Cross-modal attention-switching is impaired in autism spectrum disorders.

Reed et al. (2012) asked kids to switch attention between sights and sounds. They tested three groups: autistic tweens, kids with intellectual disability, and neurotypical peers. Each child sat at a screen that gave both visual and auditory cues. The task was simple: press when the cue changes, no matter the modality.

Autistic children were much slower when the shift crossed senses. Visual-to-visual shifts were hard, but visual-to-auditory shifts were even harder. Both comparison groups handled the switches with ease. The bigger gap on cross-modal trials tells us the deficit is about linking senses, not just shifting focus.

May et al. (2013) extend the finding into real life. They show that weak attention-switching predicts lower math scores in autistic students, even after IQ is matched. The same skill measured in the lab now shows up in report cards.

Katagiri et al. (2013) replicate the trouble with switching, but in a visual-only task. Their Asperger's group stalled when moving from small details to the big picture. Together the papers say: switching rules are hard, whether you jump senses or jump levels.

Hatta et al. (2019) seem to disagree. They found attention-switching problems in kids with somatic symptom disorder, yet call it a 'positive' result. The clash clears up once you see they studied a different diagnosis. The task was similar; the kids were not.

If you run matching-to-sample, discrete-trial, or group instruction, slow the pace when you hop modalities. Say the instruction while you point to the text, then wait. Give a clear 'look' or 'listen' cue before each shift. These tiny pauses can prevent errors that look like non-compliance but are really cross-modal overload.