RESPONSE PERSEVERATION IN AUDITORY WORD RECOGNITION

The probability (p) of an S changing his response on any trial in an auditory word-recognition task was estimated by fitting geometric distributions to observed distributions of intervals between successive response changes. Presenting 15 trials of a word sequentially with a 1-db reduction in noise on each trial (Group S) resulted in lower values of p than were obtained when 16 words were presented together in a randomized order at each noise level, so that the words did not occur in any predictable sequence (Group R). When Ss in Group S were divided into ordered subgroups on the basis of their susceptibility to response sets in nonperceptual problem-solving situations, the subgroups were homogeneous and similarly ordered with respect to p, while similar subgroups of Group R did not differ from one another. Auditory recognition points were linearly related to p. (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1969 American Psychological Association.