It was hypothesized that information about current romantic involvement could make a male target more attractive to females. A 2 (Gender) X 4 (Romantic Involvement: Current Romance, Past Romance, Friend, and Control) factorial design was created to test competing predictions from scarcity/reactance, impression-formation, and sociobiological perspectives. A total of 235 male and female subjects saw a photograph of an opposite gender target person, then read a brief description about the target that contained one of the three involvement manipulations, or contained no involvement manipulation (Control). Subjects then rated the target along a 14 item scale to measure romantic attraction. ANOVA and MANOVA results revealed main effects of Gender and Romantic Involvement. Simple effects for Romantic Involvement were found for female, but not male subjects. A priori contrasts testing the predictions from the competing theoretical perspectives provided support only for the socio-biological prediction for female subjects.