ABA Fundamentals

Response‐force changes early in extinction with and without a changing force criterion during training

Alesssandri et al. (2026) · Journal of the Experimental Analysis of Behavior 2026
★ The Verdict

Lower the force requirement right before extinction to shrink the leftover muscle output.

✓ Read this if BCBAs who use extinction with any operant response that involves push, pull, squeeze, or lift.
✗ Skip if Practitioners working only with vocal or screen-touch responses where force is flat.

01Research in Context

01

What this study did

Alesssandri et al. (2026) asked a simple question: does the force you required during training change how fast the response fades when reinforcement stops? They taught participants to press a disk with either a high or low force. Then they turned off the food and watched what happened to the push strength in the first minutes of extinction.

02

What they found

Response force dropped as soon as reinforcement ended. The key twist: people who had to press hard during training kept using more muscle for longer than people who only had to tap lightly. A high-force history slowed the slide; a low-force history let the response wilt faster.

03

How this fits with other research

Pinkston et al. (2018) ran almost the same lab setup and saw no effect of force on how long responding lasted. The two papers seem to clash, but Alesssandri looked at the very first minutes and measured push strength, while Pinkston counted total responses over full sessions. Different windows, different stories.

Lambert et al. (2024) moved the question into real life. They showed that adults with developmental disabilities fought extinction longer when they had been allowed to "consume" lots of reinforcer right before the stop. Alesssandri adds force to the list of pre-extinction levers you can adjust.

Older rat work (T et al., 1963) already warned that more continuous reinforcement makes the first extinction burst bigger. Alesssandri updates the warning: a tough response topography can do the same for human muscle output.

04

Why it matters

Before you place a behavior on extinction, lighten the response requirement for a few trials. A softer press, a lighter card swipe, or a gentler door pull today can save you from hard, emotional bursts tomorrow. You get the same final quiet, only faster and with less wear on the client—and on you.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Drop the response-force criterion to the easiest level for three trials, then start extinction.

02At a glance

Intervention
extinction
Design
single case other
Population
neurotypical
Finding
null

03Original abstract

This experiment was designed to examine the question of how different force‐exertion requirements in effect prior to extinction affect force exertion during extinction of the previously reinforced response, with an emphasis on such effects early in extinction. Human participants were exposed to one of three conditions in which making a force‐exertion response resulted in points displayed on a computer screen. In two conditions, the response‐force requirement was fixed during the reinforcement phase at a force exertion of either 50%–65% or 100%–125% of the force criterion exerted in a pretest. During the third condition, the force‐exertion criterion was decreased progressively from 100%–125% to 50%–65% of the force criterion during the reinforcement phase. After a short adjustment period, response‐force exertions generally conformed to the force requirements for reinforcement. Removing the opportunity for reinforcement reduced the number of responses relative to those occurring in the reinforcement phase, although some responding was still occurring for most participants at the end of the extinction phase. The results are discussed in relation to the variables responsible for the extinction of a force‐defined response, emphasizing changes in force early in extinction.

Journal of the Experimental Analysis of Behavior, 2026 · doi:10.1002/jeab.70066