Interactions in second-order schedules: an analysis of fixed-interval schedule control

Abstract

Pigeons were exposed to several types of two-key multiple second-order schedules. A tandem-plus-brief-stimulus schedule was used as the baseline schedule of the study. The unit schedule of the baseline was a fixed-interval (FI) and the pigeons were reinforced with food after completing on the average five FI units with associated brief stimuli. During the experimental phase of the study one of the second-order schedules was changed so that the terminal FI unit in the second-order schedule had a different key light. This schedule was a chain-plus-brief-stimulus schedule and was used in four experimental conditions, two for each key. During the control phase of the study the brief stimuli were removed from each key in separate conditions. The second-order schedule without brief stimuli was a tandem schedule. The results of the study showed that a higher rate of responding was maintained in the baseline schedule than in either of the experimental or control conditions. Scalloping in the FI units was evident in the baseline schedule but was degraded in the tandem-schedule and in the initial portion of the chain-plus-brief-stimulus schedule. The rate of responding was generally higher in the terminal unit of the chain-plus-brief- stimulus schedule than in the terminal unit of the tandem-plus-brief-stimulus schedule. These results show that fixed-interval schedule control was degraded when there was a differential cue for food in a second-order schedule. The results also suggest that the brief stimuli functioned as conditioned reinforcers. The results were discussed in terms of the "contiguity hypothesis" and "information hypothesis."