RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY April 2014 Nicola L. G. Marshall Thesis submitted in part fulfilment of the degree of Doctorate in Clinical Psychology University of East Anglia © This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognise that its copyright rests with the author and that use of any information derived there from must be in accordance with current UK Copyright Law. In addition, any quotation or extract must include full attribution. RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY ii RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY iii Abstract Aims and Objectives The primary aim of the research was to investigate the nature of the relationships between self-compassion, self-esteem and schizotypy using a non-clinical sample. A secondary aim was to investigate the mechanisms which help to explain any relationships found. In utilising a non-clinical sample the study aimed to determine whether relationships exist between the variables prior to the onset of psychosis within a continuum approach to schizotypy. A final objective was to identify specific correlates of self-compassion and schizotypy through detailed subscale analyses. Method The study utilised a quantitative, cross-sectional design. Participants completed self-report questionnaires via a secure website host measuring: self-compassion, global self-esteem, and trait schizotypy. A total of 93 participants took part in the research. Results As predicted, highly significant negative correlations were determined between self-compassion and schizotypy, and between self-esteem and schizotypy. With respect to the mechanisms through which these variables were related, self-compassion was not found to moderate the relationship between self-esteem and schizotypy. However, self- compassion and schizotypy were found to be related via both a direct and an indirect route, which was mediated by self-esteem. Conclusions The study is the first to investigate the nature of the relationships amongst self- compassion, self-esteem and schizotypy in a non-clinical population, utilising the schizotypy construct as an analogue of the psychosis continuum. The findings indicated that there may be both a direct, and an indirect route through self-esteem, RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY iv which accounted for the relationship between self-compassion and schizotypy. The results mirror associations determined within clinical populations. The authors argues that in utilising schizotypy as an analogue of the psychosis continuum the results of this study provide evidence that self-esteem and self-compassion may reflect underlying mechanisms which could underpin schizotypal symptomatology. RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY v Acknowledgements In memory of my supervisor Professor Malcolm Adams. Many thanks to Dr Sian Coker and Dr Margo Ononaiye for steadying the ship and making the transition to a new thesis supervisor seem effortless. My deepest gratitude goes to Dr Joanne Hodgekins, for taking over the supervision of this project. Her knowledge of the subject area and feedback on drafts of this thesis has been invaluable, thank you. To my beautiful children - Sylvester and Ivan. Mummy’s little book is now finished and we can all get back to family life without distraction! Finally, and most importantly, I must thank my wonderful Husband Nathan for his encouragement and unending support when times were tough. RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY vi Chapter One Introduction 1 1.1 Overview 1 1.2 Schizotypy 1 1.2.1 Definitions of schizotypy 1 1.2.1.1 Types of symptoms 3 1.2.1.1.1 Cognitive-perceptual symptoms 3 1.2.1.1.2 Interpersonal symptoms 3 1.2.1.1.3 Disorganisational symptoms 4 1.2.1.2 Quasi-dimensional versus fully-dimensional approaches 4 1.2.1.2.1 Quasi-dimensional 4 1.2.1.2.1.1 Epidemiological data 5 1.2.1.2.1.2 Cognitive markers 5 1.2.1.2.2 Fully-dimensional approach` 6 1.2.1.2.2.1 Non-help-seeking epidemiology 7 1.2.1.2.2.2 Methodological issues 7 1.2.1.2.2.3 Cognitive markers 8 1.2.1.2.2.3.1 Theory of mind 8 1.2.1.2.2.3.2 Contextual integration 9 1.2.1.2.2.3.3 Heightened threat monitoring 11 1.2.1.2.2.3.4 Emotional dysregulation 11 1.2.1.2.2.3.5 Longitudinal data 12 1.2.1.2.3 Psychometric approaches to assess schizotypy 13 1.2.1.2.3.1 Schizotypal personality scales 14 1.2.1.2.3.1.1 Quasi-dimensional scales 14 1.2.1.2.3.1.2 Fully-dimensional scales 15 1.2.1.2.3.1.3 Attenuated symptoms scales 15 1.2.1.2.3.1.4 Prodromal symptom scales 16 1.2.2 Epidemiology 17 1.2.3 Aetiology 18 1.2.4 Summary 18 1.3 Self-esteem 20 1.3.1 Definition of self-esteem 20 1.3.1.1 Global versus domain specific self-esteem 20 1.3.1.2 Temporal lability 20 1.3.2 Aetiology 21 1.3.2.1 Fennell’s (1999) model of self-esteem 21 1.3.2.2 Self-esteem and psychopathology 23 1.3.2.2.1 Self-esteem and schizotypy 23 1.3.2.2.2 Self-esteem and psychosis 25 1.3.2.2.3 Self-esteem in psychosis maintenance 26 1.3.2.2.4 Self-esteem as consequence of psychosis 26 1.3.3 Summary 27 1.4 Self-compassion 28 1.4.1 Definition of self-compassion 28 1.4.2 Aetiology 29 1.4.2.1 Social mentality theory 31 1.4.2.2 Neurobiological theory 32 RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY vii 1.4.3 Self-compassion and psychopathology 32 1.4.3.1 Anxiety and depression 32 1.4.3.2 Psychosis 33 1.4.3.3 Self-compassionate therapy for psychosis 34 1.4.4 Structure of the literature review 35 1.4.4.1 Inclusion and exclusion criteria 35 1.4.4.2 Mindfulness based approaches 35 1.4.4.3 Compassion focussed therapy 38 1.5 Conclusions and development of research hypotheses and questions 44 1.5.1 Research questions and hypotheses 48 1.5.1.1 Hypothesis 1 48 1.5.1.2 Hypothesis 2 48 1.5.1.3 Research question 1 48 1.5.1.4 Research question 2 48 1.5.1.5 Hypothesis 3 48 1.5.1.6 Hypothesis 4 49 1.5.1.7 Hypothesis 5 49 Chapter two Method 50 2.1 Design 50 2.1.1 Hypothesis 1 50 2.1.2 Hypothesis 2 50 2.1.3 Research question 1 50 2.1.4 Research question 2 51 2.1.5 Hypothesis 3 51 2.1.6 Hypothesis 4 51 2.1.7 Hypothesis 5 51 2.2 Participants 51 2.2.1 Sample size calculation 52 2.2.2 Inclusion and exclusion criteria 53 2.2.3 Response rate 53 2.2.4 Participant characteristics 53 2.2.4.1 Excluded cases 54 2.3 Measures 54 2.3.1 Schizotypy 55 2.3.1.1 Oxford-Liverpool Inventory of Feelings and Experiences 55 2.3.2 Self-compassion 56 2.3.2.1 Self-Compassion Scale 56 2.3.3 Self-esteem 57 2.3.3.1 Rosenberg Self-Esteem Scale 57 2.4 Ethical considerations 58 2.5 Procedure 59 2.6 Plan of analysis 60 2.6.1 Hypothesis 1 60 2.6.2 Hypothesis 2 60 2.6.3 Hypotheses 3-5 60 RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY viii 2.6.4 Research question 1 61 2.6.4.1 Grand mean centering 62 2.6.4.2 Multicollinearity 62 2.6.4.3 Calculating the variable interaction term 62 2.6.4.4 Forced-entry regression analysis 62 2.6.4.5 Simple slopes analysis 63 2.6.5 Research question 2 63 Chapter three Results 65 3.1 Excluded data sets 65 3.2 Descriptive statistics 65 3.3 Distribution of scores 66 3.4 Bivariate Analyses Exploring Hypotheses 1 and 2 67 3.5 Moderator analysis investigating Research Question 1 68 3.5.1 Results of regression analysis 68 3.5.2 Simple slopes analysis 69 3.6 Analyses investigating Research Question 2 69 3.7 Analyses investigating Hypotheses 3-5 71 3.7.1 Hypothesis 3 71 3.7.2 Hypothesis 4 72 3.7.3 Hypothesis 5 73 3.8 Post-hoc analyses 73 3.9 Summary of results 73 Chapter four Discussion 75 4.1 Summary of findings 75 4.2 Relevance to the literature 76 4.2.1 Hypothesis 1 76 4.2.2 Hypothesis 2 78 4.2.3 Research question 1: Moderation 80 4.2.4 Research Question 2: Mediation 82 4.2.4 Hypotheses 3-5 84 4.2.4.1 Hypothesis 3 84 4.2.4.2 Hypothesis 4 85 4.2.4.3 Hypothesis 5 86 4.3 Strengths and limitations of the study 87 4.3.1 Design 88 4.3.2 Sample 88 4.3.2.1 Non-clinical sample 88 4.3.2.2 Age 90 4.3.2.3 Gender 91 4.3.2.4 Ethnicity 91 4.3.2.5 Education level 92 4.3.2.6 Response rate 92 4.3.2.7 Responder bias 93 4.3.3 Measures 94 RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY ix 4.3.3.1 Self-compassion scale 94 4.3.3.2 Oxford-Liverpool Inventory of Feelings and Experiences 95 4.3.3.3 Rosenberg Self-Esteem Scale 96 4.3.3.4 Other variables 96 4.3.3.4.1 Depression 96 4.3.3.4.2 Stigma 97 4.4 Clinical implications 97 4.5 Future research directions 99 4.6 Summary 101 References 103 Appendices 119 Word count: 27, 690 RELATIONSHIPS AMONGST SELF-COMPASSION, SELF-ESTEEM AND SCHIZOTYPY x List of Tables and Figures Table 1 Results of systematic literature search Table 2 Descriptive statistics Table 3 Linear model of predictors of schizotypy Table 4 Correlation matrix for subscale analyses Table 5 Pearson’s correlation matrix for male participants Table 6 Pearson’s correlation matrix for female participants Table 7 Pearson’s correlation matrix for under 25 years old participants Table 8 Pearson’s correlation matrix for over 26 years old participants Figure 1 Interaction between three major affect regulation systems Figure 2 Summary of Pearson’s correlation coefficients Figure 3 Model of self-compassion as a predictor of schizotypal symptomatology mediated by self-esteem Figure 4 Diagram of conceptual mediation effect