Possibility of Artifact in Studies of Cooperation

Abstract

Stimrnary.-A comparison of cooperative and noncooperative response rate data for some Ss indicated that reinforcement influenced both rates in the same proportion during acquisition and extinction periods, even though reinforcement was contingent upon the cooperative class of behavior only.An unambiguous demonstration of reinforcement control over cooperative behavior should show that changes in cooperative rate could not be interpreted as artifact, the result of reidorcement control over generalized responding. Several studies of cooperatior: do not report noncooperative response rate (e.g.,Azrin & Lindsley, 1956; Cohen, 1962;Hingtgen, Sanders, & DeMyer, 1963; Weingold & Webster, 1964) which precludes a comparison of the effect of reinforcement on cooperative response rate with noncooperative response rate. Such a comparison would determine whether selective reinforcement control was obtained over cooperative rate or whether noncooperative rate was also influenced by reinforcement, and the relative magnitude of the effect of reinforcement on the two rates.In the srudy reported here both cooperative and individual noncooperative responses were recorded. The study closely resembled the one by Azrin and Lindsley (1956) except that the response was more difficult since the latency (not the duration) of the response was restricted to 0.05 sec. and the age range of Ss was 4 through 9, as compared with 7 through 12 in the Azrin and Lindsley sample. Ss were seated at opposite sides of a table with three holes and a stylus before each child. Ss were given non-specific instructions about the relationship between their behavior and reinforcement. The first six channels of a 20-channel Esterline Angus recorder were used to record individual noncooperative responses from each of the six holes in the table. The seventh channel recorded only cooperative responses, i.e., i t recorded only when both styli were inserted into any one pair of the three aveilable pairs of opposite holes within 0.05 sec. of each other.Seven of 12 pairs of Ss failed to develop cooperation as judged by very low cooperative rates (Coop. Rs/min. < 3 ) after 20 min. of continuous reinforcement and were terminated. Four of the remaining five pairs of Ss failed to cievelop cooperative rates which were clearly independent of noncooperative rates. Characteristic of these four pairs are the data for the pair presented in Fig. 1, 'This paper is based on a thes3 entitled "Analysis of a cooperative response" submitted to the Deparunent of Psychology of the University of Arizona in partial fulfillment of requirements for the M.A. degree. The author wishes to express his gratirude to Dr. Ralph J. Wetzel for his invaluable assistance and advice in carrying out the study. 'Now at Pomona College, Clarernonc, California.

Analysis of Discriminative Control by Social Behavioral Stimuli

“…High and steady response rates by both subjects would have produced a moderate number of reinforcers without regard to the behavior of the partner. Several studies (Brotsky & Thomas, 1967;Schmitt & Marwell, 1968;Vogler, 1968) suggest this was the case for some high-rate subjects.…”

A CLASSIFICATION AND REVIEW OF COOPERATION PROCEDURES1

“…For example, a steady response rate of 120 responses per minute by each subject would result in many reinforced or cooperative-like responses. If the responding were under the control of an individual high-rate contingency, individual responses that did not meet the 0.5-sec criterion for coordinated responding would be expected to increase and to be maintained along with responses meeting the 0.5-sec criterion (Brotsky and Thomas, 1967;Vogler, 1968;Schmitt and Marwell, 1968 It is more difficult to demonstrate that the cooperative behavior of a subject is under the control of the cooperative behavior of his partner in dependent cooperation procedures. It is simply unrealistic to expect that subjects will always respond in strict alternation such that the cooperative response of one subject is always the stimulus for the next cooperative response of his partner.…”

Cooperative responding in rats maintained by fixed‐ and variable‐ratio schedules

“…Although several other studies reported a functional relationship between temporally coordinated responding and reinforcement (Azrin & Lindsley, ; Cohen, ; Lindsley, ; Łopuch & Popik, ; Toledo & Benvenuti, ; Weingold & Webster, ), until recently no study has shown that coordinated responding occurs because of mutual reinforcement rather than as a byproduct of changes in individual response patterns (cf. Hake & Vukelich, ; Schmitt & Marwell, 1968; Tan & Hackenberg, ; Toledo et al, ; Velasco, Bevenuti, Sampaio, & Tomanari, 2017; Vogler, ).…”