Aberrant White Matter Development Underlies Atypical Visual Orienting At 6 Months in Prodromal Autism

Background: Many of the early behavioral markers associated with ASD implicate developmental impairments in flexibly allocating attentional resources to salient or biologically relevant information in the environment (e.g., orienting to name, responding to bids for joint attention, spontaneous gaze to faces and making eye contact). We reasoned that a domain-general deficit in selective visual orienting during a developmental period that requires flexible and efficient selective visual attention to support social information processing could constrain typical experience-dependent specialization with social information, and thereby contribute to the emergence of behavioral markers related to ASD symptoms.

Objectives: In the current study, we sought to examine whether infants who go on to develop ASD symptoms show different patterns of visual orienting at 6 months of age when compared to low-risk infants and genetic high-risk infants who do not develop ASD symptoms.  If so, does the organizational integrity of white matter explain atypical visual orienting? 

Methods: We used risk status and ADOS scores at 24 months to form three discrete groups; genetic low-risk or typically developing infants (hereafter LR, n = 37), genetic high-risk infants who do not meet ADOS criteria for an ASD at 24 months (hereafter HR-, n = 28), and infants who meet ADOS classification for an ASD at 24 months (hereafter ASD, n = 14).  At 6 months of age, visual orienting data was extracted from a gap/overlap paradigm and imaging data was extracted from a Diffusion Tensor Imaging sequence.  The three groups did not significantly differ in age (p = 0.297), nonverbal developmental level (p = 0.789), or number of valid gap (p = 0.306) or overlap trials (p = 0.148) completed. 

Results: A disease specific pattern emerged for the overlap latencies such that the ASD group showed longer latencies than both the HR- (p = 0.013) and the LR group (p = 0.038), who were statistically equivalent.  The data revealed a disease continuum model in gap latencies such that the ASD group showed significantly longer latencies than the LR group, but the HR- group did not significantly differ from the LR or ASD groups and their latencies fell at an intermediate position between LR and ASD (e.g., ASD > HR- > LR).  We found that the organizational integrity of white matter (RD) in the splenium moderated the association between risk status and visual orienting (overlap latencies, as per the disease specific finding above).  The results revealed a disordinal interaction, F(2, 65) = 3.35, p = .041, such that latencies in the overlap condition were not related to white matter integrity in the ASD group, whereas the association between overlap latencies and splenium RD among LR infants was quite strong.

Conclusions: These findings suggest that patterns of visual orienting at 6 months differentiate infants who go on to develop characteristics of ASD at 24 months from LR infants and HR infants who do not develop ASD symptoms, and that microstructural integrity of the splenium is putatively involved in this process.