Choice, changing over, and reinforcement delays.
Total delay to reinforcement, not the cue at the switch, decides how often learners hop between options.
01Research in Context
What this study did
Nasr et al. (2000) watched pigeons switch between two keys that gave food at different times.
The birds could hop to the other key at any moment.
The team added tiny waits between a hop and the next payoff to see which delay really controlled hopping.
What they found
Total wait time from last peck to next payoff slowed the hopping.
A short flash or tone change at the hop barely mattered.
Birds hopped less when the whole trip to food grew longer, not when the cue came quicker.
How this fits with other research
Eisenmajer et al. (1998) saw that a surprise 3-s wait slashed preference and staying power.
Nasr et al. (2000) agrees: long waits hurt, but shows the hurt comes from total delay, not the cue.
Torres et al. (2011) later used a moving-delay tool and found the same curve shape, backing the power-law link.
Vos et al. (2013) then showed the rule also holds when pigeons work on fixed-ratio jobs, not just free choice.
Why it matters
When you run concurrent schedules, watch the real travel time to reinforcement. Cut extra steps, long token lines, or slow delivery. Keep the path short and the learner stays put.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Trim any extra seconds between a response on one alternative and the next reinforcer delivery.
02At a glance
03Original abstract
In three experiments, pigeons were used to examine the independent effects of two normally confounded delays to reinforcement associated with changing between concurrently available variable-interval schedules of reinforcement. In Experiments 1 and 2, combinations of changeover-delay durations and fixed-interval travel requirements were arranged in a changeover-key procedure. The delay from a changeover-produced stimulus change to a reinforcer was varied while the delay between the last response on one alternative and a reinforcer on the other (the total obtained delay) was held constant. Changeover rates decreased as a negative power function of the total obtained delay. The delay between a changeover-produced stimulus change had a small and inconsistent effect on changeover rates. In Experiment 3, changeover delays and fixed-interval travel requirements were arranged independently. Changeover rates decreased as a negative power function of the total obtained delay despite variations in the delay from a change in stimulus conditions to a reinforcer. Periods of high-rate responding following a changeover, however, were higher near the end of the delay from a change in stimulus conditions to a reinforcer. The results of these experiments suggest that the effects of changeover delays and travel requirements primarily result from changes in the delay between a response at one alternative and a reinforcer at the other, but the pattern of responding immediately after a changeover depends on the delay from a changeover-produced change in stimulus conditions to a reinforcer.
Journal of the experimental analysis of behavior, 2000 · doi:10.1901/jeab.2000.74-311