Serial discrimination reversal learning in the pigeon: Potentiation of "pattern" by a conditional cue.

Abstract

Six experiments were performed in which pigeons were trained daily on a conditional serial discrimination reversal problem. In the standard training procedure, on odd-numbered days a vertical line was superimposed on a successive wavelength discrimination task. On even-numbered days a horizontal line was superimposed on the reverse wavelength discrimination task. Experiments 1, 2, and 3 assessed the degree of control acquired by the compound conditional cues (lines plus "days") and by the component elements (lines or "days"). Only those subjects that were trained with lines relevant mastered the reversal task. In testing, when the conditional line stimuli were present, they exerted strong control, and no days effect was noted. However, when tested in the absence of lines, those subjects that had had training with the compound conditional cue demonstrated reversal performance controlled by day, i.e., the pattern of alternating reversals had been acquired. Experiments 4, 5, and 6 tested alternate interpretations of the control exerted by the day cue. The results imply that explanations based on strict sequence learning, internal clocks, or interelement associations are untenable. Rather, it seems that the "potentiation" effect noted is due to separate storage of the wavelength discrimination problems and different retrieval processes elicited by presence versus absence of the line stimuli.

Factors Affecting Conditional Discrimination Learning by Pigeons

“…On the day following the generalization test, the birds were retrained under Problem 1/Context 1 conditions, receiving the first of 20 sessions-in which the two problems, along with their paired contexts, were alternated on a daily basis. This procedure was called conditional serial discrimination reversal (CSDR) training by Thomas and McKelvie (1982) who used it to establish conditional control over a wavelength discrimination and its reversal by vertical and horizontal lines on the response key. On the day after the 20th session of CSDR training, the birds received a second wavelength generalization test in extinction administered under the same conditions as the first.…”

“…Several points should be made about this result. In the Thomas and McKelvie (1982) CSDR study conditional control was revealed in performance during training; however, the conditional cues were lines projected on the response key, and 40 sessions of CSDR training, rather than 20, were administered. Perhaps additional training would have revealed conditional control in training performance in the present experimental situation as well.…”

“…This procedure was called conditional serial discrimination reversal (CSDR) training by Thomas and McKelvie (1982) who used it to establish conditional control over a wavelength discrimination and its reversal by vertical and horizontal lines on the response key. On the day after the 20th session of CSDR training, the birds received a second wavelength generalization test in extinction administered under the same conditions as the first.…”

Conditional discrimination learning by pigeons: The role of simultaneous versus successive stimulus presentations.

“…Thomas and McKelvie (1982) obtained conditional discrimination learning with key color and line angle combinations, but they used extensive training (i.e., 40 sessions) and a daily reversal rather than a single reversal procedure. If a means were used to prevent the configuring of these cues, brief training and a single reversal procedure might also be effective.…”

Conditional discrimination learning by pigeons: The role of training paradigms

“…Boneau & Honig, 1964;Richards , 1979). Intermediate between these extremes is conditional serial discrimination reversal learning (CSDR ) as used by Thomas and McKelvie (1982). In their study, a wavelength discrimiRequests for reprintsshouldbe addressed to David R. Thomas, Department of Psychology, Campus Box 345, University of Colorado.…”