There is little consensus in the social-cognitive literature concerning the way in which prejudice and stereotyping are related, though a number of explanatory models have been proposed. The present research program empirically examines one recent model; Lepore and Brown's Associative Strength Model (ASM: 1997; 1999; 2002). The main premise of the ASM is that differential endorsement of stereotypic content leads to individual variation in the content that is automatically activated upon categorisation. Specifically, it predicts that high-prejudice people automatically activate negative stereotypic traits, and low-prejudice people automatically activate positive stereotypic traits. The current research used a primed lexical decision task to examine prejudicelinked differences in automatic stereotype activation. In addition, an impression formation task based on that of Lepore and Brown was included to measure stereotype application. Experiments 1A and 1B attempted to evaluate the predictions of the ASM using the category and stereotype of Asians. However, neither experiment was able to demonstrate a priming effect, prejudice-linked or otherwise, using this social category. Experiments 2 and 3, in contrast, successfully induced stereotype activation using the category of gay men. Furthermore, results were consistent with the predictions of the ASM. After priming with the category of gay men, high-prejudice participants exhibited greater activation of negative stereotypic traits and low-prejudice participants exhibited greater activation of positive stereotypic traits. However, parallel patterns of stereotype application were not found in the impression formation task, with participants forming positive impressions, regardless of prejudice. Experiment 4 used an honesty manipulation to investigate the possibility that self-presentational concerns were responsible for the discrepancies between stereotype activation and application. Consistent with this argument, Experiment 4 found prejudice-linked patterns of stereotype application that mirrored the patterns of stereotype activation when self-presentation concerns were reduced. When instructed to be honest, high-prejudice participants in the gay prime condition formed negative impressions and low-prejudice participants in the gay prime condition formed positive impressions. The current program of research provides the first direct empirical support for the predictions of the Associative Strength Model concerning stereotype activation. In addition, new questions have been raised that future research should seek to explore.