Kris Giesen

Date of Award


Document Type

Thesis - ECU Access Only


Edith Cowan University

Degree Name

Doctor of Philosophy


School of Psychology and Social Science


Faculty of Computing, Health and Science

First Supervisor

Craig Speelman


This study examined the cognitive operations and responses implicated in depression vulnerability to assess whether a depressive episode can be predicted. It involved 332 never depressed (ND), formerly depressed (FD)> and currently depressed (CD) women undergoing a major life transition (motherhood or first year university), with 267 involved in the follow-up phase 3 months later. Consistent with the cognitive reactivity paradigm and mood-congruent theories, participants were exposed to mood priming procedures (sad film clips) before completing 7 computer-administered tasks. These tasks measured schematic content and organisation in the se)f-system, explicit and implicit memory, and interpretation; each by assessing positively and negatively valenced responses. Because cognitive and behavioural skills are considered to regulate the affective response, Nolen-Hoeksema,s (l 991) Response Style Questionnaire (RSQ) and Rosenbaum's (1980) Self-Control Schedule (SCS) was also administered.

Performance was analysed with two key objectives. The first assessed how depression resilient (i.e . ., ND) and depression vulnerable (i.e . ., FD) people respond to a negative mood relative to a neutral mood, and in comparison to depressed people. The ANOVAs and a priori comparisons showed that FD only evidenced depressogenic-linked biases on some of the indices, while at other times they responded more similar to ND. Further, ND did not always respond in a positive, adaptive manner, but rather evidenced mood-congruent responses typically associated with depression vulnerability.

The second objective, and the main emphasis of the study, was to determine whether a depressive episode is predictable. Contrary to expectations, only 11 non-depressed participants at Time 1 (i.e., 2 ND and 8 FD) had developed depression at follow-up., limiting the predictive capacity of the results. The descriptive results, however, evidenced a lack of uniformity in the pattern ofresponses obtained from the various indices. ND and FD who became depressed at folJow-up evidenced disparate performance patterns, supporting both the cognitive scarring hypothesis and cognitive predisposition debate. Regression analyses found mood to be predicted by a negative self-schemata and rumination in depression vulnerable people (i.e., FD and CD combined), whereas cognitive and behavioural skills predicts that in depression resilient people. Hence, the cognitive indices were not substantial contributors to the prediction models.

The results that emerged from this study suggest that depression vulnerability might be associated with a dysfunctional regulatory mechanism involved in mediating the affective response. The positive, adaptive responses observed in ND suggests they were able to compensate for a dysphoric mood and restore affective equilibrium, which was less evident in FD and almost absent in CD. This was also suppm1ed by responses on the RSQ and SCS following a similar pattern. However, the lack of uniform responses on the indices, even within the respective groups, challenges the possibility of ever being able to reliably predict depression by examining discrete cognitive responses.

Implications for theories and clinical practice are discussed, along with an emphasis on involving factors constituting depression resilience. A new perspective of conceiving depression vulnerability is also proposed. This perspective is in the context of complex systems and dynamic equilibriums, and explains why predicting depression might remain a problem.

LCSH Subject Headings

Cognitive therapy.

Emotions and cognition.

Depression, Mental -- Risk factors.