Learning and generalization of repetition-based rules in autism.

Researchers tested how well autistic teenagers can learn and apply patterns compared to non-autistic peers. Both groups could successfully learn visual patterns and apply them to new situations. However, autistic teens appeared to use their working memory (short-term memory) more when learning patterns, suggesting they may use different thinking strategies. The type of images used (faces vs. shapes) and pattern complexity affected how well both groups performed.

This study examined rule learning abilities in high-functioning autistic adolescents compared to typically developing peers. Participants were tested on their ability to detect and generalize repetition-based rules from visual sequences using three types of stimuli: simple shapes, inverted faces, and upright faces. Both groups successfully learned and generalized rules to new stimuli, with performance influenced by stimulus type (social vs. non-social) and rule complexity. A key difference emerged in the strategies used: autistic adolescents showed an association between rule learning and visual working memory, while typically developing peers did not, suggesting autistic individuals may rely on alternative cognitive strategies involving working memory resources.

Results suggest that rule learning abilities are preserved in autism but may involve different cognitive strategies. This has implications for educational interventions, suggesting that approaches supporting visual working memory may be beneficial. The finding that stimulus type affects performance indicates the importance of considering context in learning environments.

Sample size not reported, limiting assessment of statistical power. Study type unclear. Focus on high-functioning adolescents limits generalizability to broader autism spectrum. No information provided about potential confounding variables or control for other cognitive abilities.

Rule Learning (RL) allows us to extract and generalize high-order rules from a sequence of elements. Despite the critical role of RL in the acquisition of linguistic and social abilities, no study has investigated RL processes in Autism Spectrum Disorder (ASD). Here, we investigated RL in high-functioning autistic adolescents with ASD, examining whether their ability to extract and generalize rules from a sequence of visual elements is affected by the social vs. non-social nature of the stimulus and by visual working memory (WM). Using a forced-choice paradigm, ASD adolescents and typically developing (TD) peers were tested for their ability to detect and generalize high-order, repetition-based rules from visual sequences of simple non-social stimuli (shapes), complex non-social stimuli (inverted faces), and social stimuli (upright face).

Both ASD and TD adolescents were able to generalize the rule they had learned to new stimuli, and their ability was modulated by the social nature of the stimuli and the complexity of the rule. Moreover, an association between RL and WM was found in the ASD, but not TD group, suggesting that ASD might have used additional or alternative strategies that relied on visual WM resources.