A steady-state test of the demand curve analysis of relative reinforcer efficacy

The demand curve analysis of relative reinforcer efficacy predicts that preference for 1 reinforcer over another may be predicted from relative levels of consumption under single-schedule conditions. Prior experiments with human cigarette smokers have largely confirmed this prediction, but because subjects completed only 1 session under each arrangement, questions remain about the stability of single and concurrent-schedule (choice) performances. In addition, no prior tests of this prediction of the demand curve analysis have used nonhuman subjects, so questions remain about the effects of instructions and demand characteristics. In the present study, the authors examined pigeons' steady-state preferences in a closed economy across a wide range of concurrent fixed-ratio and random-ratio (RR) schedules. On the basis of the pigeons' single-schedule consumption, the demand curve analysis qualitatively predicted indifference at low ratio values and preference for the RR alternative at higher ratio values. These predictions proved correct in 88% of the cases in which the analysis predicted preferences in individual subjects at specific concurrent-schedule values. These findings offer preliminary support for the position that the predictions of the demand curve analysis hold true when steady-state performances are assessed. They also offer preliminary evidence that previous confirmations of the demand curve analysis were not due to the influence of instructions or to demand characteristics of the human laboratory setting. These findings supplement prior questioning of the coherence and utility of traditional measures of relative reinforcer efficacy.