Brief Report: Suboptimal Auditory Localization in Autism Spectrum Disorder: Support for the Bayesian Account of Sensory Symptoms.

Bouck et al. (2016) asked the adults with autism and 18 matched adults to point where a brief tone came from.

On half the trials a valid visual cue lit up first, giving a strong hint of the sound spot.

The team used Bayesian math to see how well each group folded that prior cue into their final guess.

Typical adults used the visual cue almost perfectly, slashing their average error by a large share.

Autistic adults improved only a large share. Their brains acted as if the cue were weaker and less certain.

The pattern fits a Bayesian lens: people with autism give less weight to past information when judging space.

Kaufman et al. (2010) saw a similar timing issue. Their autistic adults under-reproduced long intervals, again showing poor use of prior duration cues.

Spates et al. (2013) found delayed ear-muscle reflexes in toddlers with autism. Both studies point to sluggish or under-weighted auditory priors across the lifespan.

Russo et al. (2009) looked at speech in noise. Autistic kids processed quiet speech only as well as typical kids processed noisy speech. All three papers line up: autism dulls the brain's use of context, whether it is spatial, timing, or background noise.

If your client seems lost when sounds shift rooms, do not jump to attention deficits. Their brain may not be using early cues you think are obvious. Give spatial warnings longer to sink in: repeat the cue, pair it with a visual point, and allow extra time before the real sound happens. These small tweaks honor the Bayesian gap the data keep showing.