Abstract
College undergraduates were given repeated opportunities to choose between a fixed-ratio and a progressive-ratio schedule of reinforcement. Completions of a progressive-ratio schedule produced points (exchangeable for money) and incremented that response requirement by 20 responses with each consecutive choice. In the reset condition, completion of a fixed ratio produced the same number of points and also reset the progressive ratio back to its initial value. In the no-reset condition, the progressive ratio continued to increase by increments of 20 throughout the session with each successive selection of this schedule, irrespective of fixed-ratio choices. Subjects' schedule choices were sensitive to parametric manipulations of the size of the fixed-ratio schedule and were consistent with predictions made on the basis of minimizing the number of responses emitted per point earned, which is a principle of most optimality theories. Also, the present results suggest that if data from human performances are to be compared with results for other species, humans should be exposed to schedules of reinforcement for long periods of time, as is commonly done with nonhuman subjects.