Adult attachment syle and vulnerability to depression

This thesis explores the utility of Attachment Theory (Bowlby, 1969; 1973; 1980) as a framework for understanding both the personality and cognitive processing styles associated with depression and vulnerability to depression. In two separate but related studies, the present investigation identified depressive personality characteristics and depressive cognitive processing styles associated with each of the styles of adult attachment defined by Bartholomew and Horowitz's (1991) four-category attachment model. Using self-report data from a sample of 305 respondents (225 female; 80 male), Study 1 explored associations among each of the adult attachment styles and sociotropic and autonomous personality-based vulnerabilities to depression (Beck, 1983). Extending the work of Zuroff and Fitzpatrick (1995; Study 2), associations among the attachment styles and the specific components of the two depressive personality styles, as defined by Robins et al. (1994), were examined. Using cued recall of autobiographical memories for a subsample of 44 of the same respondents (35 female; 9 male), Study 2 explored attachment style group differences in autobiographical memory retrieval, thereby identifying specific depressive cognitive processing biases associated with each adult attachment style. Past research has focused on the assessment of memory accessibility, using either recall latency (Mikulincer & Orbach, 1995; Mikulincer, 1998a) or memory generality (Tasker, MacLeod & Maynard, 1996) to indicate accessibility. In the present study, both latency and generality were used as indicators of memory accessibility, and the content of memories was also examined. In addition, use of a four-category rather than a three-category measure of attachment style permitted distinction of cognitive biases for the fearful-avoidant and dismissive-avoidant styles. The present findings support a notion that the two adult attachment styles understood to be characterised by a negative self-view, namely fearful and preoccupied attachment, are associated with depressive vulnerability. Replicating previous findings (Carnelley, Pietromonaco & Jaffe, 1994), both the fearful and preoccupied styles were associated with state depression and with perceptions of negative parental bonding. The fearful style in particular was associated with perceptions of 'affectionless control' in childhood. As the major focus of this thesis, the fearful and preoccupied attachment styles were each associated with both personality styles and cognitive processing deficits which have previously been linked with depression and depressive vulnerability. In terms of personality styles (Study I), the fearful attachment style was broadly associated with the autonomous personality style, whereas the preoccupied attachment style was broadly associated with the sociotropic personality style, as previously demonstrated by Zuroff and Fitzpatrick (1995; Study 2). By focusing on associations with the individual components of the autonomous and sociotropic vulnerabilities, as defined by Robins et al. (1994), the more specific 'depressive' personality characteristics were identified for both these insecure attachment styles. In particular, the findings suggest that a fearful attachment style involves both avoidant and self-critical characteristics, whereas a preoccupied attachment style involves both dependent and self-critical characteristics. In terms of cognitive processing deficits (Study 2), fearful individuals had difficulty accessing autobiographical memories, indicated by delayed memory recall and reduced memory specificity across a range of memory cues. Indeed, these individuals demonstrated a globalised 'mnemonic interlock' typical of depressed individuals (Williams, 1996) and their memories were predominantly negative in content. In contrast, preoccupied individuals demonstrated a ruminative cognitive processing style, indicated by relatively fast recall of predominantly specific and negative memories across a range of memory cues. However, these individuals had difficulty accessing personally-relevant memories of abandonment, arguably due to both encoding and retrieval deficits emanating from their hypersensitivity to abandonment experiences. Dismissive attachment was associated with the avoidant but not the self-critical aspect of the autonomous personality style and involved a repressive cognitive processing style. This repressive style was indicated by delayed recall of specific negative memories. Nonetheless, unlike the fearful and preoccupied styles, dismissive attachment was not shown to be associated with state depression, suggesting that an avoidant attachment style does not necessarily contribute to depressive vulnerability. Instead, dismissive individuals' use of defensive repression of negative affects and memories appears to be an effective coping mechanism in the maintenance of a positive self-concept and a non-self-critical approach. Furthermore, in light of self-narrative models of personality (e.g., McAdams, 1993; Bruhn, 1990; 1992; 1995; Singer & Salovey, 1993), dismissive individuals' easy access to positive memories, particularly personally-relevant memories of independence, can be viewed as maintaining their positive self-concept. Overall, the findings of the present thesis support the proposition that Attachment Theory provides a framework for understanding both the personality and cognitive processing styles associated with depression. Indeed, the findings suggest that the fearful and preoccupied adult attachment styles can both be regarded as constituting vulnerability factors for depression. As expected, the fearful attachment style was highlighted as conferring greater depressive vulnerability. Thus, the findings provide further insight into the factors involved in the onset and maintenance of depression and highlight the importance of assessing adult attachment style during therapy for depression. Other clinical implications, as well as directions for future research, are outlined.