The effectiveness of eidetic imagery and induced anxiety instructions in producing quantitative and qualitative physiological changes was compared in a 2-factor design with repeated measures on phases of treatment; Ss were 15 medical school students. Both induced anxiety and eidetic imagery produced significant changes in heart rate, GSR, and respiration. As predicted, eidetic imagery produced significant eye movement changes, but induced anxiety did not. The hypothesis that eidetic imagery produces discrete arousal while induced anxiety produces diffuse arousal was confirmed for heart rate and GSR.