The effects of trial and cycle durations on automaintenance in the pigeon

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Lee Canipe Groves (Creator)
The University of North Carolina at Greensboro (UNCG )
Web Site:
Aaron J. Brownstein

Abstract: Thirty-six naive pigeons were assigned to groups which differed with respect to trial duration (6, 12, 24, 30, 48, or 96 sec), cycle duration (30, 60, or 120 sec), and, consequently, with respect to the ratio of trial to cycle (.20, .40, .80, or 1.00). All groups were magazine trained, then exposed to a negative-contingency autoshaping procedure modeled after the Williams and Williams (1969) design, except that the trial and cycle durations were specific to each group. Specifically, each subject was placed in a standard operant conditioning chamber illuminated only with an overhead houselight. Periodically (at intervals determined by the cycle duration for each group), a response key was transilluminated for a time specified by the trial duration for the respective groups. If no response occurred to the lighted key, a filled and illuminated grain hopper was presented at the termination of the keylight. If the bird pecked at the key, the key was darkened, the remainder of the trial period was Indicated by overhead green illumination, and no grain was presented. On several dependent measures (number of keylightgrain pairings prior to the first response, number of cumulated sec of trial time prior to the first response, percentage of trials containing a response over the first ten days, maximum responding reached over the first ten days, and number of sessions on which responding was maintained at equal to or greater than 10? of the trials), it was found that keypecking was more likely to be acquired, was acquired more rapidly, and was maintained longer, the smaller the ratio of trial to cycle. The absolute trial and cycle values were seen to exert some influence, however, in that any given ratio value was more effective by these criteria the smaller the absolute values of the trial and cycle. These group data were confirmed by individual manipulations in which each subject was exposed to a series of trial to cycle ratio values within the context of a constant cycle duration. Keypecking was not successfully educed in subjects which were begun on high ratios (which were ineffective in educing responding), but responding in birds which had been initially exposed to low ratios was successfully manipulated up and down over five to six reversals merely by changing the ratio of trial to cycle. These results were repeated with nine pigeons exposed to a fixed-trials autoshaping procedure in which there was no contingency between keypecking and grain presentation. Group and Individual data supported the effect of ratio of trial to cycle as a major determinant of acquisition and maintenance of pecking, with additional influence again being seen by the specific value of the trial and cycle duration. These results were discussed within the context of other autoshaplng data, as well as within the context of current reformulations of the concept of "reinforcement," in which stimuli are viewed as playing both "reinforcing" and "eliciting" roles within any given behavioral context.

Additional Information

Language: English
Date: 1976
Autoshaping (Psychology)
Operant conditioning

Email this document to