Toward a Model of Attention and the Development of Automatic Processing.
ILLINOIS UNIV CHAMPAIGN HUMAN ATTENTION RESEARCH LAB
Pagination or Media Count:
A model for the development of automatic processing is briefly described. The model is a quasi-neural model in which information processing is done through the transmission of vectors between visual, lexical, semantic, and motor processing units. Controlled processing involves gating of the output power of vectors to perform matches and to release response vectors. As subjects practice consistent tasks, associative learning enables an input vector to evoke an output vector and priority learning determines the power with which a vector is transmitted. Automatic processing involves a cascade of vector transmissions in which the output power of each transmission is determined by priority learning. The transition from controlled to automatic processing takes place in four phases. Empirical illustrations of the transition are described. Author.