Preference for multiple versus mixed schedules of reinforcement

Abstract

In a concurrent-chains procedure, Pigeons' responses in identical concurrent variable interval schedules (choice phase) produced either mixed or multiple schedules of food reinforcement. Non-independent variable interval schedules in the choice phase (initial links) ensured equal probabilities of multiple versus mixed schedule production in the terminal links. In the mixed schedule the same stimulus was correlated with different fixed interval schedules (FIx, FIy), whereas in the multiple schedule the FIx and FIy schedules were associated with different discriminative stimuli. Both relative and average reinforcement rates in the multiple and mixed schedules were varied to test Fantino's (1977) hypothesis, that the presence of stimuli correlated short interreinforcement intervals has a disproportionate influence on choice. Preference for multiple versus mixed schedules measured by changes in the ratio of initial-link responses, was not found to vary systematically with either the reinforcement ratio (FIx/FIy) or average reinforcement rate (FIx + FIy) in the multiple or mixed schedules.