The effects of response contingency and reinforcement identity on response suppression by alternative reinforcement

Abstract

Rats were trained on a multiple schedule in which lever pressing was reinforced in one component and nose poking was reinforced in the other. The reinforcers contingent on the two responses were always different, either corn oil or Noyes pellets. A schedule of alternative reinforcement was superimposed in each component, either presenting the same reinforcer as that contingent on the lever/key, or the alternative reinforcer. When the alternative reinforcers were freely presented on a variable time (VT) schedule, their suppressive effects were substantially larger when the free reinforcer was the same as the reinforcer contingent on the lever/key than when it was different. But when the alternative reinforcers were presented according to a variable-interval schedule for a third response (rod pressing), the differences caused by the nature of reinforcement were greatly reduced and not statistically significant. Similar effects occurred when the alternative reinforcement was presented on a VT schedule but preceded by a signal. The results suggest that the effects of alternative reinforcement cannot be explained solely in terms of relative reinforcement rate.