Examination of the prevalence of stimulus overselectivity in children with ASD.

Cox et al. (2015) asked a simple question: how many kids with autism still show stimulus overselectivity today? They tested 42 children diagnosed with ASD using matching-to-sample tasks. The team counted how many kids attended to only one cue when several were present.

Only 19% of the children showed overselectivity. That is about 1 in 5 kids, far below the old guess of 'most'. The authors think modern ABA programs may be shrinking the problem.

Vassos et al. (2016) seems to disagree. They found no difference in overselectivity between kids with autism, Down syndrome, and typical peers when everyone was matched for mental age. The two studies look opposite, but they tested different groups. R et al. sampled any child with an ASD diagnosis. V et al. only included kids who also had similar IQ scores. Cognitive level, not the autism label, may drive the effect.

Ploog (2010) reviewed forty years of work and warned that overselectivity is still 'prevalent' in autism. The new 19% figure updates that view. It hints that wide-use of tactics like multiple-exemplar training and stimulus control shaping is paying off.

McAleer et al. (2011) showed that overselectivity can be learned when typical adults train with overly complex stimuli. Taken together, the papers say: keep tasks simple, match developmental level, and you will see the problem less often.

Check your conditional discrimination programs this week. If 4 out of 5 kids no longer show overselectivity, your teaching may already be working. Still, probe for it—especially with learners who have lower cognitive scores. Start with simple two-cue combinations and add complexity only after the child masters each step. That small shift keeps stimulus control broad and learning fast.