Altered Neuromagnetic Evoked Responses and Neural Synchrony Related to Language in Autism Spectrum Disorder

Background:

Autism Spectrum Disorder (ASD) is characterized by atypical social and language functioning. Atypical electrophysiological signatures of auditory processing in ASD may provide evidence informing cascading impairment of higher-order social and language functioning (Brennan et al., 2016).

Objectives:  Relationships between evoked response, synchrony, phonological processing (PP) and social communication were investigated.

Methods:

Eleven ASD (Age: M = 8.8; SD= 0.9) and 10 neurotypical (NT) children (Age: M = 9.4; SD = 1.4) underwent magnetoencephalography (MEG) at rest and during an oddball paradigm, which consisted of presentation of plausible (S1) and implausible syllables (S2). M100, M200, M250, and M300 amplitudes and latencies were quantified for evoked responses (ERP). An amplitude difference score (S1-S2) was calculated to reflect auditory sensitivity to novel language. It was hypothesized that decreased S2 sensitivity (smaller S1-S2 scores) would be associated with better PP and communication. Synchronization was quantified by calculating coherence between cortical sites. Kendall Tau correlations were computed to examine relationships between ERP, coherence, PP, and communication skills.

Results:

There were no group differences in amplitude, latency, sensitivity or PP (p>.05). NT had significantly higher communication skills, per parent report (p<.001). ERP: Higher communication skills were associated with longer S1 M300 latencies in ASD (p=.02) and shorter S1 M200 latencies in NT (p=.01). Significant associations between ERPs and PP were found at different latencies. In ASD, longer S2 M250 latencies (p=.02) and higher S1 M250 amps (p=.02) were associated with higher PP. Poorer M300 sensitivity was associated with higher PP (p=.04). In NT, longer S2 M100 latencies were associated with higher PP (p=.01). Coherence: In ASD, decreased interhemispheric angular to inferior temporal coherence was associated with larger S1 amplitudes at M250. In NT, higher intra- and interhemispheric cingulo-fronto-temporo-parietal coherence was associated with shorter S1 M200 latencies. Lower intra- and interhemispheric left angular to frontal region coherence was associated with longer S2 M100 latencies.

Conclusions:

There were no significant group differences in PP; however, there were significant group differences in parent reported communication skills. Despite a lack of significant PP group differences, differential electrophysiological profiles emerged. In NT, better PP was associated with longer M100 latencies (a latency associated with exogenous attributes of auditory stimuli) for implausible words and better communication was associated with shorter M200 latencies (latency associated with endogenous attributes of auditory stimuli) in response to plausible words. In ASD, a different profile emerged. Longer latencies and higher amplitudes at a latency associated with auditory processing of endogenous information was associated with better PP. Better PP was associated with poorer M300 gating, a latency associated with responding to deviant sounds in an oddball paradigm. Additionally, better communication was associated with longer M300 latencies. ERP data at latencies linked with endogenous auditory processing were associated with more global increased connectivity in NT and increased angulo-temporo connectivity in ASD. Finally, increased angulo-frontal connectivity was associated with longer latencies at exogoneous auditory processing latencies in NT only. Findings suggest differential electrophysiological profiles supported by cortical connectivity and ERPs in ASD and NT associated with PP and communication.