Verbal overshadowing of multiple face and car recognition: effects of within‐ versus across‐category verbal descriptions

We present four experiments examining ‘verbal overshadowing’ (the phenomenon that verbally describing a stimulus can interfere with subsequent recognition of that stimulus) using a new multiple stimuli presentation paradigm. Participants are exposed to 12 to-be-remembered stimuli (faces or cars) and then describe (or not, in the control condition) a related stimulus (a 13th face or car). Subsequently, participants have to discriminate the original 12 stimuli from 12 distracters (in a ‘yes/no’ recognition decision). Verbal overshadowing occurs for both accuracy and response times for both face and car recognition, when participants have previously described a face. When participants describe a car prior to recognition, however, verbal overshadowing does not occur. We argue that (1) the paradigm provides a new tool for studying verbal overshadowing and (2) verbal overshadowing is not ‘semantic category-bound’ (i.e. limited to describing stimuli within the same semantic category). We interpret these findings within a ‘transfer-inappropriate retrieval’ framework (Schooler et al., 1997). Copyright © 2002 John Wiley & Sons, Ltd.