Interactions in multiple schedules: the role of the stimulus-reinforcer contingency.
In multiple schedules, brief rich components at the start boost initial responding.
01Research in Context
What this study did
Dove (1976) worked with pigeons in a lab. The birds pecked a key under a two-part multiple schedule.
Each part had its own colored light and its own rate of food delivery. The team varied how long each part lasted and how much food it gave.
What they found
Response rates jumped or dipped right after the color changed. The size of the jump tracked the food ratio and the component length.
Shorter, richer parts sparked the fastest early pecks. The birds acted as if they counted the local payoff odds.
How this fits with other research
Last et al. (1984) extends this idea. They showed the same early-burst pattern follows a power law and grows even stronger when the two colors look very different.
McLean (1988) seems to disagree at first. P found that changing the food rate in one part left the other part untouched. The two papers clash only on the surface: D looked at the first seconds after a switch, while P looked at steady-state minutes later. Early contrast fades into long-term independence.
Kodera et al. (1976) used the same pigeon setup but swapped in brief tones for food. They remind us that any stimulus must first be paired with food to work, a step D’s color lights already passed.
Why it matters
When you run mixed programs like DTT plus free play, think of each segment as a multiple-schedule component. Start the richer segment (high praise, high tokens) in short bursts to lift early engagement. Then stretch the leaner segment once the momentum is up. The data say the learner’s first-minute rate is set by the local payoff clock, not the session average.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open your next mixed session with a 1-minute high-reinforcement component, then fade to thinner schedules.
02At a glance
03Original abstract
In Experiments I and II, pigeons were exposed to single-key multiple schedules of response-independent and -dependent food presentation. Components were correlated with different keylights. When the rate of food presentation in the first component exceeded that in the second component, the local rate of key pecking was relatively high at onset of the first component. Overall rate in that component varied inversely with component duration and the rate of food presentation in the second component. When responding was maintained in the second component, the local rate of key pecking was relatively low at onset of that component. Overall rate in the second component varied directly with component duration and the rate of food presentation in that component. In Experiment III, pigeons were exposed to a two-key multiple schedule. Pecks on a constantly illuminated key produced food. Components were correlated with the color of a second key on which pecks had no scheduled consequences. The effects of component duration and rate of food presentation under the single-key response-dependent schedule were synthesized by combining response rates on each concurrently available key under the two-key procedure. The results support an account of multiple-schedule interactions in terms of the joint influence on responding of stimulus-reinforcer and response-reinforcer contingencies.
Journal of the experimental analysis of behavior, 1976 · doi:10.1901/jeab.1976.26-79