Response strength as a function of stimulus novelty: Concept formation in the pigeon.

Abstract

A previous study showed that a distinctive conditioned stimulus (CS) presented on the first trial of acquisition generated more responses after extinction than another CS presented regularly during acquisition. In this study all subjects were presented with a different distinctive CS on each first trial of 5 acquisition sessions. Subjects were then broken down into 7 groups which were tested, following extinction, for residual response strength of ( a) Day 1 novel CS; (b) Day 2 novel CS; (c) Day 3 novel CS; (d) Day 4 novel CS; (e) Day 5 novel CS; (/) the regular CS; and (g) a novel CS never presented before. Contrary to the previous study where a novel CS inhibited responding, Group 7 exhibited the most responding in this study. The results support a concept formation interpretation of the previous study and are difficult to explain from a "stimulus trace" or "identical elements" position.In a recent study (Burstein & Moeser, 1971), it was found that a distinctive stimulus regularly associated with the first trial of a series of acquisition trials generated more response strength when presented on a test trial following extinction than did another stimulus presented more frequently during the acquisition phase of the experiment. Specifically, for each of 5 daily acquisition sessions of 30 trials, different groups of pigeons were presented with a novel stimulus (a red conditioned stimulus, CS) on either Training Trial 1 or 20 or not at all. For all other acquisition trials, the CS was green. All responses on all trials were reinforced with 3-sec. access to a tray of Purina pigeon pellets. On Day 6, all subjects were extinguished on the "regular" green CS. Subjects were then placed in their home cages for 20 min. after which they were returned to the operant chamber and tested for residual response strength to either the distinctive red CS or the regular green CS.No significant differences were obtained in total test-trial responses of groups tested

Spontaneous recovery evoked by a nonreinforced stimulus

“…An interpretation that further develops Skinner's and Estes' explanations is supported by a series of more recent studies (Burstein & Mackenzie, 1974;Burstein & Moeser, 1971;McMillan & Burstein, 1976;Welker & McAuley, 1978). This interpretation states that, in general, any operation that emphasizes the similarity between the onset of the acquisition session and the spontaneous recovery test trial, while simultaneously stressing the dissimilarity between the extinction sessions and this test trial, will result in the characteristic increase in the response strength associated with spontaneous reThis research was supported in part by National Science Foundation Grant BNS 79-08839and was based on a BA honors thesis by the first author conducted under the direction of the second, Requests for reprints should be addressed to Mark D. Holder, Department of Psychology, University of California, Berkeley.…”

“…The measure of spontaneous recovery was the response frequency on the first test trial (the same measure employed by Burstein & Mackenzie, 1974;Burstein & Moeser, 1971;and McMillan & Burstein, 1976). An analysis of variance based on this measure was significant [F(8,S4) = 6.34, P < .001].…”

“…(3) They differed as to whether the test trial consisted of the distinctive red stimulus, the standard green stimulus, or a completely novel purple stimulus (D = distinctive; S =standard; N =novel). Presentation of the novel stimulus served as a control to determine whether any increase in responding to the distinctive stimulus, as compared with the standard stimulus, was a result of the position and discriminatory properties of the distinctive stimulus or whether any novel stimulus would increase responding, as in the Burstein and Mackenzie (1974)study. In addition to these eight groups, Group I, a control group, was presented with the standard green stimulus on all acquisition, extinction, and test trials.…”

“…Third, it was designed to further explore the development of the "concept of novelty." Although the Burstein and Mackenzie (1974) study employed five presentations of a novel stimulus in order to establish increased responding to a novel stimulus, there was some evidence that one presentation might be sufficient. The present study, therefore, determined whether a novel stimulus would evoke increased responding during the test trial after a different novel stimulus, associated with the onset of the acquisition session, was presented only once.…”

“…For example, Burstein and Mackenzie (1974) began each of five acquisition sessions with different colored keylights, By associating a novel color with the onset of each acquisition session, but not with extinction, a sixth novel stimulus presented on the test trial evoked an increase in responding. This increase in responding was not found in a control group that was presented with only one novel stimulus on the test trial.…”