Impaired Frontal Processing in 3- to 5-Year-Old Children with Autism and a Developmental Language Delay during a Mismatch Negativity Paradigm.

Background: The inferior frontal and superior temporal areas in the left hemisphere are crucial for language processing. Language abilities are highly variable in individuals with ASD, with difficulties that range from mild to severe impairments in pragmatics and/or social communication. Currently, language level is considered to be a continuous, rather than categorical, variable. Intriguingly, accumulating electrophysiological evidence suggests that deficits in the discrimination of rapid sound changes are associated with impaired speech processing in children with ASD. However, no previous magnetoencephalography (MEG) studies have focused on analysis of the Mismatch Field (MMF) source in young children with ASD in 3- to 5-year-old.

Objectives: The aim of this study was to investigate regional activity in the brain during a speech perception task in order to explain the phenotypic heterogeneity in language development among children with ASD.

Methods: Forty-six young TD children and 47 children with ASD participated in this study. we investigated the mismatch field (MMF) evoked by voice stimuli in 3- to 5-year-old typically developing (TD) children and children with autism spectrum disorder (ASD) using child-customized magnetoencephalography (MEG). A human voice pronouncing the syllable “ne” with a high falling tone was randomly presented as a rare deviant among frequent utterances of “ne” pronounced with a flat tone.

Results: A longer MMF latency in the left pars orbitalis in the children with ASD was associated with a lower performance in expressive language. Based on the results for the MMF amplitude, the children with ASD exhibited significantly decreased activation in the left superior temporal gyrus compared with the TD children in the 100 – 200 ms time window. If we classified the children with ASD according to the presence of a speech onset delay (ASD - SOD and ASD - NoSOD, respectively) and compared them with the TD children, both ASD groups exhibited decreased activation in the left superior temporal gyrus compared with the TD children in the 100 - 200 ms time window. In contrast, in the 200 - 350 ms time window, the ASD - SOD group exhibited increased activity in the left frontal cortex (i.e., pars orbitalis) compared with the other groups (Figure 1). For all children with ASD, there was a significant negative correlation between the MMF amplitude in the left pars orbitalis and language performance.

Conclusions: This investigation is the first to show a significant difference in two distinct MMF regions in ASD – SOD children compared with TD children; one region was independent of a speech onset delay (SOD), and the other region was dependent on SOD. The results from the first region (left superior temporal gyrus) suggested that reduced MMF amplitude in response to a change in the tone of a human voice may represent a biomarker for ASD regardless of the presence of SOD. The results from the second region (left pars orbitalis) suggested that enhanced and delayed activation in response to a change in the tone of a human voice reflects the developmental delay in language acquisition in young children with ASD.