An experiment is reported which tests Fazey & Hardy's (1988) catastrophe model of anxiety and performance. Eight experienced basketball players were required to perform a set shooting task, under conditions of high and low cognitive anxiety. On each of these occasions, physiological arousal was manipulated by means of physical work in such a way that subjects were tested with physiological arousal increasing and decreasing. Curve-fitting procedures followed by non-parametric tests of significance confirmed (p < ·002) Fazey & Hardy's hysteresis hypothesis: namely, that the polynomial curves for the increasing vs. decreasing arousal conditions would be horizontally displaced relative to each other in the high cognitive anxiety condition, but superimposed on top of one another in the low cognitive anxiety condition. Other non-parametric procedures showed that subjects' maximum performances were higher, their minimum performances lower, and their critical decrements in performance greater in the high cognitive anxiety condition than in the low cognitive anxiety condition. These results were taken as strong support for Fazey & Hardy's catastrophe model of anxiety and performance. The implications of the model for current theorizing on the anxiety-performance relationship are also discussed.