On the effects of noncontingent delivery of differing magnitudes of reinforcement.
Bigger free reinforcers crush problem behavior faster, yet set the stage for a later surge when you try to fade them.
01Research in Context
What this study did
The team tested how reinforcer size changes NCR. They used a multielement design with adults who had intellectual disabilities.
Sessions alternated among high, medium, and low magnitudes of the same edible. The target response stayed the same; only the amount of free food changed.
What they found
Big, free bites cut responding the most. Medium and tiny bites barely dented the behavior.
The high-magnitude condition gave fast, steady suppression. Low magnitude looked almost like no NCR at all.
How this fits with other research
Eugenia Gras et al. (2003) later repeated the test and added session length. They saw the same dose effect: larger or longer NCR packs more punch.
Saini et al. (2017) extended the idea by mixing extinction with NCR. They found big NCR alone drops behavior quickly but can bounce back later; adding extinction keeps the gains.
Shahan et al. (2025) flips the coin. They show that rich histories create bigger extinction bursts. So the same big reinforcers that suppress now may fuel a burst when you later thin the schedule.
Why it matters
Pick your NCR size with the end in mind. Use a hefty magnitude for fast crisis reduction, but plan an extinction or thinning step to block resurgence. If you only need mild dampening, keep the free reinforcer small—you will save food, time, and later grief.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Start high-magnitude NCR for the top one behavior, but schedule an extinction or thinning phase within two weeks to guard against resurgence.
02At a glance
03Original abstract
We conducted a parametric analysis of response suppression associated with different magnitudes of noncontingent reinforcement (NCR). Participants were 5 adults with severe or profound mental retardation who engaged in a manual response that was reinforced on variable-ratio schedules during baseline. Participants were then exposed to NCR via multielement and reversal designs. The fixed-time schedules were kept constant while the magnitude of the reinforcing stimulus was varied across three levels (low, medium, and high). Results showed that high-magnitude NCR schedules produced large and consistent reductions in response rates, medium-magnitude schedules produced less consistent and smaller reductions, and low-magnitude schedules produced little or no effect on responding. These results suggest that (a) NCR affects responding by altering an establishing operation (i.e., attenuating a deprivation state) rather than through extinction, and (b) magnitude of reinforcement is an important variable in determining the effectiveness of NCR.
Journal of applied behavior analysis, 1998 · doi:10.1901/jaba.1998.31-313