Punishment and rate of positive reinforcement.
Punishment cuts behavior by the same percentage whether reinforcement is frequent or scarce.
01Research in Context
What this study did
Pigeons pecked a key for food on variable-interval schedules. Some birds got food every 30 seconds on average. Others waited 3 minutes.
The team then added a mild electric shock after every 100 pecks. They watched if the extra food slowed the punishment effect.
What they found
Shock cut responding by the same percent in both groups, about a large share. High-rate or low-rate baseline did not matter.
After shock ended, birds overshot their old rate for a few sessions. Mild shocks even gave short bursts of faster pecking during punishment.
How this fits with other research
McKearney (1970) later showed ratio schedules drop faster than interval schedules when shock grows, but both recover once shock stops. The 1968 study used only interval schedules, so the schedule type rule still holds.
Lea et al. (1977) found brain-stimulation reinforcement resists shock better than food. Their food birds matched the 1968 pattern, tying the rate-independence rule to reinforcer type.
Last et al. (1984) flipped the story: shock for long pauses sped birds up, not down. Timing of shock matters more than the grain size of food.
Why it matters
When you add a punisher, expect the same percent drop no matter how rich the reinforcement schedule is. Do not count on thicker reinforcement to cushion the blow. Watch for brief response bursts right after mild punishment and plan for rebound once the punisher stops. Use this rule to set realistic suppression targets and to explain temporary spikes to staff or parents.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Count baseline rate, then predict a straight percent drop after the punisher is added.
02At a glance
03Original abstract
This experiment investigated the effect of several punishment intensities on two responses maintained by contrasting rates of reinforcement. The responses were concurrently reinforced according to two different variable-interval schedules. Because these schedules were independent of one another and programmed different rates of reinforcement, the two responses occurred at dissimilar rates. When responses were simultaneously suppressed by punishment, both rates were reduced proportionately until suppression was virtually complete. In other words, the per cent suppression resulting from punishment was independent of the rate at which the response was reinforced. Phenomena found in single-response studies were duplicated here. Responding tended to increase both within and between punishment sessions at mild and moderate punishment intensities. Cessation of punishment led to a "compensatory" overshooting beyond the prepunished response rate.
Journal of the experimental analysis of behavior, 1968 · doi:10.1901/jeab.1968.11-285