Effects of reinforcement history on responding under progressive-ratio schedules of reinforcement.
Reinforcement history tweaks the first few responses on a new ratio schedule, then disappears.
01Research in Context
What this study did
Matson et al. (1994) tested how past reinforcement affects new work. Pigeons first worked on easy or hard fixed-ratio schedules. Later the birds moved to a progressive-ratio schedule where the work needed for food kept rising.
The team watched how fast the birds started and how long they paused after food. They also gave drug and extinction probes to see if history would pop back up.
What they found
Birds with the harder history started faster and paused less on the first few ratio steps. These differences vanished after only a few rewards. When the drug or extinction came, history did not return.
In short, reinforcement history tweaked the opening moves, then washed out.
How this fits with other research
Rasing et al. (1992) showed that when the same colored key signaled the old schedule, history controlled responding for many sessions. L et al. removed the signal, so the brief effect fits: stimulus control keeps history alive.
Adkins et al. (1997) found that DRH versus DRL history biased variable-interval rates for months. The target study saw effects for minutes. The gap looks like a contradiction, but the earlier work kept the same response form across phases while L et al. switched to a new progressive task, letting the history fade.
Bennett et al. (1998) later showed interval schedules create tougher response persistence than ratio schedules. L et al. adds that whatever persistence you start with, ratio requirements quickly overwrite recent history.
Why it matters
When you shift a client to a new reinforcement schedule, do not assume yesterday’s contingencies will guard or haunt you. If the teaching setup changes—new materials, new room, new staff—history effects may vanish as fast as they did here. Probe early performance, but trust the current contingencies to do the heavy lifting. If behavior sticks around, look for stimulus cues, not ancient history.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →When you switch a client to a new ratio, take data on the first three trials—then relax; history won’t bite.
02At a glance
03Original abstract
The effects of experimental history on responding under a progressive-ratio schedule of reinforcement were examined. Sixteen pigeons were divided into four equal groups. Groups 1 to 3 were trained to peck a key for food under a fixed-ratio, variable-ratio, or differential-reinforcement-of-low-rate schedule of reinforcement. After training, these pigeons were shifted to a progressive-ratio schedule, later were shifted back to their original schedule (with decreased rates of reinforcement), and finally were returned to the progressive-ratio schedule. Pigeons in Group 4 (control) were maintained on the progressive-ratio schedule for the entire experiment. To test for potential "latent history" effects, pigeons responding under the progressive-ratio schedule were injected with d-amphetamine and given behavioral-momentum tests of prefeeding and extinction. Experimental histories affected responding in the immediate transition to the progressive-ratio schedule; response rates of pigeons with variable-ratio and fixed-ratio histories were higher than rates of pigeons with differential-reinforcement-of-low-rate and progressive-ratio-only histories. Pigeons with differential-reinforcement-of-low-rate histories, and to a lesser degree pigeons with variable-ratio and fixed-ratio histories, also had shorter postreinforcement pauses than pigeons with only a progressive-ratio history. No consistent long-term effects of prior contingencies on responding under the progressive-ratio schedule were evident. d-Amphetamine and resistance-to-change tests failed to reveal consistent latent history effects. The data suggest that history effects are sometimes transitory and not susceptible to latent influences.
Journal of the experimental analysis of behavior, 1994 · doi:10.1901/jeab.1994.61-375