Behavioral control by the response–reinforcer correlation
The link between how often a response happens and how often it gets reinforced can drive behavior even when every pellet is delayed—so audit your schedules for accidental correlations.
01Research in Context
What this study did
The team built a simple operant box for rats. A lever press gave food only if the animal’s own rate of pressing matched a hidden rule.
They switched the rule across sessions. Sometimes high pressing paid off. Sometimes low pressing paid off. Sometimes the pay-off was random no matter how fast the rat pressed.
Each rat served as its own control. The only thing that changed was the correlation between the rat’s response rate and the food that followed.
What they found
When high pressing produced more food, rats pressed hard. When low pressing produced more food, rats eased off.
Even tiny time gaps between the press and the pellet still let the correlation rule the show. The rats acted like they could ‘see’ the pattern even without immediate, one-to-one pairing.
How this fits with other research
LeBlanc et al. (2003) already showed that richer schedules lift both response rate and resistance to change. Kuroda’s work tightens the lens: it is the correlation itself, not just the amount of food, that does the lifting.
Bensemann et al. (2015) warned that DRO can accidentally strengthen untargeted behavior because any response that lands near reinforcer delivery gets blessed. The new study gives the mechanism—correlation is enough to pull behavior toward whatever happens to be paired, even if you never planned it.
Appel (1968) used conjunctive and interlocking schedules to tweak response–reinforcer dependencies. Kuroda strips away the extra bells and whistles and still gets clean control, showing the older findings are a special case of the broader correlation principle.
Why it matters
Check your schedules for hidden correlations. A DRL meant to reduce talking might accidentally reinforce long pauses if the only praise lands right after silence. A rich VI you think is ‘safe’ could still strengthen off-task behavior if the learner’s fidgets line up with your timer. Run a quick scatterplot of response rate versus reinforcer rate across sessions; if the line slopes up or down, the contingency is talking louder than you meant it to.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Graph each client’s response rate against reinforcer rate for the last five sessions; rescale or add extinction windows if the slope drifts from your target.
02At a glance
03Original abstract
Using a discrete-trials procedure, two experiments examined the effects of response-reinforcer correlations on responding while controlling molecular variables that operated at the moment of reinforcer delivery (e.g., response-reinforcer temporal contiguity, interresponse times preceding reinforcement). Each trial consisted of three successive components: Response, Timeout, and Reinforcement, with the duration of each component held constant. The correlation between the number of responses in the Response component and reinforcer deliveries in the Reinforcement component was varied. In the Positive-correlation condition, a larger number of responses in the Response component programmed a higher reinforcement rate (Experiment 1) or a shorter time to reinforcement (Experiment 2) in the Reinforcement component. Although programmed in this way, the actual reinforcer delivery was dependent on, and occurred immediately after, a response in the Reinforcement component. In the Zero-correlation condition, the programmed rates of reinforcement (Experiment 1) or the times to reinforcement (Experiment 2) in the Reinforcement component of each trial were yoked to those in the preceding Positive-correlation condition. Responding in the Response component was higher in the Positive- than in the Zero-correlation condition, without systematic changes in molecular variables. The results suggest that the response-reinforcer correlation can be a controlling variable of behavior.
Journal of the Experimental Analysis of Behavior, 2018 · doi:10.1002/jeab.461