Human perceptual learning: the effect of pre-exposure schedule depends on task demands.

The effects of the pre-exposure schedule (concurrent, intermixed, and blocked) to two similar visual stimuli were assessed in three different tasks. Participants were more accurate identifying one of two pre-exposed stimuli as the target by means of same/different judgments after concurrent than intermixed or blocked pre-exposures. Regardless of pre-exposure schedule, participants were accurate in identifying the same target stimulus in a subsequent multiple choice task. However, the other pre-exposed stimulus was incorrectly chosen as the target in a greater proportion after blocked than intermixed or concurrent pre-exposure. Finally, participants who received the blocked schedule showed a greater ability to construct the target in a puzzle test than those who received a concurrent or intermixed schedule. These results suggest that the effect of pre-exposure schedule may depend on task-specific demands. But all these results might be explained by a selective attention mechanism like that proposed by Gibson (1969) to account for perceptual learning.