Probabilistic reinforcement learning in adults with autism spectrum disorders.
Adults with autism catch on fine when rewards are big and steady, but they miss low-probability cues and need stronger signals to switch choices.
01Research in Context
What this study did
Solomon et al. (2011) asked adults with autism and neurotypical adults to play a simple card game. Each card pair had different chances of winning. Some pairs paid off often. Others paid off rarely.
The team watched who learned which pairs were lucky and who kept choosing the best cards.
What they found
Both groups learned the high-pay pairs equally well. When the payoff was rare, the autism group fell behind. They also kept picking losing cards longer after a rare win.
In short, adults with autism struggle to read shaky odds and to use small bits of good news.
How this fits with other research
van Noordt et al. (2017) helps explain why. They measured brain waves while teens and adults with autism got reward feedback. Their frontal theta rhythm was weaker, showing the brain had trouble locking onto the feedback signal.
Kohls et al. (2011) saw the same broad problem earlier. Kids with autism had smaller brain responses to both money and social rewards. Together, the three studies point to one theme: people with autism notice rewards less, especially when the rewards are small or unlikely.
Johnson et al. (2017) tested ways to fix this in therapy. They tried quality, size, and timing of reinforcers. Quality won most often, but results jumped around. The lab data now make sense: if the learner's brain under-reacts to weak rewards, you must boost size or quality until the signal is clear.
Why it matters
When you run a token board, slot machine schedule, or social praise, remember thin payoffs may not register. Start with rich, sure rewards. Fade the odds later, and watch for persistence on old choices. If the learner stalls, raise reinforcer quality before you raise task difficulty.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Start new tasks with a 100% reinforcement schedule and a high-value item; thin the schedule only after the learner responds accurately three sessions in a row.
02At a glance
03Original abstract
BACKGROUND: Autism spectrum disorders (ASDs) can be conceptualized as disorders of learning, however there have been few experimental studies taking this perspective. METHODS: We examined the probabilistic reinforcement learning performance of 28 adults with ASDs and 30 typically developing adults on a task requiring learning relationships between three stimulus pairs consisting of Japanese characters with feedback that was valid with different probabilities (80%, 70%, and 60%). Both univariate and Bayesian state-space data analytic methods were employed. Hypotheses were based on the extant literature as well as on neurobiological and computational models of reinforcement learning. RESULTS: Both groups learned the task after training. However, there were group differences in early learning in the first task block where individuals with ASDs acquired the most frequently accurately reinforced stimulus pair (80%) comparably to typically developing individuals; exhibited poorer acquisition of the less frequently reinforced 70% pair as assessed by state-space learning curves; and outperformed typically developing individuals on the near chance (60%) pair. Individuals with ASDs also demonstrated deficits in using positive feedback to exploit rewarded choices. CONCLUSIONS: Results support the contention that individuals with ASDs are slower learners. Based on neurobiology and on the results of computational modeling, one interpretation of this pattern of findings is that impairments are related to deficits in flexible updating of reinforcement history as mediated by the orbito-frontal cortex, with spared functioning of the basal ganglia. This hypothesis about the pathophysiology of learning in ASDs can be tested using functional magnetic resonance imaging.
Autism research : official journal of the International Society for Autism Research, 2011 · doi:10.1002/aur.177