One hundred thirty-two male and female college undergraduates were randomly assigned to three different arousal conditions-sex, anxiety, laughter-and a nonaroused control, as a test of the hypothesis that aggressive and sexual imagery would appear only under conditions of sexual arousal. The subjects wrote four TAT stories, two to male-dominant pictures and two to female-dominant As predicted, the sexually aroused subjects wrote stories higher in sexual and aggressive imagery than did those in the other two arousal conditions, and the scores of those in the anxiety and laughter conditions were not significantly different from the controls'.
The results are discussed in terms of a connection between sex and aggression. Hull's theory of drive level and Schachter and Singer's nonspecific arousal are rejected as explanations for the results. Support is also given to earlier findings in the area.