Computational Models for the Analysis of Attention

Abstract

The resurgence of interest in the cognitive sciences since the early 1970s has been accompanied by a dramatic acceleration in the development of theoretical models. Cognitive scientists develop models in order to characterize the formal properties of the processes that they are studying. Efforts to model cognition led to a recent focus on “neural networks,” computer-based computational systems that simulate the characteristics of actual neural brain processes. Neural networks are designed by means of computational modeling, which is a direct result of advances in the fields of mathematical psychology and computer science. Computational models are created with the hope that they will provide a formal mathematical basis for understanding cognitive and behavioral processes. These models seek to specify the parameters underlying information processing and therefore are of considerable interest to investigators of attentional phenomena. Although at times these models seem to do little more than provide a mathematical description of processes that are typically characterized less formally, they may be useful for identifying the constraints and parameters underlying attentional operations.