This study is one in a series of experiments aimed at an analysis of the effects of stress upon performance. Previous studies have met with little success in analyzing individual differences in the effects of stress (2). In spite of the negative results obtained thus far, there is reason to believe that differences between individuals constitute the most important variable in the area of stress. We have also pointed out elsewhere (2) that some kind of interaction must exist between the type of stress and these individual differences.Recently, a series of papers dealing with anxiety, learning, and avoidance conditioning suggested a possible lead in the study of individual differences and their interaction with type of stress. In the first of these studies, Taylor (6) selected two groups of Ss on the basis of a personality questionnaire which was presumed to measure "anxiety." She compared the rate of acquisition of a conditioned eyelid response in these two groups which she labeled "high anxiety" and "low anxiety," respectively. The "high anxiety" group conditioned much more rapidly than did the "low anxiety" group. A subsequent study (5) showed the same effect irrespective of the strength of the UCS. These data