Characterizing the Cortical Basis of Motor Impairments in ASD Using EEG

Background:  

Atypical visual motion perception is a feature of autism spectrum disorders (ASD) and may contribute to the abnormal development of motor milestones observed in this population. The EEG Rolandic mu rhythm, related to motor cortex function, has reduced suppression during action observation in ASD. Individuals with ASD also exhibit reduced global alpha (Berger) power; the alpha rhythm shares a frequency range, but not topography, with the mu rhythm. Exploring the relationship between mu and alpha rhythms and motor performance may help to better characterize the cortical basis of motor deficits seen in ASD.

Objectives:  

The present study investigated relative power at each frequency in the mu and alpha band (8-12 Hz) in children with and without ASD during a visual motion task. Differences in motor performance between the two groups were assessed using the Movement Assessment Battery for Children (M-ABC) and the Beery-Buktenica Developmental Test of Visual-Motor Integration (Beery VMI). Power at each frequency was correlated with performance on these motor tests.

Methods:  

Participants were 11 females and 18 males (n=29) between 7-8 years old, all but 2 were right handed. The ASD group’s mean age was 95.2 months (+/- 6.3) and mean FSIQ was 100.1 (+/- 12.1). The control group’s mean age was 95.1 months (+/- 7.6) and mean FSIQ was 107.6 (+/- 23.9). ASD diagnosis was based on ADOS and ADI-R. EEG was recorded as subjects watched sand fall through an hourglass. Time series data were transformed via FFT with 1 Hz resolution and standard electrodes were considered (C3, C4, P3, P4, F3, F4, O1, O2).

Results:  

The ASD group had lower standard scores on Beery VMI (t=3.22, p<0.01) and M-ABC Total (t=3.88, p<0.001). At 9 Hz, the ASD group had lower relative power at C3 (t=1.82, p<0.05) and F3 (t=2.37, p<0.05). At 10 Hz, the ASD group had lower relative power at F3 (t=2.43, p<0.05) only. Combining ASD and TYP groups, the Beery VMI scores correlated with 9 Hz relative power at C3 (r=.67, p<0.001), F3 (r=.68, p<0.001), and F4 (r=.64, p<0.001). No correlations were observed at 8 Hz, 10 Hz, 11 Hz, or 12 Hz. Separately, the ASD group revealed no correlation between performance and 9 Hz power, while the TYP group showed strong positive correlations at C3 (r=.70, p<0.01), F3 (r=.67, p<0.01), and F4 (r=.65, p<0.05).

Conclusions:  

Children with ASD scored lower on both motor tests, highlighting a motor execution deficit in the autism group. The ASD group exhibited lower 9 Hz power in left central (C3) and left frontal (F3) scalp regions and lower 10 Hz power in a left frontal (F3) region. No group differences were observed at 8 Hz, 11 Hz, or 12 Hz, at occipital or parietal electrodes (O1, O2, P3, P4), or in the right hemisphere (C4, F4). The group differences in 9 Hz power at C3 and F3 and the strong correlations with a visuo-motor integration assessment suggest that EEG power at this frequency may serve as an index of visual motor impairment in 7-8 year old children.