The present study tested Abramson, Seligman, and Teasdale's (1978) reformulation of the learned helplessness hypothesis. Specifically, the study employed a laboratory paradigm to investigate: (a) whether attributions about uncontrollable events mediate subsequent deficits, particularly self-esteem loss; and (b) if lower socio-economic class individuals are more susceptible to helplessness following uncontrollability than are middle class individuals, All subjects were Black female college students, 25 from the lower class and 25 from the middle class. These subjects were randomly assigned to one of five experimental conditions: internal attribution provided for failure to a concept-identification task, external attribution provided for failure, no attribution provided for failure, no attribution provided for success, and no pretreatment task. All subjects were then tested for performance deficits on an anagram-solving task. Additional dependent variables included mood change, self-esteem change, and persistence at Rubick's Cube. A two-way analysis of variance using the factors social class and experimental condition revealed few differences across groups on any of the measures. Possible reasons for the failure to obtain differences were discussed.