Suppression of response rates in variable‐interval schedules by a concurrent schedule of reinforcement

Herrnstein's equation, which describes the relationship between reinforcement frequency and response rate in variable‐interval schedules predicts that the introduction of a concurrent source of reinforcement will increase the reinforcement frequency needed to obtain the half‐maximal response rate, without affecting the theoretical maximal response rate itself. This prediction was tested in the present experiment. During Phase I four rats were trained to press a lever (lever A) in variable interval schedules, using a range of reinforcement frequencies. During Phase II, a second lever (lever B) was introduced; responses on lever B were always reinforced at the same frequency, while the reinforcement frequency for lever A responses was varied. During Phase I the data obtained from all four rats conformed closely to Herrnstein's equation. Comparing the absolute response rates in Phase I and Phase II it was found that the predictions based on Herrnstein's equation were qualitatively confirmed, although there was some quantitative discrepancy between the predicted and actual degree of response suppression during Phase II.