The effects of ratio and interval schedules on the location variability of pecking responses in pigeons: Application of Bayesian statistical model.

Many studies have investigated how variability in animal behavior is induced by different reinforcement schedules. However, the animal species and experimental settings have varied between these studies. The present study investigated the variability of pigeons' pecking location to operandum under fixed-ratio, fixed-interval, variable-ratio, and variable-interval schedules. A circular response area that was 22.4 cm in diameter was used so that the pecking responses would be effective over a wide range. Pigeons were exposed to a multiple-ratio yoked-interval schedule; the pairs of schedules were fixed-ratio and fixed-interval or variable-ratio and variable-interval. We used a Bayesian statistical approach to probabilistically evaluate the effects of reinforcement schedules on the variability. Bayesian statistical analysis showed the following: (1) interval schedules provided greater variability than did the ratio schedules under the same interreinforcement intervals, (2) fixed schedules provided greater variability than did variable schedules for both the ratio and interval schedule requirements, (3) the larger schedule requirement generated greater variability under the fixed schedules but this effect was not observed on the variable schedules, and (4) the variability did not change as time elapsed in each trial. These results suggested that each reinforcement schedule has a specific effect on the variability of the response location.