Changing behavior within session: Cyclicity and perseverance produced by varying the minimum ratio of a variable-ratio schedule

Four pigeons repeatedly chose between a fixed-ratio (FR) 20 and a variable- ratio (VR) 40 schedule of reinforcement, in which the minimum ratio of the VR cycled within each session. The minimum ratio ascended and descended (AS CDESC), descended and ascended (DESCASC), or remained constant (unchanging) . In Phase 1, 2 birds (Group 1) were exposed to ASCDESC series and 2 birds (Group 2) were exposed to the DESCASC series. Choice proportions changed wi th the cycling minimum ratio for Group 2 but not for Group 1. In Phase 2, G roup 1 subjects were exposed to the DESCASC series and Group 2 subjects wer e exposed to the unchanging condition, Although Group 1's choice proportion s appeared to be undifferentiated in Phase 2, Group 2's choice proportions continued to cycle for more than 100 sessions. Group 2 subjects were then m oved to the ASCDESC series in the third phase, and choice proportions cycle d with the minimum ratio as in the first phase. The descending portion of t he series was the more powerful determinant of cyclicity. Response rates al so changed with the minimum component ratio, a finding that goes against th e claim of universality of a rise-and-fall within-session pattern of respon ding. That preference varied despite the constancy of the average ratio req uirement suggests nonlinear averaging in quantitatively representing a vari able schedule's value. The strong perseverance observed also lends support to a growing body of literature on history effects.