Interaction between the components of a chained schedule.

The following study was carried out as a part of the current interest in our laboratory 'in the interrelationships of various components of complex programs of reinforcement. The schedule selected for study, a fixed-interval, fixed-ratio chain (chain FI FR), has been pre-viously reported (Ferster & Skinner, 1957). This schedule is programmed so that the first response after the expiration of x minutes produces a stimulus change (a fixed-interval Fl, schedule with a "secondary reinforcer"). A primary reinforcement, food, is then delivered immediately following the emittance of the "nth" response (a fixed-ratio, FR, schedule with food reinforcement), the stimulus being present from the first FR response until the actual presentation of the reinforcement. In Ferster and Skinner's experiments using pigeon subjects, the performance during the FR component of the schedule was typical of that generated by a simple FR schedule. A high rate of responding started abruptly with the onset of the stimulus and continued with almost no pausing until the reinforcement was delivered. The responding during the FI part of the schedule was typified by a pause after reinforcement, followed by steady or accelerating rates of responding until the stimulus change occurred. However, the chained performance showed less of the smooth acceleration in responding following the postreinforcement pause normally generated by an FI schedule. Responding during the FI portion of the chain frequently started suddenly, after a some-what longer postreinforcement pause, and reached a high rate more quickly than with a simple FI schedule. In 4