Secondary reinforcement and information as determinants of observing behavior in monkeys (Macaca mulatta)

Abstract

Responses on a food lever were reinforced on alternating Variable Ratio and Extinction schedules, while concurrent responses on an observing lever produced positive and negative cues correlated with the two schedules. In Experiment I the Variable Ratio requirement was reduced, thereby increasing the density of reinforcement in the presence of the positive cue but reducing average uncertainty as to the effective schedule. The result was a reduction in observing response rates. In Experiment II uncertainty was reduced through presentations of an informative light rather than primary reinforcement, and again the result was a decrease in observing behavior. In Experiment III presentations of the negative cue associated with nonreinforcement were discontinued, and omission of this presumably aversive stimulus again significantly reduced observing behavior. These results were interpreted as generally supporting an information analysis of observing behavior.