Shock intensity and duration interactions on free-operant avoidance behavior.
In avoidance, doubling shock duration works like doubling intensity—manipulate either, but multiply both for the same behavioral shift.
01Research in Context
What this study did
Kelly (1973) worked with lab rats in a free-operant avoidance box.
Each lever press postponed a brief foot-shock.
The team varied both how strong the shock was (0.2–0.8 mA) and how long it lasted (0.1–1.6 s).
They counted how fast the rats pressed and how many shocks they actually got.
What they found
Response rate climbed in a straight line when the researchers multiplied intensity × duration on a log scale.
Shock rate dropped the same way.
Doubling the strength had the same punch as doubling the length—no need to crank the dial harder alone.
How this fits with other research
McKearney (1970) and Hake et al. (1967) already showed that stronger shocks suppress or punish faster, but they used punishment, not avoidance.
Thomas et al. (1968) used the same two-knob design (intensity × duration) and saw monkey biting rise the same way lever pressing rose here—parallel laws, different response.
Last et al. (1984) flips the story: shock can also reinforce behavior when it is tied to long pauses, proving the same electrons can punish, maintain, or allow escape depending on contingency.
Why it matters
When you adjust aversive stimuli in treatment or analog studies, think product, not just power.
Lengthening a mild consequence can match shortening a strong one, giving you a wider, kinder menu.
Always pair the parametric change with a clean contingency so the learner knows exactly what to do.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →If you must thin a warning stimulus, try doubling its length before you raise its volume or intensity.
02At a glance
03Original abstract
Shock intensities (1 to 4 mA) and shock durations (0.3 to 0.75 sec) were concurrently varied over a range commonly used in free-operant avoidance studies using a lever-press response. Response rates were a positive linear function of the log of the product of intensity times duration. Shock rates were a negative linear function of that log. The increase in response rates was primarily due to a selective increase in the conditional probability of making responses with long interresponse times. The disproportionality of receiving shocks early in the session (warm-up) was also a linear function of the log of the intensity-duration product, with increasing disproportionality as the value of the intensity-duration product was increased. Thus, with all measures of the avoidance performance, shock intensity and shock duration combine in a multiplicative fashion to determine the avoidance performance.
Journal of the experimental analysis of behavior, 1973 · doi:10.1901/jeab.1973.19-481