Memory Recruitment: A Backwards Idea about Masked Priming
Glen Bodner
University of Calgary

Accounts of priming typically assume that primes preactivate existing representations, thus proactively altering target processing. In contrast, Mike Masson and I have put forward a memory-recruitment account of priming (e.g., Bodner & Masson, 1997; Masson & Bodner, 2003) that suggests the processing applied to primes, even masked primes, is encoded into a memory resource, which in turn is retrospectively recruited to contribute to target processing if the context and task conditions support this recruitment. My talk will emphasize two lines of research that have attempted to contrast these prospective and retrospective accounts. The first line examines whether nonwords, which do not have existing representations, show masked priming. The second line examines whether masked priming is sensitive to context and task factors in ways that contradict prospective accounts. After summarizing the successes and failures of our backwards account, I will provide suggestions for moving accounts of priming forward.