On Cognitive Maps, Vicarious Trial-and-Error, and Impulsivity
Abram Amsel Department of Psychology and Institute for Neuroscience University of Texas at Austin Austin, TX 78712 Telephone: 512-471-3827/Fax: 512-471-6175 E-mail: email@example.com
The most prominent hypothesis of hippocampal function likens the hippocampus to a cognitivemap, a term used by a famous learning theorist, E. C. Tolman, to explain maze learning. The usual current application of this concept of cognitive map, as it applies to the hippocampus, is to what is called spatial learning, mainly in the radial-arm maze of Olton and the Morris water maze. In a recent Hippocampus Forum in the journal "Hippocmpus", evidence for the cognitive map hypothesis was reviewed in a lead article by L. Nadel, followed by a series of commentaries by leading investigators of hippocampal function. In a later issue of Hippocampus, I offered an alternative not represented in the forum--that the function of the hippocampus in spatial learning is not as a cognitive map, but that it subserves another function also proposed by Tolman in his work on simple discrimination learning, vicarioustrial-and-error (VTE), based on incipient, conflicting dispositions to approach and avoid.
Experimental data are presented to show that reduced hippocampal function affects learning--single alternation and the partial reinforcement extinction effect in a straight runway--that is in no way spatial. They support the hypothesis that intact VTE may itself be a function of the hippocampus, and that there may be connections among deficits in VTE, impulsivity, and attentional disorders. Some recent experiments on this latter conjecture are also presented. These experiments, also in infant and preweanling rats, involve reducing hippocampal function pharmacologically, following pre-and postnatal exposure to alcohol, and via x-irradiation, and restoring some of this loss of function with d-amphetamine.
The Hullians of my generation, and, particularly, my mentor, Kenneth W. Spence, would probably not be amused by my emphasis, in this presentation, on Edward Tolman and some of his followers, particularly in the opening part of my remarks; but, except for the S's and R's and the little rg's in a later allusion to my own theorizing, the work of Tolman, a famous learning theorist of the 1930's and 1940's, will be my main point of departure.
A major focus of this presentation is on the hippocampus, and the most prominent hypothesis of hippocampal function, originated in a book by O'Keefe and Nadel ( 1978), likens the hippocampus to a "cognitive map," a term coined by E. C. Tolman ( 1948) to explain, among other things, complex maze learning. The more recent application of this concept of cognitive map, particularly as it applies to functions of the hippocampus, is to what is called "spatial learning," mainly in the radial-arm maze of Olton and the Morris water maze.
In a Hippocampus Forum "Hippocampus and Space Revisited," in the journal Hippocampus, in 1991, evidence for the cognitive map hypothesis was reviewed in a lead article by L. Nadel, followed by a series of commentaries by leading investigators of hippocampal function. In one of the next issues of Hippocampus, I offered, as an epilogue to this love feast-like commentary, an hypothesis that had not been represented in the Forum ( Amsel, 1993).