Frequency of reinforcement as a parameter of conditioned suppression.
Dense reinforcement protects against fear-based drops, yet can fuel later resurgence.
01Research in Context
What this study did
Evans (1963) worked with pigeons on a two-key setup. One key paid every minute on average. The other paid every four minutes.
A scary tone came on sometimes. It signaled a mild shock. The birds could still peck for food. The team watched how the tone cut pecking.
What they found
Birds on the richer key almost kept pecking. Birds on the lean key froze. When the tone stopped, the rich-key birds bounced back faster.
More pay meant less fear effect. Less pay meant bigger fear effect.
How this fits with other research
Fisher et al. (2019) flips the story. In kids with problem behavior, richer pay during FCT made more resurgence later. Lab says dense = safe. Clinic says dense = risky.
LYOSLOANE (1964) shows timing matters. A scary cue early in a long run stops birds cold. A cue near the end lets them finish. Same fear, different spot, different result.
MIGLEMOORHEARSKELLEHER et al. (1964) adds that fear cuts rate but leaves timing intact. Birds peck less often, yet still know when to peck.
Why it matters
When a client shuts down, check the pay rate. A thicker schedule can shield the skill while you ride out the scare. But watch for later resurgence when you thin. Start thick, then fade slow.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Boost reinforcement to at least one per minute during new or scary tasks.
02At a glance
03Original abstract
Two pigeons were trained on a multiple VI-1; VI-4 min schedule for food reinforcement until a stable differential rate of response was established. Each component of the multiple schedule was in effect for 10 min and was separated from the other by 1 min time out. An Estes-Skinner conditioned suppression procedure was superimposed on each component of the schedule. The relative magnitude of the suppression behavior was measured during its acquisition and extinction, and at CS durations of 100, 200, and 300 sec. The initial magnitude of the suppression behavior was less severe on the VI-1 baseline than on the VI-4, and it extinguished more rapidly on the VI-1. As the relative duration of the CS was increased, the suppression behavior became less severe on both baselines, but the initial differential magnitude in the suppression remained intact.
Journal of the experimental analysis of behavior, 1963 · doi:10.1901/jeab.1963.6-95