Direct Learning 论文

摘要

The theory of direct perception holds that perception is specific to properties of ambient energy arrays. This article explores a similar approach to perceptual and perceptual-motor learning: Change due to learning is portrayed as specific to properties of ambient energy arrays. This type of learning is labeled direct learning. It is argued that a theory of direct learning explains a wide range of phenomena in ecologically relevant and informationally rich situations as well as in simpler experimental situations. Previous experimental results are reanalyzed and reinterpreted from the perspective of direct learning. To achieve this, the notion of information space is introduced. In information spaces, points represent ambient energy variables, paths represent change due to learning, and vector fields represent information that guides learning. These concepts also allow the theory of direct learning to be connected to and to take advantage of the theory of ordinary differential equations.