The authors tested the approach/inhibition theory of power by examining teasing interactions between women and men in conditions in which either one was given elevated power or they were in an equal-power control condition. Consistent with hypotheses, high-power individuals behaved in a disinhibited fashion and were less accurate judges of their partner's emotion, whereas low-power individuals behaved in a more inhibited, indirect fashion and reported more self-conscious/anxiety-related emotion. Additional contrast analyses revealed only modest support for the claim that men would act in powerful fashion in the absence of explicit power differences, and that power-based differences were greatest when the man had power over the woman. Discussion focuses on different perspectives on the interaction between power and gender.