Poster #190 - EEG correlates of social engagement during naturalistic parent-child interaction in typical development and ASD

• In Event: 3-053 - Poster Session 10
  In Poster Session: PS 10 Section - Social Cognition

Integrative Statement

Language and social development occur in contexts of social engagement, especially between a child and parent. Atypical social engagement is also a key diagnostic criterion for Autism Spectrum Disorder (ASD). Previous studies have investigated the neural correlates of social engagement in hopes of improving identification and treatment of developmental disorders; however, most studies use artificial experimental conditions rather than true-to-life interaction. The goal of this pilot study was to investigate the neural correlates of social engagement during a live, naturalistic parent-child interaction. Based on previous literature, we predicted that during moments of social engagement as compared with non-social engagement, children would demonstrate increased EEG Theta band power (representing greater social attention and cognitive engagement) and suppression of Alpha and Mu power (reflecting more general attention and motor processing) (Jones et al., 2015; Orekhova et al., 1999; St. John et al., 2016); these effects were expected to be attenuated in ASD.
Six children, 3 typically-developing (TD 1 female; age 29-36 months) and 3 diagnosed with ASD (2 female; age 32-40 months) have participated to date. EEG was recorded using 32 active Ag-AgCl scalp electrodes (Biosemi ActiveTwo). The parent-child dyad sat next to each other at a table and engaged in contexts designed to elicit frequent social engagement (e.g., playing with a puzzle together) or nonsocial engagement (e.g., watching a nonsocial movie together). Sessions were videotaped and microcoded offline (using a coding system based on Adamson, Bakeman, & Deckner, 2014) to identify moments of social engagement (child engaged exclusively with their parent or in an activity jointly with their parent) and nonsocial engagement (child was engaged with an object/movie without interaction) within each context. These behavioral codes for social vs. nonsocial engagement conditions were time-locked to the EEG data, and EEG data for the two conditions were extracted. Moments of off-task/unengaged behavior were not used for these analyses.
The EEGLab/ERPLab moving window function was used to reject data with artifact (e.g., eye blinks, head/body motion). Accuracy of artifact rejection was visually confirmed for each subject. Each subject had > 30 clean 1s epochs. Fast Fourier transforms were computed using a window of 1 second with 50% overlap and application of a Hanning window. Power was calculated by averaging across windows in the bands of interest: Theta (4-6 Hz, frontal regions), Alpha (6-9 Hz, parietal regions), and Mu (6-9 Hz, central regions).
Figure 1 shows log-transformed power for each participant and condition. Figure 2 shows the percent difference between social and nonsocial conditions for each participant. Initial results show a pattern of increased Theta in the social vs. nonsocial engagement conditions for all TD children, but only one child with ASD. Patterns of Alpha and Mu suppression are variable across children. Our final sample size (N=24; 12/group) will allow systematic analysis of how group (TD vs. ASD) and context (social vs. nonsocial engagement) relate to EEG power. Group differences or group*condition interactions could indicate underlying early differences in the brain in ASD that could provide insights for theory, diagnosis, and treatment monitoring.

Authors

• Brittany L. Manning, Northwestern University
  Presenting Author

• Lauren H Hampton, Northwestern University
  Non-Presenting Author

• Megan Y. Roberts, Northwestern University
  Non-Presenting Author

• Elizabeth S. Norton, Northwestern University
  Non-Presenting Author