Autism & Developmental

Delayed reversal learning and association with repetitive behavior in autism spectrum disorders.

South et al. (2012) · Autism research : official journal of the International Society for Autism Research 2012
★ The Verdict

Autistic learners need more trials to adjust when rules flip, and the lab delay mirrors daily rigidity.

✓ Read this if BCBAs working with school-age autistic clients who struggle when routines change.
✗ Skip if Clinicians focused only on early intensive behavior intervention with toddlers.

01Research in Context

01

What this study did

South et al. (2012) ran a lab game with youth who have autism and youth who do not. The kids first learned that one picture meant "safe" and another meant "threat." After several trials the rule flipped: the old safe picture now meant threat and vice-versa.

The team counted how many extra trials each child needed before they stopped following the old rule. They also asked parents about everyday rigid behaviors like needing the same route to school.

02

What they found

Kids with autism needed more trials to switch to the new rule. The slower they were in the game, the more intense their real-world repetitive behaviors were reported to be.

Neurotypical kids adjusted faster; their game scores did not relate to everyday flexibility.

03

How this fits with other research

Faja et al. (2015) and Leezenbaum et al. (2019) saw the same delay pattern in younger children using marshmallow-style "wait for a treat" tasks. These studies conceptually replicate the target: autistic learners show slower self-control across different ages and tasks.

Adams et al. (2021) extends the idea to eye-movement control. They found the slowdown is specific to proactive control — getting ready before the signal — not to quick emergency stops. This hints that extra warning time, not louder stop cues, may help.

Konke et al. (2026) looks like a contradiction at first: they report a positive outcome, showing that stronger delay skills protect adaptive functioning in toddlers. The twist is measurement direction. Mikle shows autism predicts poorer delay; Andersson shows that when autistic kids do delay well, life skills are better. Both can be true and together argue for teaching delay skills early.

04

Why it matters

If a learner stalls when classroom rules change, the issue may be cognitive flexibility, not defiance. Build extra practice trials into discrimination tasks and give pre-signals before rule switches. Pair these with everyday delay training like waiting for the iPad. Strengthening flexible waiting in clinic may soften rigid routines at home and school.

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

Insert two extra reversal trials after each discrimination set and pre-cue the upcoming rule change with a visual prompt.

02At a glance

Intervention
not applicable
Design
quasi experimental
Sample size
59
Population
autism spectrum disorder, neurotypical
Finding
negative
Magnitude
medium

03Original abstract

An important aspect of successful emotion regulation is the ability to adjust emotional responses to changing environmental cues. Difficulties with such adaptation may underlie both marked symptoms of behavioral inflexibility and frequent severe anxiety in the autism spectrum disorders (ASDs). Thirty children and adolescents diagnosed with ASD and 29 age- and intelligence quotient-matched controls completed a reversal learning paradigm following partial reinforcement Pavlovian fear conditioning, using a surprising air puff as the unconditioned stimulus. After initial reversal of cue contingencies, where a previously safe cue now predicted the air puff threat, the control group but not the ASD group responded more strongly to the new threat cue. The ASD group showed evidence for reversal learning only during later trials. Reversal learning in the ASD group was significantly negatively correlated with everyday symptoms of behavioral inflexibility but not with everyday anxiety. Understanding shared associations between inflexibility, anxiety, and autism, with regard both to clinical symptoms and neurobiological mechanisms, can provide important markers for better characterizing the substantial heterogeneity across the autism spectrum.

Autism research : official journal of the International Society for Autism Research, 2012 · doi:10.1002/aur.1255