav 2008;9:62-72. 18. Moizé V, Gluck ME, Torres F, Andreu A, Vidal J, Allison K. Transcultural adaptaion of the Night Eating Questionnaire (NEQ) for its use in the Spanish population. Eat Behav 2012;13:260-263. 19. Harb AB, Caumo W, Hidalgo MP. Translation and adaptation of the Brazilian version of the Night Eating Questionnaire. Cad Saude Publica 2008;24: 1368-1376. 20. Gallant AR, Lundgren J, Allison K, Stunkard AJ, Lambert M, O’Loughlin J, Lemieux S, Tremblay A, Drapeau V. Validity of the night eating questionnaire in children. Int J Eat Disord. 2012;45:861-865. 21. Lundgren JD, Allison KC, Vinai P, Gluck ME. Assessment instruments for night eating syndrome. In JD Lundgren, KC Allison, AJ Stunkard, editors. Night Eating Syndrome: Research, assessment, and treatment. New York: Guilford, 2012: pp. 197-217. 22. Allison KC, Engel SG, Crosby RD, de Zwaan M, O’Reardon JP, Wonderlich SA, et al. Evaluation of diagnostic criteria for night eating syndrome using item response theory analysis. Eat Behav 2008;9:398-407.

Book Reviews

Winnicott’s Children edited by Ann Horne & Monica Lanyado London: Routledge, 2012

he twelve chapters of this book encompass Winnicott’s understanding of children and the impact of this understanding on therapeutic work with children and adolescents. Each chapter, written by a senior member of the British Association of Psychotherapists, describes that particular clinician’s method of treating children following Winnicott’s thinking. The Foreword by Helen Tailor Robinson takes us directly to the heart of Winnicott’s thinking and his profound understanding of infants and mother-infant interactions. Robinson emphasizes that “Winnicott’s children” include “…therapists who have derived from, and descended from, him and his work, and who have also grown up enough to use him…” (p. xvi). Adam Phillips, in his prologue “On reading Winnicott,” enlightens us as to how Winnicott’s writing should be understood. “… It is important that we do not read Winnicott too literally… It is that Winnicott is not really talking about mothers and babies, but he is also using mothers and babies to talk about other things as well: both sexuality and a person’s relationship themselves…” (p. XX) In their Introduction, Ann Horne and Monica Lanyado consider the twelve chapters in light of the book’s three parts: Concepts, concerning Winnicott’s thinking about communication; Transition, meaning transitional space in therapy and in supervision; and The Outside World, for example, children in residential schools or community intervention programs. Chapter 1: “Winnicott in his time,” by Caldwell and Joyce, describes Winnicott’s thinking as the leading theorist of the Independent Group in British psychoanalysis, with reference to Freud, Ferenczi, Melanie Klein, Anna Freud, and Bowlby. The authors of this chapter go on to describe Winnicott’s work, for example, the therapeutic consultation, his theory about delinquency, and so on. A leitmotif of this chapter is Winnicott’s notion of the internal analytic setting maintained by the analyst who is available to the patient (child, parent or adult) regardless of external circumstances. Chapter 2, “What is therapeutic about communication?” by Lanyado, emphasizes the way Winnicott communicated with his patients from a position of deep 74

understanding, as well as his way of communicating his ideas to a variety of audiences through various media, from BBC radio programs to lectures before the British Association of Psychotherapists to his books. In this chapter Lanyado refers to communication as a mode of making contact and relating which gives the other (the patient, the baby) the feeling that he or she is understood. Without this deep feeling of being understood, the baby, for survival reasons, complies with the world, which eventually leads to the formation of a False Self. The consequences of difficulties in the type of communication which is deeply understanding of the other are manifested in intimate relations in the form of a sense of rejection. Lanyado gives an example of a therapy with a 13-year-old boy in which the therapist “survived” the boy’s negativity towards her, preparing the stage for this boy to dare to communicate. Another example is of a therapy with another boy whose mother was emotionally deprived herself. Chapter 3, “A joy to be hidden, a disaster not to be found” by Kitchener, deals with children at various developmental stages and across a range of clinical presentations, whose “private self ” is not communicating. The hidden communication and the threat of not being understood give this chapter its name. Chapter 4, “Reflection on mirrors” by Colloms, focuses on Winnicott’s ideas about the mirroring function of the mother, giving several examples, most of which concern physical handicaps that might interfere with the mirroring function. Chapter 5, “Hate in the counter-transference” by Dowling, refers to Winnicott’s landmark paper, which later appeared as a chapter in his book Through Paediatrics to Psycho-Analysis and which sheds light upon and legitimizes the therapist’s negative countertransference as a necessary part of his or her work. The author describes how hate in the counter-transference enabled her to create an empathic connection with abused and deprived children and their parents even when they were full of hate towards her as a transference object. The author’s ideas are derived from her rich clinical experience, and make an important contribution to therapists working with deprived and abused children and their parents who find themselves experiencing mixed and sometimes strongly negative feelings towards their patients. Chapter 6, “Body and soul” by Horne, is concerned with trauma experienced at the pre-verbal stage, as described by Winnicott in his paper, “Mind and its rela-

Book Reviews

tion to psych-soma.” It describes the sensitive understanding of trauma worked out in the body and acted out as a repetition compulsion. Chapter 7, “On psychoanalytic supervision: Avoiding omniscience, encouraging play” by Vastardis and Phillips, opens the second part of the book and portrays the supervision process, the supervisor-supervisee relationship, boundaries concerning supervision, and the patient-therapist-supervisor triangle. Chapter 8, “Transition and change” by Lanyado, presents this author’s view that the meditative states of mind are an area similar to transitional space, which she views as a “therapeutic transitional experience” felt by the patient as the “presence of the therapist.” The therapist’s ability to be in a meditative state of mind is particularly helpful in the treatment of severely traumatized or deprived children. Chapter 9 opens the third part of the book with “Space for growth” by Onions and Browner, who describe psychotherapy combined with milieu therapy conducted in a therapeutic residential school for primary-aged children with severe emotional disturbances. Chapter 10, “A word in your ear” by MelvilleThomas, describes her weekly radio broadcasts on the subject of children and adolescents. Chapter 11, “The adolescent, the therapist and the school environment” by Alexander, describes the

author’s therapeutic attitude towards treating adolescents, in which she encourages therapists “not to pathologise unnecessarily.” She shows how this view affects her considerations in her school work as to when to refer an adolescent to a clinic or other agencies and when to intervene in the school setting. Chapter 12, “On delinquency” by Horne, is based on Winnicott’s view of delinquency as a sign of hope. The author describes therapies with delinquent adolescents who are viewed as struggling against their early experiences of “not being wanted at the beginning of life.” Clear suggestions for intervention are presented at the end of the chapter. This book is a rich and inspiring collection for psychotherapists in general and for child and adolescent therapists in particular. It highlights the dilemmas and concerns met by child and adolescent therapists in their work, and consider them in the context of Winnicott’s thinking. Authors of various professions from the British Independent Group have contributed from their rich experience in order to “translate” Winnicott’s thinking into practice. This book will appeal to those who work with children and their parents in child and family mental health settings, regular and residential schools, and social welfare settings. Ricky Finzi-Dottan

‫ מצב חלקי של‬,)‫ משתתפים‬59( ‫ קבוצת תסמונת אכילה לילית‬‫) וקבוצת ביקורת‬8( ‫ הפרעות אכילה אחרות‬,)42( ‫התסמונת‬ ‫ ובדיקת‬,‫ נערכו בדיקת מהימנות פנימית וניתוח גורמים‬.)32( ‫תוקף על פי השוואת מסקנות הריאיון לתוצאות השאלון תוך‬ .‫בדיקת נקודות חתך שונות‬ ‫וניתוח הגורמים הצביע‬0.78 ‫ אלפא קרונבך נמצאה‬:‫תוצאות‬ ‫ נמצאה המאוזנת‬21 ‫ נקודת החתך של‬.‫על חמישה גורמים‬ .‫ביותר מבחינת אבחנת אמת של התסמונת‬ ‫ שאלון תסמונת אכילה לילית הוא כלי סקירה יעיל‬:‫מסקנות‬ ‫ושימושי לאיתור סימפטומים של תסמונת אכילה לילית ויכול‬ .‫להיות בשימוש בקהילה ובמרכזים לטיפול בהפרעות אכילה‬

‫מהימנות ותיקוף צולב של שאלון להערכת‬ ‫ גרסה עברית‬:‫תסמונת אכילה לילית‬

‫ חיפה‬,‫ אליסון‬.‫ חסון רוזנשטיין וק‬.‫ מ‬,‫ צ'ישינסקי‬.‫ א‬,‫ לצר‬.‫י‬

‫ בשנים האחרונות חלו כמה התפתחויות באבחון תסמונת‬:‫רקע‬ ,‫ לאחרונה נקבעו קריטריונים חדשים לאבחנה‬.‫האכילה הלילית‬ ‫ ופורסם ותוקף שאלון להערכת תסמונת אכילה‬,‫המוצעים למחקר‬ ‫ מטרות המחקר הנוכחי הן לתקף את תרגום השאלון‬.‫לילית‬ ‫ לזהות את נקודת החתך האופטימלית באוכלוסייה‬,‫לעברית‬ .‫המקומית ולבדוק את המהימנות הפנימית של השאלון‬ ‫ משתתפים מילאו את השאלון ועברו במקביל‬141 :‫שיטה‬ ‫ הם חולקו לארבע קבוצות על פי הריאיון‬.‫ריאיון קליני אבחנתי‬

List of reviewers for Israel Journal of Psychiatry, 2013 The Editors would like to thank the following for their contribution as reviewers of manuscripts during 2013

Henry Abraham

Carl Feinstein

Andrea Letamendi

Shiri Sadeh Sharvit

Henry Abramovitch

Shmuel Fennig

Itzhak Levav

Isaac Sakinofsky

Idan Aderka

Sylvana Fennig

Nava Levit Binnun

Aya Secker-Einsbecker

Abraham Adunsky

Gabriele Fischer

Shaul LevRan

Harold Sgan-Cohen

Kelly Allison

Annette Gallant

Alon Liberman

Gaby Shefler

Paul Appelbaum

Marc Gelkopf

Pesach Lichtenberg

Gal Shoval

Alan Apter

Saralee Glasser

Rachel Liebman

Lea Shelef

Rachel Bachner-Melman

Michal Granot

Mark Lukowitsky

Edward Shorter

Yoram Barak

Jon Grant

Ido Lurie

Shaul Shreiber

Yehuda Baruch

Alexander Grinshpoon

Paul Lysaker

Emi Shufman

Haim Belmaker

Cornelius Grop

David Mankuta

Zahava Solomon

Yechayaou Beloosesky

Leon Grunhaus

Binyamin Maoz

Eli Somer

Gershon Ben Shachar

Ilanit Hasson-Ohayon

Yuval Melamed

Dan Stein

Moshe Bensimon

Helen Herrman

Beth Merenstein

Rael Strous

Yael Benyamini

Jonathan Huppert

Roberto Mester

Tom Trauer

Howard Berger

Iulian Iancu

Piper Myers

Anne-Marie Ulman

David Blass

Yulia Kartalova-O’Doherty

Hanan Munitz

Onno van der Hart

Avi Bleich

Gregory Katz

Ora Nakash

Gary Walter

David Brent

Arad Kodesh

Susan Ochoa

Michael Weingarten

Anat Brunstein-Klomek

Natan Kellerman

Femi Oyebode

Myrna Weissman

Orna Chishinski

Sean Kidd

Avi Peled

Laura Widman

Rena Cooper-Kazaz

Robert Kohn

Einat Peles

Vicky Wing

Giancarlo Dimaggio

Anatoly Kreinin

Alexander Ponizovsky

Eliezer Witztum

Zippi Dolev

Jesper Krogh

Paul L. Prather

Zvi Zemishlany

Adiel Doron

Sefi Kronenberg

Nicky Pugh

Nava Zisapel

Adrienne Einarson

Julian Leff

Jamie Ringer

Shlomo Zusman

Smadar Even Tov

Stephen Levine

Hamidreza Roohafza

Liana Fattore

Hilik Levkovitz

Amir Rotem

‫מטה־קוגניציה בקרב מתבגרים מחפשי‬ ‫עזרה שאינם פסיכוטיים‪ :‬מידת הקשר שלה‬ ‫לסימפטומים פרודרומליים‪ ,‬מצוקה וירידה‬ ‫בתפקוד הפסיכו־חברתי‬

‫ר‪ .‬שייר‪ ,‬נ‪ .‬רזניק‪ ,‬מ‪ .‬אדרס‪ ,‬א‪ .‬אפטר‪ ,‬ל‪ .‬ג'‪ .‬סיידמן וד‪ .‬קורן‪ ,‬חיפה‬

‫מטרה‪ :‬לבחון את הרעיון שלפיו קשיים בתפקוד המטה–‬ ‫קוגניטיבי הם מאפיין מרכזי פרה–פסיכוטי של סכיזופרניה‬ ‫והפרעות נוספות בספקטרום‪ .‬שיטה‪ :‬שבעים ושמונה מתבגרים‬ ‫מחפשי עזרה שאינם פסיכוטיים (בני ‪ )18-13‬עברו הערכה‬ ‫באמצעות שאלון פרודמלי (‪,)Prodromal Questionnaire - PQ‬‬ ‫ריאיון מובנה לסימפטומים פרודרומליים ‪Structured Interview‬‬ ‫‪for Prodromal Syndromes‬נ(‪ ,)SIPS‬שני סולמות לתפקוד חברתי‬ ‫ואקדמי‪ ,‬גרסה מטה–קוגניטיבית של שתי מטלות למדידת‬ ‫קוגניציה לא חברתית (זיכרון מילולי ותפקודים ניהוליים)‬ ‫ושתי מטלות למדידת קוגניציה חברתית (תפיסת רגשות‬ ‫בפרצופים ו–‪ .)Theory of Mind‬נוסף להעברה הסטנדרטית של‬ ‫המטלות‪ ,‬המשתתפים התבקשו להעריך את רמת הביטחון‬ ‫שלהם בנכונות של כל תשובה שנתנו‪ ,‬ולבחור אם הם רוצים‬ ‫"להחשיב" את התשובה שלהם לצורך חישוב הציון להערכת‬ ‫התפקוד שלהם במטלה‪ .‬כל תשובה שהנבדק החליט להחשיב‬ ‫קיבלה בונוס של ‪ ₪ 1‬אם הייתה נכונה‪ ,‬או קנס של ‪ ₪ 1‬אם‬ ‫הייתה לא נכונה‪.‬‬ ‫תוצאות‪ :‬רמות התפקוד הקוגניטיבי והמטה–קוגניטיבי‬ ‫לא היו שונות באופן מובהק בין משתתפים בקבוצת הסיכון‬ ‫הגבוה לבין נבדקים בקבוצת הסיכון הנמוך‪ .‬עם זאת‪ ,‬הניבוי‬ ‫של התפקוד הפסיכו–חברתי היה מובהק כאשר המדדים‬ ‫החדשים המטה–קוגניטיביים נוספו למדדים הקוגניטיביים‬ ‫והקוגניטיביים–חברתיים המקובלים‪.‬‬ ‫דיון‪ :‬תוצאות אלה מאתגרות את הספציפיות של הקשר בין‬ ‫ליקויים נוירו–קוגניטיביים לבין סיכון לסכיזופרניה‪ .‬עם זאת‪,‬‬ ‫עולה מהם כי ייתכן שלמטה–קוגניציה תפקיד חשוב במיתון‬ ‫הקשר בין נוירו–קוגניציה לבין סוגים שונים של תפקוד‪ ,‬גם‬ ‫בשלב שלפני הפריצה האקוטית של המחלה‪.‬‬ ‫מטה־קוגניציה בסכיזופרניה וסכיזוטיפיה‪:‬‬ ‫הקשר לסימפטומים של סכיזופרניה‪ ,‬מאפיינים‬ ‫סכיזוטיפיים ואיכות חיים חברתית‬

‫ש‪.‬י‪ .‬רבין‪ ,‬א‪ .‬חסון־אוחיון‪ ,‬מ‪ .‬אבידן‪ ,‬ס‪ .‬רוזנצוויג‪ ,‬ה‪ .‬שלו‬ ‫וש‪ .‬קרביץ‪ ,‬רמת גן‬

‫רקע‪ :‬במחקר זה נבחן קיומו של מודל תיווך‪ ,‬שלפיו סימפטומים‬ ‫של מחלת הסכיזופרניה ומאפיינים סכיזוטיפיים מתווכים את‬ ‫הקשר החיובי בין מטה–קוגניציה לבין איכות חיים חברתית‬ ‫בקרב אנשים הסובלים מסכיזופרניה ואנשים ללא מחלת נפש‪.‬‬ ‫שיטה‪ :‬במחקר השתתפו ‪ 39‬אנשים שאובחנו כמתמודדים‬ ‫עם סכיזופרניה ו–‪ 60‬אנשים ללא מחלת נפש‪ .‬הכלים שבהם‬ ‫נעשה שימוש הם ריאיון עומק חצי מובנה (‪ )A-MAS‬להערכת‬ ‫מטה–קוגניציה‪ QLI-MH ,‬לבדיקת איכות חיים‪ LIFE-O ,‬לבחינת‬ ‫‪77‬‬

‫וה–‪PANSS‬‬

‫רמות הסכיזוטיפיה בקרב אנשים ללא מחלת נפש‪,‬‬ ‫עבור אנשים הסובלים מסכיזופרניה‪.‬‬ ‫תוצאות‪ :‬אנשים הסובלים מסכיזופרניה הפגינו רמות מטה–‬ ‫קוגנטיביות ירודות ודיווחו על איכות חיים חברתית נמוכה‬ ‫בהשוואה לאנשים ללא מחלת נפש‪ .‬בקרב אנשים הסובלים‬ ‫מסכיזופרניה‪ ,‬נמצא כי הסימפטומים השליליים מתווכים את‬ ‫הקשר החיובי בין היכולת להבנת האחר לבין איכות החיים‬ ‫החברתית‪ .‬ברם‪ ,‬על אף שעבור אנשים ללא מחלת נפש‪ ,‬ליכולת‬ ‫להבין את האחר מתאם שלילי עם אנהדוניה ועם איכות חיים‬ ‫חברתית‪ ,‬לא אושש מודל תיווך עבור אוכלוסיה זו‪.‬‬ ‫דיון‪ :‬ממחקר זה עולה כי היכולת להבין את האחר‬ ‫קשורה באופן חיובי לאיכות החיים החברתית בקרב שתי‬ ‫אוכלוסיות המחקר‪ .‬נוסף לכך‪ ,‬סימפטומים שליליים של מחלת‬ ‫הסכיזופרניה ושל אנהדוניה שהיא המקבילה הסכיזוטיפית‬ ‫לסימפטומים השליליים‪ ,‬קשורים אף הם לאיכות חיים חברתית‬ ‫בקרב אנשים הסובלים סכיזופרניה ובקרב אנשים ללא מחלת‬ ‫נפש‪ ,‬בהתאמה‪ .‬מחקר זה‪ ,‬שבו לא אושש קיומו של מודל‬ ‫תיווך עבור אנשים ללא מחלת נפש‪ ,‬מחזק תיאוריות קודמות‬ ‫אשר אינן רואות את הסכיזוטיפיה כתמונת ראי של מחלת‬ ‫הסכיזופרניה‪ ,‬ושלפיהן רמות סכיזוטיפיות גבוהות אינן מעידות‬ ‫בהכרח על התפרצות של מחלת הסכיזופרניה‪ .‬נערך דיון על‬ ‫מגבלות המחקר ועל כיוונים למחקרים בעתיד‪.‬‬ ‫מטה־קוגניציה בספקטרום הפרעות הסכיזופרניה‪:‬‬ ‫שיטות הערכה ואסוציאציות עם נוירו־קוגניציה‪,‬‬ ‫סימפטומים‪ ,‬סגנון קוגניטיבי ותפקוד‬ ‫פ‪.‬ה‪ .‬לייסקר‪ ,‬ב‪.‬ל‪ .‬לאונהרדט‪ ,‬מ‪ .‬פיננבורג‪ ,‬ר‪ .‬ון דונקרסגוד‪,‬‬ ‫ס‪ .‬דה יונג‪ ,‬ג‪ .‬דימג'יו ורודבוש‪ ,‬ארה"ב‬

‫אפשר להגדיר ליקויים ביכולת המטה–קוגניטיבית בקרב‬ ‫הסובלים מסכיזופרניה כנמצאים על רצף מסוים ‪ -‬מפעילויות‬ ‫דיסקרטיות יותר עד לפעילויות סינטטיות יותר‪ .‬אמנם כל ליקוי‬ ‫מייצג מוקד מחקר חשוב באותה מידה‪ ,‬אך פעילויות מטה–‬ ‫קוגניטיביות סינטטיות עשויות להיות קשות יותר למדידה‪,‬‬ ‫היות שהן כרוכות בהערכה של מורכבות החשיבה יותר מאשר‬ ‫בדיוק קונקרטי‪ ,‬ולכן הן זכו לתשומת לב פחותה‪.‬‬ ‫סקירה זו מסכמת מחקר בנושא מטה–קוגניציה סינטטית‬ ‫באמצעות פרדיגמה‪ ,‬שלפיה היכולת המטה–קוגניטיבית נמדדת‬ ‫בנרטיבים אישיים‪ .‬תוצאות העבודות שנסקרו מספקות ראיות‬ ‫לכך שאפשר לגלות את הליקויים הללו בקרב חולי סכיזופרניה‬ ‫ושהליקויים קשורים לחומרת הסימפטומים ולתפקוד נוירו–‬ ‫קוגניטיבי ירוד יותר‪ ,‬אך לא רק אליהם‪.‬‬ ‫ללא קשר לסימפטומים ולנוירו–קוגניציה‪ ,‬ליקויים במטה–‬ ‫קוגניציה קשורים למגוון רחב של פעילויות שכליות‪ ,‬לרבות‬ ‫סגנון חשיבה‪ ,‬פוטנציאל למידה ותובנה‪ .‬הליקויים הללו עלולים‬ ‫להשפיע גם על התוצאות לטווח הארוך בשל השפעתם על‬ ‫יכולת התפקוד במסגרות עבודה‪ ,‬ועל היכולת לחשוב על קשרים‬ ‫חברתיים ולקיימם‪.‬‬

‫כתב עת ישראלי‬ ‫לפסיכיאטריה‬ ‫תקצירים‬ ‫שינוי מטה־קוגניטיבי כמנבא תוצאות של טיפול‬ ‫קוגניטיבי במקרים של פסיכוזה‬ ‫ס‪ .‬פרקר‪ ,‬ע‪ .‬וולס וא‪ .‬מוריסון‪ ,‬מנצ'סטר‪ ,‬אנגליה‬

‫רקע‪ :‬מעט מאוד ידוע על מנבאים של תגובות לטיפול קוגניטיבי‬ ‫(‪ )CT‬במקרים של פסיכוזה‪ .‬מחקר זה בודק את ההשערה כי‬ ‫שינוי מטה–קוגניטיבי בסוף הטיפול ולאורך זמן קשור לתוצאות‬ ‫חיוביות בקרב אנשים הסובלים מפסיכוזה ומקבלים טיפול‬ ‫קוגניטיבי‪.‬‬ ‫שיטה‪ :‬למטופלים הסובלים מפסיכוזה אשר הופנו לקבלת‬ ‫טיפול קוגניטיבי הוצע טיפול קוגניטיבי למשך ‪ 30‬מפגשים לכל‬ ‫היותר‪ .‬הערכות הכוללות ראיונות למדידת סימפטומים ושאלון‬ ‫מטה–קוגניטיבי המודד דאגה‪ ,‬נעשו לפני ואחרי קבלת הטיפול‬ ‫הקוגניטיבי וכן במעקב לאחר שנה‪ .‬ממצאי המחקר מתייחסים‬ ‫לנתונים שהתקבלו מ־‪ 32‬מטופלים‪.‬‬ ‫תוצאות‪ :‬נמצאו קשרים חיוביים מובהקים בין שינויים‬ ‫מטה–קוגניטיביים בדאגה במעקב לאחר שנה לבין סימפטומים‬ ‫חיוביים של פסיכוזה וממדים ספציפיים של הזיות‪ .‬לא נמצא‬ ‫קשר מובהק בין שינוי מטה–קוגניטיבי לבין סימפטומים של‬ ‫הזיות‪.‬‬ ‫מסקנות‪ :‬ממצאי המחקר מעידים על־כך ששינוי מטה–‬ ‫קוגניטיבי קשור לשינויים בסימפטומים לאורך זמן‪ .‬ממצאים‬ ‫אלו בולטים במיוחד בקרב אנשים אשר חווים סימפטומים‬ ‫של הזיות‪ .‬בהמשך נערך דיון על המגבלות המתודולוגיות של‬ ‫המחקר ועל ההשלכות הקליניות שלו‪.‬‬ ‫טיפול מבוסס מנטליזציה בפסיכוזה‪:‬‬ ‫חיבור בין מודל מבוסס התקשרות‬ ‫לבין פסיכותרפיה לצורך הבנת המצב‬ ‫המנטלי הלקוי בקרב אנשים‬ ‫הסובלים מהפרעות פסיכוטיות‬

‫ב‪.‬ק‪ .‬ברנט‪ ,‬ד‪.‬ג'‪ .‬הול‪ ,‬מ‪.‬ס‪ .‬קשאבאן‪ ,‬ל‪.‬ג'‪ .‬סיידמן ופ‪ .‬פונאגי‪,‬‬ ‫מסצ'וסטס‬

‫נמצא כי הפרעות ביכולת לבצע מנטליזציה (היכולת לחשוב על‬ ‫מצבים נפשיים של הפרט ושל אחרים) קשורות לסימפטומים‬ ‫ולליקוי בתפקוד של אנשים הסובלים מהפרעות פסיכוטיות‪.‬‬

‫‪israel journal of‬‬

‫‪psychiatry‬‬ ‫כרך ‪ ,51‬מס' ‪2014 ,1‬‬

‫הועלתה השערה כי פסיכותרפיה‪ ,‬שבמהלכה יתמקדו בהבנה‬ ‫של הפרט ושל אחרים (למשל טיפול מבוסס מנטליזציה ‪-‬‬ ‫‪ ,)MBT‬עשויה לסייע להחלמה מפסיכוזה‪.‬‬ ‫במאמר זה תוצג פרשנות מבוססת התקשרות לליקויים‬ ‫ביכולת לבצע מנטליזציה‪ .‬במאמר יתואר מודל נוירו–פסיכולוגי‬ ‫המחבר בין שיבושים ביכולת לבצע מנטליזציה הקשורים‬ ‫להפרעות בסביבה המטפלת לבין הפיזיופתולוגיה של פסיכוזה‬ ‫בקרב אנשים המצויים בסיכון מבחינה גנטית לסבול מכך‪.‬‬ ‫בהמשך מוצגות דוגמאות לכמה מן הטכניקות הבסיסיות‬ ‫של טיפול מסוג ‪ MBT‬לצורך שיקום היכולת לבצע מנטליזציה‪,‬‬ ‫כפי שהן מיושמות בטיפול במטופל הסובל מהפרעה‬ ‫פסיכוטית‪.‬‬ ‫הזיות רדיפה כרוניות וזיכרונות אוטוביוגרפיים‬ ‫בקרב חולי סכיזופרניה‪ :‬מחקר יומן‬

‫פ‪ .‬ברנה‪ ,‬ק‪ .‬הורן‪ ,‬מ‪ .‬קזיס‪ ,‬א‪ .‬אופרין־מאיר‪ ,‬ד‪ .‬ווילרד‪ ,‬פ‪ .‬וורי‪,‬‬ ‫ג‪ .‬הדלין‪ ,‬מ‪.‬קרבס וג'‪ .‬דניון‪ ,‬שטרסבורג‪ ,‬צרפת‬

‫רקע‪ :‬הזיות רדיפה כרוניות מעוגנות בדרך כלל בחיי היומיום‬ ‫של מטופלים‪ ,‬אך טרם נערכו מחקרים על האופן שבו מצבים‬ ‫הקשורים לתחושות רדיפה נשמרים בזיכרון ועל האופן שבו הם‬ ‫נשלפים בשלב מאוחר יותר‪.‬‬ ‫שיטה‪ :‬מחקר של יומן בשילוב עם משימת זיהוי נערך בקרב‬ ‫עשרה מטופלים המתמודדים עם סכיזופרניה‪ ,‬אשר סובלים‬ ‫מהזיות רדיפה כרוניות‪ ,‬ובקרב עשרה משתתפים בקבוצת‬ ‫ביקורת‪ .‬יומנים של אירועי רדיפה יום–יומיים ושל אירועים‬ ‫שאינם אירועי רדיפה נרשמו‪.‬‬ ‫תוצאות‪ :‬בשתי הקבוצות התקבלו התוצאות הבאות‪)1 :‬‬ ‫אירועי רדיפה יום–יומיים היו קשורים לציוני חרדה גבוהים‬ ‫יותר מאשר אירועים שאינם אירועי רדיפה; ‪ )2‬אירועי רדיפה‬ ‫יום–יומיים נחוו כפחות מיוחדים וכיותר סטריאוטיפיים‬ ‫מאירועים שאינם אירועי רדיפה; ‪ )3‬תדירות הזיהוי השגוי של‬ ‫תיאורים של אירועי רדיפה יום–יומיים שהשתנו‪ ,‬הייתה גבוהה‬ ‫יותר מזו של אירועים שאינם אירועי רדיפה‪.‬‬ ‫מגבלות‪ :‬בשל המוטיבציה הגבוהה שנדרשת מיומנאים‪,‬‬ ‫המדגם הנוכחי קטן‪.‬‬ ‫מסקנה‪ :‬זיכרונות של אירועי רדיפה היו מאוד רגשיים‬ ‫וסמנטיים‪ .‬לעתים קרובות הם זוהו באופן שגוי‪ ,‬דבר שמצביע‬ ‫על קיומה של הטיה כתוצאה מאינטראקציות בין עיבוד לבין‬ ‫הזיות הרדיפה‪.‬‬ ‫‪78‬‬