Everything has Its Time: Narrow Temporal Windows are Associated with High Levels of Autistic Traits Via Weaknesses in Multisensory Integration.

Kawakami et al. (2020) asked adults with no autism diagnosis to watch short flashes and beeps. The team slowly moved the sound earlier or later until each person said the pair no longer felt simultaneous.

They also gave everyone a short quiz about social and sensory habits that hint at autistic traits. The goal was to see if people who scored higher on these traits also had a tighter 'window' for pairing sight and sound.

Adults who reported more autistic-style traits needed the flash and beep to land almost exactly together. If the gap was wider than a tiny slice of time, they stopped calling it 'together.'

The same high-trait adults also showed weaker multisensory integration on a separate task. Tighter window, weaker merge — the two went hand in hand.

Zhou et al. (2021) used the same flash-beep setup but found NO link between window width and trait level. The difference: they tested university students with very low trait scores, while Sayaka sampled the broader community. The contradiction disappears when you see one study hunted for tiny shifts within a narrow range and the other looked across the full spectrum.

Johnston et al. (2017) showed adolescents diagnosed with ASD also have a narrow window, but only for speech clips, not for objects. Sayaka extends that downward into everyday adults who carry mild traits, hinting the timing quirk sits on a continuum.

Capio et al. (2013) found adults with ASD detect 17 ms visual lags better than controls. Sayaka’s high-trait adults show the same tightening, linking sub-clinical traits to clinical-level sensory precision.

If your client says the lip-sync on the tablet ‘looks wrong,’ believe them. A narrow window means even 100 ms delays can feel off and break attention. Use tight, well-synced media and check that your voice and gestures line up when you model skills. For higher-trait learners, give extra processing time between cues instead of piling on multisensory input all at once.