Brain plasticity underlying acquisition of new organizational skills in children: A Rashomon analysis

Luis Martinez Agulleiro¹Bowen Deng¹Richard Gallagher¹Howard A Abikoff¹Yuliya N Yoncheva¹Lauren Robinson²Greta R Conlon^1,3Maleeha Haroon¹Chao-Gan Yan⁴Adriana Di Martino⁵Yihong Zhao⁶Francisco Xavier Castellanos^2*

• ¹Department of Child and Adolescent Psychiatry, New York University Grossman School of Medicine, New York, United States
• ²Child and Adolescent Psychiatry NYU Langone Medical Center, Grossman School of Medicine, New York University, New York, United States
• ³Department of Psychology, The University of Alabama at Birmingham College of Arts and Sciences, Birmingham, United States
• ⁴Department of Psychological and Cognitive Sciences, Tsinghua University, Beijing, China
• ⁵Autism Center, Child Mind Institute Inc, New York, United States
• ⁶School of Nursing, Columbia University, New York, United States

Objective: We used resting-state functional magnetic resonance imaging to identify changes in brain functional connectivity (FC) associated with Organizational Skills Training (OST). Method: In an open, waitlist-controlled, randomized clinical trial (NCT04108273), 51 children aged 8–12 years with deficient organizational skills were assigned to immediate tele-health OST treatment (twice weekly, 10 weeks) or waitlist. We obtained Children's Organizational Skills Scale-Parent version (COSS-P) scores and examined FC changes between dorsal anterior cingulate cortex (dACC) and preregistered subcortical anterior ventral striatum (aVS) regions-of-interest. Results: OST produced significantly lower COSS-P scores compared to waitlist, with a large effect size (Cohen's f² = 0.77). Initial imaging analyses revealed a significant increase (instead of the predicted decrease) in FC between dACC and the aVS component of the default mode network in the immediate treatment group (FC = 0.092  0.041, 95% CI [0.009, 0.175], p < 0.05). Analyses were then performed with two additional analytic pipelines, neither of which detected any significant effects. Conclusion: Although improvements in organizational deficits were associated with increased FC within a circuit linking dACC and the default mode network region of the aVS in one analysis, the direction was the opposite of predicted and results did not replicate. Thus, we highlight the tentativeness of our findings; we have de-identified all the data and made it available for investigators to examine and to combine with other datasets in mega-and meta-analyses. Future studies should also include alternative control conditions and larger samples.