Analysis of potentiation and overshadowing effects in the instrumental performance of pigeons

In each of two experiments, pigeons were trained on a multiple VI (variable interval) schedule with a 3-s delay of reinforcement. Different components were associated with different key colors. Experiment 1 (stage 2) confirmed a previous finding that the response rate is higher in a component having a signal (illumination of the houselight) filling the delay interval than in a component lacking the signal. This potentiation effect was replaced by an overshadowing effect (i.e., the rate was low in the signaled component) when in stage 1 of Experiment 1 the birds received concurrent experience of a component containing houselight presentations not correlated with reinforcement. In Experiment 2 it was found that this overshadowing effect was abolished when the signal used was the presentation of a pattern on the response key rather than illumination of the houselight. These results are interpreted in terms of an interaction between the rate-enhancing properties of the signal (perhaps a consequence of its conditioned reinforcing power) and the tendency of the signal in some conditions to evoke behavior that competes with the response being recorded.