Testing for controlled variables: a model-based approach to determining the perceptual basis of behavior

Abstract

Perceptual control theory (PCT) views behavior as being organized around the control of perceptual variables. Thus, from a PCT perspective, understanding behavior is largely a matter of determining the perceptions that organisms control-the perceptions that are the basis of the observed behavior. This task is complicated by the fact that very often the perceptions that seem to be the obvious basis of some behavior are not. This problem is illustrated using a simple pursuit-tracking task in which the goal was to keep a cursor vertically aligned with a target set at various horizontal distances from the cursor. The "obvious" perceptual basis of the behavior in this task is the vertical distance between cursor and target. But a control model suggests that a better description of the perceptual basis of the behavior is the angle between cursor and target. The experiment shows how a control model can be used to do the test for the controlled variable, a control-theory-based approach to distinguishing the actual from the apparent perceptual basis of any behavior.