When less is sometimes more: investigating the interplay between meta-cognition and member-to-group generalisation

Fluency is a meta-cognitive process whereby people apply the ease associated with retrieving or processing stimuli to a subsequent content-relevant judgment (Schwarz 1998; Winkielman & Cacioppo, 2001). In the present work, retrieval fluency and, to a certain extent, processing fluency are applied to the social cognitive process of member-to-group generalisation. During member-to-group generalisation, information about the individual members of a social group affects the judgment of the group as a whole (Paolini, Hewstone, Rubin, & Pay, 2004). In Chapter 1, I trace the development of member-to-group generalisation research and identify the contact hypothesis and the social-cognitive approach as two distinct approaches to generalisation. In Chapter 2, I meta-analytically review social cognitive studies that used an impression formation paradigm to investigate member-to-group generalisation. In Chapter 3, I review retrieval fluency research that had been applied to the investigation of social group judgments and meta-analytically review impression-formation studies that had investigated meta-cognitive processes. In Chapter 4, I introduce a dual process model of the effects of retrieval fluency on member-to-group generalisation and advance a self-generation paradigm to test for the influence of fluency on member-to-group generalisation. In Chapter 5, I present a first test of the new self-generation paradigm and demonstrate that retrieval fluency influences member-to-group generalisation. In Chapter 6, I combine the self-generation paradigm with the impression formation paradigm in a single research design and demonstrate that both approaches to retrieval fluency are appropriate to predict stereotype change. In Chapter 7, I test alternative accounts for the dual process model, and argue that retrieval fluency effects automatically influence subsequent out-group judgments unless participants are engaged in bias correction. In Chapter 8, I present a final test of fluency and demonstrate that both impression formation and self-generation paradigms are influenced by different sources of meta-information and by the typicality of the exemplar. In Chapter 9, I present a summary of the evidence, highlight the main themes of the research, and identify implications for future investigations and social policy interventions.