Choice of timeout from fixed-time schedules: Comparison of two procedures

Four rats obtained food on fixed-time (FT) 1-, 4-, and 16-min schedules. During FT schedules, a lever press produced a timeout period during which food could not be delivered. When timeout was in effect, a lever press ended the timeout period and reinstated the FT schedule. In different conditions, the timing of the FT interval either stopped for the duration of the timeout period (Stop condition) or the timing of the FT interval continued during timeout (No Stop condition). The percentage of session time spent in timeout periods increased as the time between food deliveries increased regardless of whether or not timing of the FT interval stopped during timeout. The timeout percentage was larger on Stop than No Stop conditions if the obtained interfood interval was also longer. The rate of initiation of timeout periods tended to increase from FT 1-min to FT 4-min schedules, and then not increase further or decrease on FT 16-min. The mean time per timeout period tended to be similar on FT 1- and FT 4-min schedules and then increase on FT 16-min. A given percentage of session time spent in timeout periods appears to be due to interactions between rate of timeout, duration of timeout periods and interfood interval. The data are not consistent with the suggestion that timeout initiation represents escape from aversive aspects of intermittent reinforcement. Instead, choice of timeout shares some of the properties of facultative behaviors.