It has been observed that in maze learning, a blind alley near a goal tends to be eliminated more rapidly than one more remote from the goal. The modeling of this behavior assumes that an event occurring near the goal is reinforced more strongly than one farther away from the goal, and additionally that the preference for one path over another is a function of the ratio of the path lengths. The purpose of this paper is to show that this maze-learning behavior is analogous to optimization of a path-length problem using a technique from Operations Research called Dynamic Programming An example of Dynamic Programming is shown to illustrate the analogy.
Get full access to this article
View all access options for this article.
References
1.
HilgardE. R.Introduction to psychology. New York: Harcourt, Brace, & World, 1962.
2.
NemhauserG. L.Introduction to dynamic programming. New York: Wiley, 1966.
