Autism spectrum disorder (ASD) is associated with atypical speech processing early in life, which can have negative cascading effects on language development, one of the strongest predictors of long-term outcomes. While the frequency-following response (FFR)—a sound-evoked potential that reflects synchronous neural activity along the auditory pathway—has revealed speech processing differences in children and adolescents on the autism spectrum, what has yet to be established is the extent to which such differences are present during the ASD prodromal period. Investigating speech-evoked FFRs in infant siblings of children with ASD (infants at elevated familial likelihood for autism [EL infants]) has the potential to reveal the earliest time window when neural function related to speech processing may begin to diverge. In this talk, I will present results from a prospective, longitudinal study characterizing the developmental time course and the predictive value of the FFR to speech sounds in EL infants compared to infants with a typical likelihood for developing ASD. Findings reveal that the FFR may be sensitive to subtle speech processing differences in EL infants within the first year of life, well before the emergence of behavioral precursors. This study will set the stage for a broader discussion of the utility of the FFR for characterizing speech processing differences and in predicting distal language and social communication outcomes across the range of heterogeneity associated with the autism phenotype.

Frequency-following responses (FFRs) to speech stimuli are non-invasive and easily deployable measurements of speech encoding integrity in the auditory system that show great potential as a biomarker for many pathological conditions. But because fundamental questions remain with respect to how the activity and modulation of underlying neural circuits affect speech FFRs, interpreting and attributing changes in FFRs to particular circuit elements is challenging. As a first step towards gaining more biological insight into these issues, our lab has been involved in a collaborative effort to determine the extent to which speech FFRs and their response properties are conserved in an animal model, setting the stage for future mechanistic experiments. In this talk, I will present results from the guinea pig, a rodent with excellent low-frequency hearing, that demonstrate that FFRs to speech sounds show remarkable similarities with those recorded in humans and monkeys. I will discuss how stimulus fundamental frequency, stimulus statistics, arousal state, and manipulation of auditory cortical activity affect FFR amplitudes and fidelity in this species. Using intracranial translaminar recordings, I will discuss our initial efforts to map scalp recordings to neural ensemble activity. Finally, I will present early studies on non-stimulus-following intrinsic oscillatory activity in the auditory cortex, and discuss what these may reveal about underlying circuits.