PERFORMANCE ON CONCURRENT VARIABLE-INTERVAL EXTINCTION SCHEDULES

Six homing pigeons were trained on concurrent variable-interval extinc tion schedules in a switching-key procedure. The discriminative stimul i and associated schedules in operation were selected randomly after e ach switch and each reinforcer. More than 80 daily sessions were arran ged in each of five experimental conditions that varied the reinforcer rate on the variable-interval schedule. Behavior allocation remained nonexclusive even after extended training and did not change systemati cally as a function of the reinforcer rate. Both of these findings are predicted by a contingency-discriminability description of choice and are incompatible with a generalized matching description.