Chapter 78 - Mechanisms of Perceptual Learning

Abstract

Perceptual learning—improvement in the performance of a perceptual task as a function of practice or training—is a widely observed phenomenon that may have important practical and theoretical consequences. Perceptual learning may reflect plasticity in different levels of perceptual analysis, including changes in early visual, auditory, or somatosensory cortices, as well as higher-order changes in the weighting of information in task performance. Perceptual learning, as distinct from cognitive learning or strategy selection, often exhibits significant specificity to the trained stimuli or tasks, and is assessed by transfer (or, conversely, generalization) tests. At a behavioral level, the effects of perceptual learning on an observer's performance are characterized by external noise tests within the framework of noisy ideal observer models. In visual perceptual tasks, behavioral analysis, combined with evidence from neuroscience, supports perceptual learning at several levels that has the function of improving two separable mechanisms: tuning of the task relevant perceptual template (external noise exclusion) and enhancing the stimulus (reducing absolute threshold). These two mechanisms of improvement are separable in certain circumstances, but often coexist, albeit with decoupled magnitude. Many improvements because of perceptual learning reflect retuning through reweighting of unchanged early sensory representations.