Name: Vicky Yu
Master's Thesis Defense Meeting
Date: Thursday, April 17th, 2025
Time: 2:15 pm - 3:15 pm
Location: JS Coon 148

Thesis Chair/Advisor:
Agata Rozga, Ph.D. (Georgia Tech)

Thesis Committee Members:
Woon Ju Park, Ph.D. (Georgia Tech)
Christopher Stanzione, Ph.D. (Georgia Tech)

Title: Longitudinal Trajectories of Early Coordinated Communication in Toddlers with Autism: Predicting Language Outcomes at Age Three

Abstract: Early coordinated communication behaviors—such as gestures, gaze, and vocalizations—are known predictors of language development in children with Autism Spectrum Disorder (ASD), yet few studies have examined how these behaviors unfold longitudinally across early development greater than two timepoints and how their trajectories predict later language skills. This study examined whether longitudinal growth in early coordinated communication, measured through linear, quadratic, and cubic components of Communication Complexity Scale (CCS) scores, and gesture frequency at the first visit predicted language outcomes at age three in toddlers with ASD. Using data from 20 children across at least four time points, I modeled communication growth and assessed language acquisition via the Mullen Scales of Early Learning. Neither growth trajectories nor gesture use alone significantly predicted language outcomes. However, in models including interaction terms, all predictors—linear, quadratic, and cubic growth, as well as gesture frequency—emerged as significant, suggesting suppression effects or overlapping variance among predictors. These findings highlight the importance of modeling developmental processes in context rather than isolation. Static, single-timepoint measures may obscure important patterns of growth that unfold over time. Despite limitations including small sample size and measurement ceilings, this study underscores the need for dynamic, longitudinal approaches in understanding early communication and informing individualized interventions in ASD.