BRIEF REPORT Self-Organization in Bipolar Disorder: Replication of Compartmentalization and Self-Complexity Y. Alatiq • C. Crane • J. M. G. Williams • G. M. Goodwin Published online: 12 May 2010 Ó Springer Science+Business Media, LLC 2010 Abstract Remitted bipolar patients were compared to remitted unipolar patients and healthy controls on a self concept task assessing degree of self-compartmentalization (clustering of self-aspects based on valence) and self- complexity (the degree of relatedness versus differentiation across self-aspects). Similar to the findings of Taylor et al. (Cognitive Therapy and Research, 31(1), 83-96, 2007) the bipolar and unipolar groups showed higher levels of self compartmentalization than healthy controls. No differences were found on the self complexity measure. Keywords Bipolar disorder Á Self Á Compartmentalization Á Self-complexity Introduction Although bipolar disorder is characterized by extreme shifts in self evaluation between different affective states, research into the self-concepts of people with bipolar dis- order has largely been restricted to the single dimension of self-esteem. A number of authors have argued, however, that viewing the self as a unitary, one-dimensional structure is not sufficient in explaining the diversity of people’s emotions and behaviour. Instead, they have proposed that the self should be viewed as a multi-dimensional, multi- faceted dynamic phenomenon that includes a set of images, schemas or concepts. For example one self could include a set of different aspects such self as a student, as a brother, as a friend and as a football player. (e.g. Carver & Scheier 1981; Epstein 1980; Markus & Wurf 1987). Each aspect of the self contains a set of specific beliefs which vary in their degree of elaboration, valence and importance. Each aspect also contains affective and cognitive components, attached to it, mediating information processing and affect regula- tion (Markus 1977; Markus & Nurius 1986; Markus & Wurf 1987). Two models of self-concept are of interest in the study of bipolar disorder because of their potential to explain mood swings in general. First is the self-compartmentali- sation model (Showers 1992a, 1992b), which refers to the organisation of positive and negative information about the self. An ‘integrated-self’ reflects the ability to integrate inversely valenced information within self-aspects so that each self-aspect contains a mixture of positive and negative beliefs. For example one sees his self as a student to be curious and expressive but also tense and easily distracted. In contrast to this, the ‘compartmentalised’ self reflects the tendency to cluster information of the same valence toge- ther in separate self-aspects so that each self aspect con- tains primarily positive or primarily negative information. For example one sees his self as a student to be curious, motivated and creative (purely positive) another sees his self as a student to be tense, insecure and moody (purely negative). According to this model, the organisation of self- knowledge is believed to influence the accessibility of positive or negative beliefs about the self. In compart- mentalised organisation, life events that activate purely negative self-aspects will increase the accessibility of exclusively negative beliefs about the self, leading to increased dysphoric mood. With an integrated organisa- tion, however, the same event will activate a self-concept with a mixture of positive and negative beliefs about the self, leading to a more balanced reaction. The Y. Alatiq (&) Á C. Crane Á J. M. G. Williams Á G. M. Goodwin Department of Psychiatry, Warneford Hospital, University of Oxford, Warneford Lane, Oxford, Headington OX3 7JX, UK e-mail: yalatiq@gmail.com 123 Cogn Ther Res (2010) 34:479–486 DOI 10.1007/s10608-010-9315-1