UCLA

Autism spectrum disorder (ASD) is a highly prevalent neurodevelopmental disorder characterized by social communicative deficits and restricted, repetitive behaviors. Early interventions are likely to be more effective for treating disorders that impact language development such as ASD, highlighting the urgency of being able to identify precursors of the disorder very early in life. Prior research has demonstrated that altered developmental trajectories in brain structure, function, and connectivity are hallmark features of ASD, but little is known about the developmental origin of these atypicalities. Although overt behavioral symptoms indicative of ASD begin to emerge in the second year of life, a recent and growing body of work has shown that brain-based markers can be detected well before the first birthday. This dissertation aims to elucidate the development of neural networks subserving language acquisition from a multimodal perspective by relating early brain structure and function to later behavioral outcome in infants at high familial risk for ASD. Chapter 1 provides a general introduction to the research outlined in the following chapters. Chapter 2 describes a study which examined whether very early differences in structural connectivity of canonical language pathways are related to later outcome in infants at high and low risk for developing ASD. As early as 1.5 months of age, infants at high familial risk for ASD showed altered white matter integrity and lateralization of language tracts. In addition, structural metrics were associated with later language outcome at 18 months of age as well as ASD symptomatology at 36 months of age. Chapter 3 presents data charting longitudinal changes in the functional connectivity of language-related networks to investigate differences associated with ASD risk from 1.5 to 9 months of age. This study takes a two-pronged approach using (1) seed-based methods to examine primary and secondary auditory cortex connectivity within each time point, and (2) network-based analyses to model longitudinal changes in functional connectivity across time. As early as 1.5 months of age, at-risk infants already showed differences in networks underlying auditory-motor integration. By 9 months of age, high risk infants showed hyperconnectiivty between primary auditory cortex and sensory regions, whereas low risk infants showed robust connectivity with higher-order cortical regions. Over time, low risk infants showed widespread changes that followed a normative developmental profile with increasing long range and decreasing short range connectivity. By contrast, high risk infants showed limited changes over time with more static developmental profiles. Lastly, Chapter 4 describes a study which explored whether the neural basis of implicit language learning is altered in infants at high risk for ASD. Study findings showed that high risk infants exhibited less sensitivity to speech cues that are critical for language acquisition. In addition, learning-related neural activation was associated with better language outcome and less severe ASD symptomatology at 36 months. Low risk infants also displayed increasing functional connectivity between bilateral primary auditory cortex and the right anterior insula, suggesting that language stimuli may be intrinsically more salient for low risk compared to high risk infants. Taken together, findings from these studies indicate that the multimodal integration of early brain-based measures and later behavioral outcome can shed insights into mechanisms underlying altered developmental trajectories associated with ASD risk. As such, these studies enhance our understanding of the early structural and functional architecture of the developing brain, which has implications for the development of early interventions which may possibly prevent the onset of full ASD symptomatology and steer development toward more optimal developmental trajectories.