Effects of within-chain response delay upon post-delay operant performance

Abstract

Operant responding of rats was delayed by withholding the response levers between the two components of a response chain.Latency and rate of responding in the second component were facilitoted by a 10 sec dewy with respect to a 4 sec dewy. When the delay was increased to 50 sec, the facilitation of wtency declined whereas the facilitation of response rate increased.Several theorists {cf. Brown. 1961 ) have viewed delay or interruption of an ongoing response as a source of frustration analogous in its effects to nonreinforcernent. However, it has not been established that the behavioral effects of response delay are. in fact, similar to those of nonreinforcement.It has been consistently demonstrated with rats that frustrative nonreinforcement increases post-nonreinforcement response levels in both runway (cf. Amsel, 1958) and lever pressing contexts (e.g., Carlson, 1968). But several investigators have reported inconsistent effects of introducing an enforced delay into an instrumental response chain. For example, Holder, Marx, Holder, & Collier (1957) have shown that delay of a running response produces an increase in speeds in the portion of a runway following the point of delay. In contrast, Wist (1961) and others have demonstrated that delay of responding produces a decrease in post-delay running speeds. The source of this disagreement is not clear since the procedures used in these studies differed in a number of ways. But it is possible that a lack of control inherent in the runway delay setting itself may have contributed to the differences.For this reason. and in view of earlier success in adapting operant procedures to the study of nonreinforcement, the present experiment investigated delay of responding in an operant response setting. Delay was produced by withholding response levers for a period of time between the first and second components of a two-lever response chain.Method. Eight 9Q..day-illd male Long-Evans hooded rats served as Ss.Eighty-five per cent of ad lib body weights were maintained thwughout the experiment.Two operant conditioning chambers and programming equipment described previously (Cadson, 1968) were used. The distance from the response wall of the chamber to the opposite wall was reduced from 14 in. to 8 in.The S5 were given two 30 min adaptation sessions, three days of magazine training (60 reinforcements per session), and ten days of lever1'ress training using the training procedures descnoed in Carlson (l %8). During the next lO sessions, the following reinforcement schedule was in effect. A trial began witlt the onset of the house light and \eft jewel light and the insertion of the left (L) lever into the chamber. Fifteen responses on this lever (fIXed ratio, fR{.15) retracted the lever from the chamber and extinguished the jewel light. No reinforcement was delivered at this point. A mid·trial interval (MTI) of :2 sec was then followed by the insertion ~f the right (R) lever into the u :r . . Psyclton. Sci.. 1968. Vol. t I (9) chamber and the onset of its corresponding jew...

Delay of reinforcement in instrumental discrimination learning of rats.

“…Under two conditions, however, delay of reinforcement has been shown to facilitate instrumental performance. Specifically, when a sequence of nondiscriminative instrumental responding is interrupted for some period of time prior to the end of the sequence, rate of responding following the delay point may be enhanced (Carlson, 1968;Holder, Marx, Holder, & Collier, 1957). Similarly, delay of reinforcement in the form of time-out from reinforcement following incorrect responses may improve accuracy in a matching-tosample discrimination task (Zimmerman & Ferster, 1963).…”