Magnetoencephalographic (MEG) studies examining brain activity in children with autism spectrum disorder (ASD), including published studies from this R01's previous funding period, have identified functional markers that discriminate children with ASD from typically developing (TD) controls. Of note, the ~100 ms (M100) component of the superior temporal gyrus (STG) auditory evoked response to simple tones demonstrates a short but profound delay (~10ms) in children with ASD, especially in the right hemisphere, and especially for tones with frequencies dominant in human speech (300-500Hz). Two dominant hypotheses have emerged to account for M100 delays: (1) conduction velocity of the thalamo-cortical acoustic radiations projections, with immature white matter (WM) in ASD, and (2) abnormal STG synaptic transmission, with diminished gamma- band phase synchrony in ASD. Data from the prior funding period provides support for both hypotheses and motivates a stratification model, where both factors contribute (although not necessarily equally) to M100 delays. The contribution of these mechanisms to M100 latency can be independently examined, using diffusion MRI to assess the acoustic radiations (the thalamo-cortical projections of the auditory pathway WM) (Specific Aim 1) and using MEG measures of neural oscillatory activity to probe integrity (or lack thereof, labeled as 'oscillopathy') of local circuitry and by implication synaptic transmission (Specific Aim 2). Using the MEG and diffusion MRI measures the ASD cohort will be divided into 4 subtypes, based on a median-split in white-matter deficit (WM+/-) and oscillopathy (Osc+/-), thus identifying a 'dominant deficit', or combination thereof, for each subject: WM-/Osc-, WM-/Osc+, WM+/Osc-, WM+/Osc+. The clinical significance of these STG auditory system deficits will be examined by evaluating the ability of the WM/oscillopathy subtypes to reduce heterogeneity in ASD via assessment of between-group phenotypic differences and within-group reduction in phenotype variance as well as identification of associations between clinical symptoms (e.g., language function, phonological processing, and general cognitive ability) within specific subtypes (Specific Aim 3). Thus, we will determine the extent to which electrophysiological measures such as the auditory M100 latency, as well as subordinate measures of WM microstructure and oscillatory activity, can be considered as biomarkers for use in the characterization and stratification of children with ASD. Future implications of such stratification can be anticipated in differential patient management strategies as well as selection for, and monitoring the activity of, targeted pharmaceuticals. Finally, this R01 renewal will determine whether the auditory evoked response abnormalities in ASD are specific to the auditory domain or reflect a more widespread phenomenon, with latency delays identified also in primary visual and somatosensory cortex (Specific Aim 4).