Related Experiment Video
Updated: Jun 26, 2026
Probing the Brain in Autism Using fMRI and Diffusion Tensor Imaging
Published on: September 12, 2011
Functional brain network organization during the 40-Hz auditory steady-state response in children with and without
Mai Yasumoto1, Yoshiaki Miyagishi2, Tetsu Hirosawa1
1Department of Child and Adolescent Psychiatry, Graduate School of Medical Science, Kanazawa University, Kanazawa, Japan.
Introduction:
The auditory steady-state response (ASSR) at 40 Hz provides a well-established probe of gamma-band neural synchronization and is thought to reflect excitation-inhibition balance in cortical microcircuits-a mechanism implicated in autism spectrum disorder (ASD). Previous studies focusing on regional ASSR measures have yielded inconsistent findings for ASD, suggesting that system-level approaches may provide complementary information about ASD-related neurophysiological alterations.
Methods:
We examined source-level functional brain networks during the 40-Hz ASSR in children with ASD (n = 19) and typically developing (TD) children (n = 34) aged 5-8 years using magnetoencephalography (MEG). Functional connectivity was estimated using the phase lag index (PLI) within the 30-50 Hz band, and graph-theoretical metrics-clustering coefficient (CC), characteristic path length (PL), and small-worldness (SW)-were computed from binary undirected networks. Group differences were assessed using multiple linear regression adjusting for age, sex, and cognitive ability. Associations between network measures and autistic traits, indexed by the Social Responsiveness Scale (SRS), were examined using a regression model including diagnosis and the interaction of diagnosis × graph measure.
Results:
The experimental paradigm elicited ASSR-related cortical activity in both groups. A significant main effect of diagnosis was observed for characteristic PL, with children with ASD exhibiting shorter PL than TD children, indicating altered network topology during 40-Hz ASSR (p = 0.0023). No group differences were found for CC or SW. In the secondary analysis, shorter PL was associated with higher SRS total T-scores in the omnibus model (p = 0.027). The diagnosis × PL interaction was not significant, indicating that this association did not differ detectably between ASD and TD groups.
Discussion:
These findings suggest that graph-theoretical analysis of functional brain networks during the 40-Hz ASSR may provide complementary information about ASD-related neural organization beyond conventional ASSR measures. Shorter characteristic PL in ASD, together with its association with autistic-trait severity across the full sample, points to altered large-scale network configuration during externally driven gamma-band synchronization. The results support the value of examining functional brain network topology during the 40-Hz ASSR in ASD.
