With rats as Ss, an analog of shock escape-avoidance was developed in which withdrawal of positive reinforcement served in lieu of noxious stimulation. Efficient escape behavior developed with latencies varying as a function of the rate of reinforcement reinstated by the conditioned response, but avoidance behavior was infrequent. The results were considered from the standpoint of whether the escape response was positively reinforced through onset of the period of reinforcement or negatively reinforced through termination of the period of non-reinforcement.
83 Ss were run in two experiments involving delay of shock termination. In the first, shock was terminated a brief interval after a shock-escape response, while in the second, only responses occurring a brief interval after shock onset were effective in terminating it. Although both procedures were designed to facilitate avoidance acquisition by minimizing freezing behavior, in both cases the use of delay led to decrements in acquisition.
Rats were given shock-escape training in a 4-ft. runway, then divided into 4 groups of 9 Ss each. During extinction, shock was present in the 4-ft. alley, the first 1-ft. segment, or the last 1-ft. segment for 3 groups, but was never present in the start box. A control group received no shock. In general, punished Ss ran faster and longer than non-shocked Ss. The more immediate the punishment, the more vigorous and sustained was the punished act. The immediate 1-ft. shock led to very high resistance to extinction, indicating that an intermediate shock duration may yield optimal facilitation.
Resistance to extinction of an escape response as a function of the number of reinforcements
