Timing of avoidance responses by rats.
Learners time avoidance acts in proportion to the interval length, so prompt placement should follow the same ratio, not fixed seconds.
01Research in Context
What this study did
Mulvaney et al. (1974) watched rats in a shuttlebox with no warning lights or tones.
The rats could cross to the other side at any time to avoid mild shock.
The team tested three interval lengths: 10, 20, and 40 seconds.
What they found
The rats did not wait a fixed number of seconds.
Instead, they spaced their crossings in proportion to the interval in force.
Longer intervals produced wider, steady timing waves of avoidance.
How this fits with other research
Leander et al. (1972) first showed that unsignalled shuttlebox avoidance is learned quickly and reaches high efficiency. Mulvaney et al. (1974) zoom in and explain how the animal times those same responses once the skill is stable.
KIEFFETHOMAS (1965) found rats pause longer before tougher ratio schedules. Mulvaney et al. (1974) extend this timing skill to negative reinforcement, showing proportional waiting inside each shock-avoidance cycle.
Blough (1971) showed rats time their responses from clear warning signals. Mulvaney et al. (1974) remove all signals and reveal the animal can still use elapsed clock time, demonstrating flexible temporal control.
Why it matters
If you run avoidance or escape programs, do not lock prompts to fixed seconds. Space your cues at one-quarter, one-half, and three-quarters of the total interval you set. This matches the natural gradient the rat data show and should keep human learners in better rhythm with the schedule.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Cut your next avoidance interval into quarters and deliver prompts at the 25%, 50%, and 75% marks instead of at set second counts.
02At a glance
03Original abstract
Three rats were trained on an unsignalled shuttlebox-avoidance task under three response-shock intervals (10, 20, and 40 sec). Under all conditions, subjects developed excellent temporal gradients of avoidance; that is, response rate was an increasing function of time since last response. Although the response rate at any given interval of time after the previous response was inversely related to the response-shock interval, there was an underlying similarity in the temporal gradients for the three intervals. In all cases, response rate relative to the maximum response rate was approximately equal to the proportion of the interval that had elapsed. This suggests that rats in unsignalled avoidance are estimating time from response completion, and that the units of the estimate are proportional parts of the response-shock interval.
Journal of the experimental analysis of behavior, 1974 · doi:10.1901/jeab.1974.22-513