Discrimination of Temporally Remote Causal Relations by Pigeons: Effects of Signals that Mediate the Temporal Gaps

The discriminative effect of the response-reinforcer relation may contribute to the change in response rates that occur when reinforcement is delayed. The present study investigated this possibility using a discrete-trials conditional discrimination procedure. Each trial began with a sample component where a variable-interval schedule was assigned either to a left or right key. The key peck that ended the schedule, which served as the sample response, in different conditions initiated delays with a signal fully mediating the delay interval, delays with a signal present only during the first second of the interval, or delays with the absence of signal. The delay in turn was followed by a choice component where one alternative was correct if the sample response had been a left-key response and the other alternative was correct if the sample had been a right-key response. Correct discrimination of the location of the sample response resulted in reinforcement. Accuracy was high with a full signal; slightly lower with a partial, relative to a full, signal; and lowest without a signal. Thus, responses producing delayed reinforcers were detected, but only when a stimulus change accompanied the response. The results parallel the way response rates change when behavior is maintained under a conventional reinforcement schedule as a result of adding delays prior to reinforcement at each type of delay. This suggests a possible role for the discriminative effect of the response-reinforcer relation in the control of behavior by (delayed) reinforcement