Depressive automatic processes as vulnerability markers in depression

Previous research has found little support for cognitive vulnerability factors in depression. I argue that this lack of support is due to the use of tasks that tap into effortful processes, as seen in previous research. I propose that support for cognitive vulnerability factors would be found by using tasks which tap into automatic processes. Depressive automatic processes were assessed by three tasks: dichotic listening, probe detection, and implicit memory tasks. For the dichotic listening task, subjects shadowed brief neutral passages while depression- and positive-content words were played in the unattended channel. Concurrently, subjects were required to detect the word "press" presented intermittently on a computer screen. Detection latencies were recorded. For the probe detection task, pairs of words were briefly presented on a computer screen, one appearing in the upper half of the screen, and one appearing in the bottom half. Three types of word pairs were used: neutral-neutral, depression-neutral, and positive-neutral. Subjects were asked to read the top word aloud. They were also required to detect the presence of a "+", which sometimes appeared in either location of the words. Detection latencies were recorded. For the implicit memory task, half of the subjects were presented with a list of words and were required to rate each word on how much they liked each word. Four types of words were used: depression-related, happiness-related, types of flowers, and types of diseases. These subjects were then asked to generate eight exemplars for each word type. The other half of the subjects were simply asked to generate eight exemplars for the word types. Depressive effortful processes were defined as self-report of cognitions. This was assessed by three self-report questionnaires: Dysfunctional Attitude Scale, Automatic Thoughts Questionnaire, and the Hopelessness Scale. Three groups of subjects were used: currently depressed patients (n=20), remitted depressed individuals (n=20), and nondepressed individuals (n=20). The currently depressed group consisted of 13 women and 7 men, the remitted depressed group consisted of 16 women and 4 men, whereas the nondepressed group consisted of 13 women and 7 men. Their ages ranged from 23 to 65 years, with an overall average of 39.9 (SD = 11.28) years. Subjects were tested individually on each of the tasks. Three months after testing, they were asked to complete the Beck Depression Inventory. Four main hypotheses were examined: (a) currently depressed individuals would show a bias for depression-related stimuli on the automatic tasks; (b) remitted depressed individuals' pattern of performance on the automatic tasks would resemble that of the currently depressed patients; (c) remitted depressed individuals' pattern of performance on the effortful tasks would resemble that of the nondepressed individuals; and (d) measures of depressive automatic processes would be predictive of follow-up depressive symptoms. Analyses of variance and regression analyses were used. The first hypothesis was not supported. The second hypothesis was only supported for the dichotic listening task. The third hypothesis was supported, whereas the fourth hypothesis was not supported. Implications of the results to schema theory, vulnerability methodology, and clinical assessment procedures were dlscussed.Arts, Faculty ofPsychology, Department ofGraduat