Differential outcome expectancies of reinforcement and nonreinforcement and memory for temporal and nontemporal stimuli

The effects of differential outcome expectancies of food and no-food on memory for temporal and non-temporal stimuli were examined. Pigeons matched short (2s) and long (8s) sample durations to red and green comparison stimuli, and vertical and horizontal lines to vertical and horizontal lines. In Experiment 1, the Nondifferential Outcome group (NDO) received food or no-food on a random half of all trials. The Differential Outcome groups (DO) received food for correct responding to one temporal sample and one nontemporal sample, and no-food following the other samples. In the Differential Outcome-Short-Food group (DO-SF), the short sample stimulus was followed by food, whereas in the Differential Outcome-Long-Food group (DO-LF), the long sample stimulus was followed by food. On linetilt trials, food followed vertical for half of the subjects in each DO group. During a subsequent delayed matching-to-sample (DMTS) testing the NDO group displayed a typical choose-short bias. Other than at the 0s delays interval on no-food trials, the DO groups displayed equivalent biases to the most favourable outcome (choose-favourable bias). In Experiment 2 outcome expectancies were removed off-baseline for the DO groups, followed by a reintroduction of MTS with nondifferential outcomes for all groups. On session 1 of matching-to-sample (MTS), the DO groups performed less accurately than the NDO group. Performance of all groups was equivalent during delay testing. Apparently the performance of the DO groups had been guided by outcome expectancies which overshadowed sample stimulus control. These findings suggest that nonanalogical coding of event duration occurred