EFFECTS OF HUNGER AND VI VALUE ON VI PACING¹

Abstract

In two experiments, each involving four rats, responses preceded by an inter-response time between 8 and 10 sec in duration were intermittently reinforced. In Experiment I, final performance was compared under two hunger levels, while the frequency of reinforcement was held constant by a VI 5 schedule. In Experiment II, hunger was held constant and VI 3 was compared with VI 8. Both hunger and frequency of reinforcement increased the over-all rate of response, but the exact effects of these operations on temporal discrimination were different for different rats. Usually, a peak "response probability" (IRTs/Op ratio) was obtained 8 to 10 sec after the preceding response, indicating adaptation to the reinforcement contingency, but in some cases this peak was about 2 sec earlier. One rat exhibited unusually pronounced bursting which seemed to alternate with adaptive temporally spaced responding. Prolonged pauses, observable in the cumulative records, particularly following reinforcement, were attributed to the fact that inter-response times greater than 10 sec were not reinforced, so that as the interval of time since the preceding response became discriminably greater than 10 sec, the probability of a response became small.

AN EXPERIMENTAL SPACE COMBINING INDIVIDUAL AND SOCIAL PERFORMANCES1

“…In the delayed-reinforcement experiments (Ferster and Hammer, 1965) where rhesus monkeys were introduced to long delays of reinforcement without the required history of successive approximations, the animals lost weight continuously without any corresponding increase in the food-reinforced behavior. These data are in contrast to many classical operant experiments (Clark, 1958;Revusky, 1963;Perin, 1942;Conrad, Sidman, and Herrnstein, 1958) which demonstrated a z I Fig. 3.…”

RESPONSE RATE, REINFORCEMENT FREQUENCY, AND CONDITIONED SUPPRESSION1

“…Finally, it should be noted that the effects of the schedules on Rats 7 to 12 do not appear to be in agreement with an observation by Revusky (1963). He also examined the effects of VI value on behavior generated by response pacing schedules, and concluded that the pacing contingency "did not eliminate the usual effects of .…”

THE REINFORCEMENT OF SHORT INTERRESPONSE TIMES1

“…These variables undoubtedly affect the paced distributions and therefore need to be included in the rationale. Some work has been done (Revusky, 1963) but so far the results have been too inconclusive to suggest any specific modifications or generalizations of the present formulation.…”