RETARDATION IN HUMAN DISCRIMINATION-LEARNING AS A CONSEQUENCE OF PREEXPOSURE - LATENT INHIBITION OR NEGATIVE PRIMING

A series of experiments with human subjects, using black-and-white che querboard patterns, demonstrated that non-reinforced pre-exposure coul d impair performance in a subsequent learning task. Subjects were invi ted to take part in a scenario similar to that of a computer simulated card game. Their task was to turn over a series of cards by mouse-cli cking on a pack of cards lying face-down, and then to classify these c ards into one of two categories. In a subsequent task, subjects were a sked to discriminate between pairs of chequerboards, some of which had previously appeared in the initial categorization phase: either direc tly (''fronts'') or incidentally (''backs'') involved in categorizatio n. In Experiment 1, for those stimuli used as the backs of the cards ( that is, those visible on top of the pack of cards), there was a signi ficant impairment in performance relative to non-pre-exposed control s timuli. Although the impairment appeared to be specific to the stimuli pre-exposed, when the pre-exposed ''backs'' were minimally distorted in the discrimination task of Experiment 2, performance was still sign ificantly impaired relative to non-pre-exposed control stimuli. The re sults of Experiment 2 do not support the interpretation that retardati on in learning following masked pre-exposure in human experiments is c omparable to latent inhibition following simple preexposure in other a nimals. Whilst the impairment in performance appears to be similar to that of latent inhibition, the results may instead, be better understo od in terms of the inhibitory processes involved in negative priming. If this is so, then serious doubt is cast on whether latent inhibition has ever been reliably demonstrated in adult humans.