Answers are grounded in this paper and cite their sources.
Investigates word learning differences between autistic and neurotypical children, suggesting unique learning environments.
PURPOSE: For neurotypical children, accurate generalisation of newly learnt object names is facilitated by 'desirable difficulty' during initial referent selection; holding colour constant across competitors strengthens encoding of the intended referent's category-defining shape. We investigated whether variability in colour at referent selection similarly influences novel label generalisation in vocabulary-delayed autistic children - a population potentially characterised by differences in shape encoding. METHODS: Thirty-two autistic (M age = 116.63 months) and 32 neurotypical (M age = 56.72 months) children matched on receptive vocabulary (autistic M age equivalent = 67.78 months; neurotypical M age equivalent = 68.25 months) used a touchscreen computer to disambiguate new noun-object mappings with objects of the same or different colours. Children were then tested on their retention and shape-based generalisation after a 5-minute delay. RESULTS: Neurotypical children's generalisation was significantly more accurate - and only above-chance - in the same colour condition. Conversely, autistic children displayed similar above-chance generalisation accuracy in both the same and different colours conditions. For referent selection and retention, children in both groups achieved above-chance accuracy, with no significant differences between conditions. Finally, response time at referent selection, retention, and generalisation did not significantly differ between populations or conditions. CONCLUSION: Unlike neurotypical children, autistic children were not similarly disadvantaged by variability in object colour at referent selection, suggesting robust encoding of shape across perceptually varied learning conditions. Crucially, these findings demonstrate that beneficial and non-beneficial word learning environments may differ for autistic and neurotypical children.
No takes yet. Share an insight, caveat, or question.
Lund et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: