Depressed unipolar male patients (n = 30) were more likely to attribute bad outcomes to internal, stable, and global causes than were nondepressed schizophrenics (n = 15) and nondepressed medical patients (n = 61). Also, the depressed patients were more evenhanded in their attributions for good versus bad events than the other patients. These results support the existence, in clinical depression, of .the depressive attributional style postulated by the reformulated learned helplessness model and indicate that it is not a general characteristic of psychopathology.According to the reformulated learned helplessness model, depressive symptoms are associated with a particular attributional style-one in which bad outcomes are attributed to internal, stable, and global causes (Abramson, Seligman, & Teasdale, 1978). Recent research has investigated this hypothesis in samples of college students. Some studies have shown the hypothesized attributional style to be associated with mild depressive symptoms, as is (to a lesser degree) the opposite style for attributing good events (e.g., Seligman, Abramson, Semmel, & von Baeyer, 1979), but other studies have yielded contrary evidence (Blaney, Behar, & Head, 1980;Hammen & Cochran, 1981;Harvey, 1981).Any attempt to generalize from mild depression to clinical depression is hazardous because the two phenomena may be different in kind and not just in degree (e.g., Arieti

Forty-eight inpatients hospitalized for 1, 3, or 9 weeks and 24 outpatients were tested on cognitive tasks: poor performance and depressive symptoms increased with length of hospitalization, even as illness resolved. Further, increased hospitalization made patients more susceptible to the debilitating effect of uncontrollable events. Taken together, these results imply that the passive, compliant, and inanimate behavior of the "good patient" may be the result of learned helplessness engendered by hospitalization.

According to the attributional reformulation of learned helplessness, depressive symptoms are associated with an attributional style that points to internal, stable, and global causes for bad events involving the self. Recent research has yielded contradictory support for this proposal. Thus, 61 published tests of the attributional reformulation were analyzed to determine factors that might distinguish those that corroborated the reformulation's predictions from those that did not. To the degree that a study used a large sample, a large number of events about which attributions were made, and hypothetical events, it tended to support the reformulation with respect to stable and global attributions. However, these characteristics were highly intercorrelated across studies, making it impossible to isolate their independent effects. None of the factors examined consistently distinguished supporting from nonsupporting studies with respect to internal attributions.

This study was undertaken to extend the learned helplessness phenomenon to a clinical population and to test the competing hypotheses of Seligman and Lewinsohn. Ninety-six male hospitalized psychiatric and medical patients were divided into three levels of depression according to their scores on the Short Form of the Beck Depression Inventory. Subjects were randomly assigned to one of four experimental conditions: (a) One group was treated with an 80-dB tone, which could be terminated by making an active response; (b) a second group was treated with the tone with a passive escape contingency; (c) a third group was treated with an inescapable tone; and (d) a no-noise group served as a control. After treatment, subjects were tested on an anagram-solving task. Inescapable noise produced as much deficit in the low-depressed subjects as was present in the depressed no-noise control subjects. Passive escape subjects did as well as active escape subjects. Results replicate the learned helplessness phenomenon in a group of clinical depressives and support Seligman's model of depression.The learned helplessness phenomenon has been proposed as a model for clinical depression (Miller & Seligman,

Therapeutic implications of the learned helplessness model of depression were tested in a clinical population. In pretreatment, two groups of nondepressed medical patients waited, two groups of nondepressed medical patients received helplessness training, and two groups of psychiatric patients (diagnosed as Primary Affective Disorder) waited. In treatment, subjects received either Velten's mood-elation procedure as "therapy" or Velten's mood-neutral procedure as placebo. Performance on a cognitive task and on a mood task was assessed. Three separate administrations of the Depression Adjective Check List indicated that helplessness training induced depressive affect, and the mood elation procedure decreased depressive affect for both helpless and depressed patients. The mood neutral procedure and the waiting periods were associated with no affective changes. On the cognitive (anagrams) task, performance deficits were associated with helplessness and depression but were reversed by mood elation. Results are interpreted as consistent with the learned helplessness model of depression.