Background: Language ability is one of the strongest predictors of prognosis and developmental course in Autism Spectrum Disorder (ASD). A range of language abilities occur in ASD and although many have delays in language it remains unclear why some children’s language continues to lag, while others do not. Abnormal anatomy and function of language-related regions has been found in ASD, however, how these differences relate to language development over time is undetermined.

Methods: This study examined longitudinal changes in language functions in children with ASD and investigated whether cortical language region anatomy was related to these changes in language. Eighteen boys with ASD, 2-8 years old were evaluated (Time 1) and re-examined about 3.5 years later (Time 2) at ages 7-10. MRIs were collected at Time 2 to evaluate gray matter volume of anterior (Pars Triangularis, PTR; pars opercularis, POP) and posterior (Planum Temporale, PT; Posterior Superior Temporal Gyrus, pSTG) language regions and the microstructure of the arcuate fasciculus.

Results: Eleven boys had relative decline in language functions (decline group) and 7 boys had no relative change in language (no change group). The no change group had larger PT and right PTR volume relative to the decline group. In addition, the right PTR was correlated with the language change score, with larger right PTR associated with less language decline. There was a trend for non-right-handers to have more language decline than right-handers.

Conclusions: Results suggest differences in cortical language anatomy may play a role in language development, with further studies warranted.

From the observation of different F-wave waveforms, we introduce a new method of differentiating these waveforms, by assigning each with an “F-wave waveform value”, which can be used in the clinic to evaluate the effects of rehabilitation. F-wave waveform values were determined by creating a window from minimum onset latency to maximum onset latency in measurable waveforms. We then calculated the correlation coefficient of each waveform, using Microsoft Excel, and identified F-waves as those with a correlation coefficient of greater than 0.9 or equal to 1.0. The number of different F-wave waveforms types was determined from the number of identified waveforms. We applied F-wave waveform values to evaluate neurophysiological change and the effects of rehabilitation following hemiplegia. In the future, F-wave waveform values should be considered as an important tool when assessing the effects of rehabilitation on impaired neurological responses.