A facilitative effect of punishment on unpunished behavior.
Punishing one response can briefly inflate other untreated responses through behavioral contrast, so monitor the whole response class.
01Research in Context
What this study did
The team worked with pigeons that pecked a key for food on a VI schedule.
While the birds pecked, the researchers added mild electric shock to some key pecks.
They counted how shock changed both the shocked pecks and the pecks that never got shocked.
What they found
Shock quickly cut the punished key pecks.
At the same time, the unpunished key pecks rose above their old rate.
When shock stopped, both rates flipped back, showing a clear contrast effect.
How this fits with other research
Dunham et al. (1969) ran the same setup with an ABAB design and saw the same jump in off-key pecks, a direct replication.
Winett et al. (1972) and Dunham (1972) swapped in different responses—drinking, running, avoidance—and still got the same boost in unpunished behavior, proving the effect holds across topographies and schedules.
Shearn et al. (1997) later revisited the question with both pigeons and humans and found contrast fades fast while induction is more common, suggesting the original finding is real but fragile and context-bound.
Why it matters
When you punish one behavior, watch for a surge in untreated behaviors that serve the same function. The boost may feel like progress, but it is just contrast. Track all relevant responses, not just the target, so you can adjust your plan before unintended behaviors steal the show.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a simple tally of untreated alternate behaviors before and after you introduce any punishment procedure.
02At a glance
03Original abstract
The key pecking of two pigeons was reinforced on a variable-interval schedule of reinforcement during the presentation of each of two stimuli. In various phases of the experiment, punishment followed every response emitted in the presence of one of the stimuli. In general, when the rate of punished responding changed during the presentation of one stimulus, the rate of unpunished responding during the other stimulus changed in the opposite direction. This sort of change in rate is an example of behavioral contrast. When punishment was introduced, the rate of punished responding decreased and the rate of unpunished responding increased as functions of shock intensity. When the rate of previously punished responding increased after the termination of the shock, the rate of the always unpunished responding decreased. When the procedure correlated with a red key was changed from variable-interval reinforcement and punishment for each response to extinction and no punishment, the rate of reinforced responding during presentations of a green key decreased and then increased while the rate of the previously punished responding during red first increased and then decreased during extinction.
Journal of the experimental analysis of behavior, 1962 · doi:10.1901/jeab.1962.5-191