Daniel A. Ignacio ^1,2* Talin Babikian ^1,3 Emily L. Dennis ⁴ Kevin C. Bickart ^1,5 Meeryo Choe ^1,6 Aliyah R. Snyder ^3,7 Anne Brown ^1,6 Christopher C. Giza ^1,6 Robert F. Asarnow ²

• ¹ University of California, Los Angeles, Steve Tisch BrainSPORT Program, Los Angeles, CA, United States
• ² Department of Psychiatry and Biobehavioral Sciences, Semel Institute for Neuroscience and Human Behavior, University of California, Los Angeles, Los Angeles, California, United States
• ³ Department of Psychiatry and Biobehavioral Sciences, UCLA School of Medicine, Los Angeles, CA, United States
• ⁴ Department of Neurology, School of Medicine, The University of Utah, Salt Lake City, Utah, United States
• ⁵ Department of Neurology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, California, United States
• ⁶ Division of Pediatric Neurology, UCLA Mattel Children's Hospital, Los Angeles, California, United States
• ⁷ Department of Clinical and Health Psychology, College of Public Health and Health Professions, University of Florida, Gainesville, Florida, United States

INTRODUCTION Neuroimaging has expanded our understanding of pediatric brain disorders in which white matter organization and connectivity are crucial to functioning. Paralleling the known pathobiology of many neurodevelopmental disorders, traumatic brain injury (TBI) in childhood can alter trajectories of brain development. Specifically, diffusion tensor imaging (DTI) studies have demonstrated white matter (WM) abnormalities in TBI to identify microstructural disruptions that may underlie atypical neurodevelopment. The neurocognitive correlates of these previous findings will be explored here. METHOD Indicators of WM organization were collected in 44 pediatric patients with moderate/severe TBI and 76 controls over time: T1 (8-20 weeks post-injury) and T2 (54-96 weeks). Our previous work identified two TBI subgroups based on information processing differences: a TBI subgroup with slower interhemispheric transfer times (IHTT) of visual information than controls and a TBI subgroup with comparable IHTT. We extend this prior work by evaluating cognitive trajectories associated with divergent WM structure post-injury in slow and normal IHTT TBI subgroups. RESULTS At T1, both TBI subgroups performed significantly worse than controls on a norm-referenced working memory index (WMI), but only the Normal IHTT TBI subgroup significantly improved over the 12-month follow-up period (p = .014) to match controls (p = .119). In contrast, the Slow IHTT TBI subgroup did not show any recovery in working memory performance over time and performed more poorly than the control group (p < .001) at T2. Improvement in one of the two WMI subtests was associated with DTI indicators of WM disorganization in CC tracts to the precentral, postcentral, frontal, and parietal cortices. IHTT and WM mean diffusivity predicted 79% of the variance in cognitive recovery from T1 to T2, even when accounting for other known predictors of TBI recovery. DISCUSSION Over one year post-pediatric TBI, some patients experienced persisting working memory disturbance while others exhibited recovery; stratification was based on an event-related potential marker. More or less improvement in neurocognition was associated with the degree of WM disorganization measured by indicators of diffusivity. IHTT, measured post-acutely after TBI, and WM disorganization progression predicted neurocognitive trajectories at the chronic timeframe – potentially representing a prognostic biomarker.