Autism spectrum disorder (ASD) is associated with broad differences in perceptual, cognitive, and motor functioning. They are increasingly recognized as core features of the autistic phenotype. Increasing evidence suggests that these differences can reflect decreased functional and hemispheric specialization of cortical systems in autism. A robust behavioral marker of cerebral lateralization is handedness, which in neurotypical individuals reflects strong left hemisphere specialization for fine motor control. Atypical handedness and reduced lateralization are frequently reported in autism; however, findings have been inconsistent, partly due to reliance on self-report questionnaires rather than direct observation of motor behavior. Most earlier studies have focused on children, leaving open the question of whether altered lateralization persists into adulthood. Naturalistic visuomotor tasks that allow continuous measurement of hand use, spatial methods, and movement variability may provide a more sensitive assessment of functional specialization in autism.
The aim of this study was to characterize handedness and visuomotor strategies in autistic and non-autistic adults using a naturalistic LEGO-building task. It also determines whether autistic adults show reduced behavioral lateralization despite being classified as right-handed and whether this reduced specialization is linked to altered spatial methods, reduced motor efficiency, and elevated variability in movement patterns. It also assessed the relationship between handedness measures derived from questionnaires and those derived from observed motor behavior.
The study included 27 autistic adults and 27 non-autistic adults matched for gender and age. All participants were classified as right-handed based on writing preference and had normal or corrected-to-normal vision. Autism diagnoses were confirmed using the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2). Participants completed the Autism Spectrum Quotient, Edinburgh Handedness Inventory (EHI), and Test of Nonverbal Intelligence 4 before performing a standardized LEGO-building task. Participants reconstructed five LEGO models by blocks distributed symmetrically in a construction board, which allows for free spatial choice and naturalistic bimanual interaction during the task. Movements were video recorded and coded by independent raters. Key outcome measures involved proportion of right-hand grasps, ipsilateral and contralateral movements, spatial block utilization (left–right and near–far), model completion time, and movement trajectory variability. Trajectory similarity in participants was quantified by using dynamic time warping (DTW), which captures differences in the sequence and spatial pattern of block selection. Statistical analyses compared group performance and examined the correlation between handedness measures and movement variability.
Despite all participants being classified as right-handed, autistic adults showed significantly reduced right-hand dominance compared with non-autistic controls, both in EHI scores and in observed hand use during the task. Importantly, questionnaire-based handedness and behavioral hand use were not correlated, indicating that these measures capture distinct aspects of lateralization. Analysis of the first grasp in each trial revealed no group differences, with both groups predominantly initiating movements with the right hand. This finding suggests that reduced lateralization in autism emerges during the ongoing, dynamic demands of the task rather than at movement onset. Autistic participants also exhibited less asymmetry in contralateral movements, using both hands more equally across the workspace, whereas non-autistic participants showed stronger right-hand specialization.
Spatial analyses showed that both groups preferentially used blocks from the right side of the board, with no group difference. However, autistic participants demonstrated a significant bias toward using blocks located closer to their body, unlike non-autistic participants, who showed no near–far preference. Autistic participants also took longer to complete the LEGO models, indicating reduced visuomotor efficiency. DTW analyses revealed greater within-group variability in movement trajectories among autistic participants, reflecting more idiosyncratic and less consistent action sequences. Moreover, weaker behavioral right-hand preference was associated with greater movement variability, whereas EHI scores were not predictive of movement pattern variability.
In conclusion, autistic adults exhibit reduced behavioral specialization of hand use and altered visuomotor strategies during a naturalistic object-manipulation task, even when classified as right-handed. Reduced right-hand dominance, more symmetrical contralateral movement, preference for nearer objects, slower task performance, and increased movement variability collectively support the hypothesis of weaker hemispheric specialization in autism. The dissociation between questionnaire-based and behavioral measures of handedness underscores the limitations of self-report tools and highlights the importance of naturalistic, performance-based assessments. By focusing on adults, these findings suggest that reduced lateralization in autism reflects a persistent difference in neural organization instead of developmental delay. Overall, the results support reduced specialization accounts of autism and emphasize motor variability as a key dimension of heterogeneity within the autism spectrum, with important implications for assessment and individualized intervention.
Reference: Fewster E, Hadad BS, Freud E. Reduced Hand Specialization and Idiosyncratic Visuomotor Strategies in Autism During Naturalistic Object Manipulation. Autism Res. 2026. doi:10.1002/aur.70180
