The suggestion that movement disorders exhibited by people with central nervous system (CNS) dysfunction should be considered normal and therefore not rehabilitated is rejected from three standpoints: (1) the CNS does not always select the best movement patterns for optimal long term outcome, (2) there is literature demonstrating that rehabilitation enhances motor function and independence, and (3) there exists a capacity for motor recovery and motor skill learning following brain damage.

We investigated gait in newly diagnosed children with autism. From our previous study with 6- to 14-year-olds, we hypothesized that motor symptoms indicative of basal ganglia and cerebellar dysfunction would appear across the developmental trajectory of autism. Two groups were recruited: children with autism (eight males, three females; mean age 5y 10mo [SD 9mo]; range 4y 4mo–6y 9mo) and a comparison group of typically developing children (eight males, three females; mean age 5y 9mo [SD 1y 1mo]; range 4y 3mo–7y 2mo). The GAITRite Walkway was used to gather data from average gait and intra-walk measurements. Experienced physiotherapists analyzed gait qualitatively. Groups were matched according to age, height, weight, and IQ; although not statistically significant, IQ was lower in the group with autism. Spatiotemporal gait data for children with autism were compatible with findings from patients with cerebellar ataxia: specifically, greater difficulty walking along a straight line, and the coexistence of variable stride length and duration. Children with autism were also less coordinated and rated as more variable and inconsistent (i.e. reduced smoothness) relative to the comparison group. Postural abnormalities in the head and trunk suggest additional involvement of the fronto-striatal basal ganglia region. Abnormal gait features are stable across key developmental periods and are, therefore, promising for use in clinical screening for autism.

Autism and Asperger's disorder (AD) are neurodevelopmental conditions that affect cognitive and social-communicative function. Using a movement-related potential (MRP) paradigm, we investigated the clinical and neurobiological issue of ‘disorder separateness’ versus ‘disorder variance’ in autism and AD. This paradigm has been used to assess basal ganglia/supplementary motor functioning in Parkinson's disease. Three groups (high functioning autism [HFA]: 16 males, 1 female; mean age 12y 5mo [SD 4y 4mo]; AD: 11 males, 2 females; mean age 13y 5mo [SD 3y 8mo]; comparison group: 13 males, 8 females; mean age 13y 10mo [SD 3y 11mo]) completed a cued motor task during electroencephalogram recording of MRPs. The HFA group showed reduced peak amplitude at Cz, indicating less activity over the supplementary motor area during movement preparation. Although an overall significant between-group effect was found for early slope and peak amplitude, sub-analysis revealed that the group with AD did not differ significantly from either group. However, it is suggested that autism and AD may be dissociated on the basis of brain–behaviour correlations of IQ with specific neurobiological measures. The overlap between MRP traces for autism and Parkinson's disease suggests that the neurobiological wiring of motor functioning in autism may bypass the supplementary motor area/primary motor cortex pathway.