| Abstract: | A discrimination was established between two fixed-ratio schedules of reinforcement. In one, fixed ratio 25, the reinforcer was delivered on the twenty-fifth response; on the other, fixed ratio 50, the fiftieth response was reinforced. In the first component of a chain, either fixed ratio 25 or fixed ratio 50 was randomly programmed on the center key of a three-key pigeon box. Reinforcement of a single peck on the side key was contingent upon discriminating which schedule had just been completed on the center key. During test trials, a timeout was introduced after the first response on fixed ratio 25 and after either the first or twenty sixth response on fixed ratio 50. When the timeout followed the first response on fixed ratio 25 and fixed ratio 50, the accuracy of the discrimination was unaffected. When the timeout followed the first response on fixed ratio 25 and the twenty sixth response on fixed ratio 50, the accuracy of the discrimination decreased rapidly to chance as a function of the duration of the timeout. The loss of discrimination was primarily due to errors after fixed ratio 50 was completed. The timeout appears to weaken the control over the choice response by the response-produced stimuli which preceded the timeout. The results are consistent with the interpretation that the discrimination between fixed ratio 25 and fixed ratio 50 is maintained by chaining of response-produced stimuli within the ratio cycle. |
