Do self-presentation concerns moderate the relationship between implicit and explicit homonegativity measures?

This study investigated whether the relation between implicit and explicit homonegativity measures is affected by self-presentation concerns, since previous research in this area has been inconclusive. In Experiment 1, 70 high-school students made evaluative ratings of pictures of homosexual and heterosexual couples. Self-presentation was manipulated by either instructing participants that the study concerned attitudes regarding sexual orientation (socially sensitive) or attitudes regarding age (less sensitive). The age-instruction led to increased homonegativity but not to a stronger correlation with an Implicit Association Test (IAT; Greenwald, McGhee & Schwartz, 1998) with homo/heterosexual couples. Concerns regarding the construct validity of the present implementation of the IAT were alleviated in Experiment 2, where heterosexual (n = 30) but not homosexual (n = 30) participants showed implicit homonegativity. The current findings are problematic for the interpretation of low correspondence between implicit and explicit attitude measures as being primarily an effect of self-presentation concerns.