Performance in multiple fixed-interval schedules.
In multiple FI schedules, response rates follow reinforcement rates, not component durations, across birds and people.
01Research in Context
What this study did
Duncan et al. (1972) worked with six pigeons in a small chamber.
Each bird faced two fixed-interval (FI) schedules in one session.
Pecks on the left key paid grain after 60 s. Pecks on the right key paid after 120 s.
The team switched the two parts every few minutes and counted how fast the birds pecked.
What they found
Birds pecked twice as fast on the 60-s key as on the 120-s key.
The length of each part (2, 4, or 8 min) did not change the speeds.
Response rates tracked how often grain came, not how long the part lasted.
How this fits with other research
Timberlake et al. (1987) later ran the same FI setup with adult humans.
Money replaced grain and a green light told them when the FI was running.
People also showed the classic FI “break-and-run” pattern, proving the rule works across species.
Tracey et al. (1974) used handwriting instead of key pecks.
Two adults wrote faster when money came every 30 s than every 60 s.
Again, rate followed the pay schedule, not the session length, matching the pigeon data.
Why it matters
When you set multiple FI schedules, clients will work harder where pay-offs come sooner, no matter how long each activity block lasts.
Use this to balance work and rest periods in skill-building or DRL programs.
Keep the richer schedule short and frequent to keep engagement high.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Put the richer reinforcer on the shorter FI and watch responding rise.
02At a glance
03Original abstract
The performances of five pigeons were studied under a variety of multiple fixed-interval schedules in which both component duration and reinforcement rate were varied. The three series of experimental conditions were: (a) when the ratio of component durations equalled the reciprocal of the ratio of component reinforcement rates; (b) when the component durations were equal; and (c) when the ratio of component durations equalled the ratio of component reinforcement rates. Relative response rates were related to relative reinforcement rates in the same manner as in multiple variable-interval schedules, but no effect of component duration was found.
Journal of the experimental analysis of behavior, 1972 · doi:10.1901/jeab.1972.17-375