A vulnerability model of depression was tested by hypothesizing that depressogenic self-schemas that interact with schema-congruent negative life events will be associated with depression. Ninety-three college students were followed prospectively for four monthly assessments of both interview-and questionnaire-measured stressful life events, and clinical interview-and questionnaire-measured depression. An information-processing schema model of vulnerability was used to define subgroups of 46 dependent and 32 self-critical schematic individuals at the beginning of the study; the schema groups were based on clinical subtypes discussed both by Beck and by psychodynamically oriented theorists. As predicted, the dependent subgroup showed significantly stronger associations between depression and schema-relevant interpersonal life events than between depression and schema-irrelevant negative achievement events. The predicted opposite pattern for self-critical schematics was observed, but was less often statistically significant. The results are discussed in the terms of the need for integration of cognitive and life-stress models of depression.Depression research that integrates developments in life-stress approaches and cognitive approaches is virtually nonexistent, despite numerous calls for such efforts (e.g.,

A sample of 93 persons selected for diversity of initial depression levels was studied longitudinally to explore the relation between stressful events and depression, both assessed with interview and selfreport methods. Previous research has often indicated modest associations but has been limited by methodological issues of assessment, sampling, and design, which are addressed in the present work.Regression analyses and inspection of individual patterns supported the hypothesis that initial depression status is a critical factor in depression-event associations, and concurrent high-impact negative events contribute significantly but modestly to outcomes. It appeared that nonsymptomatic persons were relatively resistant to onset even when exposed to high-impact stress events, whereas a subset of initially symptomatic persons continued to have both more depression and more high-impact events over time. It is suggested that future research on event-depression associations should carefully consider these different outcome patterns: symptom resistance and symptom onset in nondepressed persons and symptom remission and symptom maintenance or recurrence in initially depressed persons.Whereas stress-illness research flourishes in general (e.g., Goldberger & Breznitz, 1982), nowhere is such activity more intense than in the study of depression. In the few years since Brown and Harris's major Camberwell study (Brown & Harris, 1978), numerous studies have investigated the stress-depression relationship (e.g.,

The role of depressive self-schemas in vulnerability to depression was explored in a longitudinal design. Five groups of subjects hypothesized to be at differential risk for depression according to a schema model were identified: depressed schematic, depressed nonschematic, nondepressed schematic, nondepressed nonschematic, and a psychopathology control. They were followed regularly for 4 months with selfreport and clinical interview measures of depression. There was no evidence of risk for depression associated with schema status apart from initial mood and no interaction of life stress events and schemas. In a second experiment with the same subjects, it was shown that depressive self-schemas do not exert an ongoing, active influence on everyday information processing; instead current mood affected information processing. Remitted depressed persons resembled nondepressed rather than depressed ones. The results support Kuiper and colleagues' distinction between concomitant and vulnerability schemas, and help to clarify differences between cognitions that are symptoms or correlates of depression and those that may play a causal role under certain conditions.Beck's seminal contributions to the study of depression have included an emphasis on the characteristic negativism of depressive thinking and a major new system of treatment for a difficult and heretofore discouraging population (Beck, 1967;Beck, Rush, Shaw, & Emery, 1979). What is less well-established at this date, however, is the etiological significance of depressive thinking. The vast bulk of studies confirming the characteristic negativism of depressed persons has been essentially correlational in form (see Coyne & Gotlib, 1983, for a review), and although descriptively informative, such work fails to illuminate the causal relation between depressive cognition and depressive affect. Although it is intuitively apparent that gloomy and pessimistic thinkingWe gratefully acknowledge the assistance of those who served as interviewers through the course of the follow-up:

Stressful life events as well as individual appraisals have been hypothesized as important antecedents of depression. However, the examination of characteristics of life events has been a relatively neglected topic. The present study investigated depression and life-event characteristics by applying an event classification scheme proposed by Fairbank and Hough (1979). Cognitive appraisal properties of event types were also examined to shed light on event qualities. Four hundred undergraduates completed the Beck Depression Inventory, a Life Events Inventory, and an Attribution Questionnaire. The results indicated that the event classification scheme was successful in separating types of events by their degree of association with depression. Also, event, types were associated with shared cognitive appraisals. Thus, event types differed in depression associated with them as well as in cognitions associated with them. Theoretical implications of the results are discussed, with a particular focus on the implications for current psychological theories of depression.

Dysfunctional cognitions have been hypothesized to be important determinants of unassertive responding, although treatment strategies aimed directly at modifying cognitions have generally proved no more effective than skills training interventions alone. However, the possibility that individual differences in dysfunctional cognitions may interact with type of treatment has not been previously explored. Nonstudent adults with assertion problems and with high or low levels of dysfunctional attitudes were assigned to cognitivebehavioral, skills training, or waiting list control groups. Both self-report and behavioral assessments of behavior change following treatment failed to show interaction of treatment type and level of dysfunctional attitudes. Skills training and cognitive-behavioral treatments were equally effective and significantly more effective for persons with low levels of dysfunctional attitudes. The results are discussed in terms of matching individuals to treatments and highlighting the particular challenge presented by the unassertive, high-dysfunctional-attitude individuals.