Self-report and processing measures of cognitive vulnerability to depression: Relationship to dysphoric symptoms in a nonclinical sample

Abstract

Some studies have demonstrated higher sensitivity of processing measures compared to self-reports in predicting dysphoric symptoms. Inasmuch as both types of measures predict future depression, it was concluded that they assess different aspects of cognitive vulnerability to depression. The main aim of this study was to determine to what extent self-reports versus processing indicators of cognitive vulnerability to depression were related to dysphoric symptoms. An additional aim was to check the relationship between a processing measure and dysphoric symptoms under two conditions: cognitive load and nonload (retaining a six-digit number in memory). In a sample of 1082 students (671 females) the following instruments were administered: Depression, Anxiaty, and Stress Scale (DASS-21), Serbian scrambled sentences test (SSST) and Dysfunctional Attitudes Scale (DAS). Hierarchical regression analyses showed that both self-report (DAS) and processing measures were significant and independent predictors of dysphoric symptoms, which supports the assumption of two measures assessing different aspects of cognitive vulnerability to depression. Two dimensions of the DAS, Perfectionism and Dependence, made an equal contribution to the prediction (i.e., 4%). SSST proved to be a better predictor of dysphoric symptoms then both dimensions of the DAS under both load conditions (with load: unique variance was 9%, and without load: unique variance was 19%). This finding is consistent with the theoretical assumptions stipulating that processing measures are more reliable indicators of vulnerability as they are not susceptible to intentional distortion. When considering the effect of the cognitive load manipulation, the results showed a stronger relationship between symptoms and depressive content on the SSST obtained without cognitive load, which is contrary to the results of foreign studies and can be explained by the peculiarities of our sample.