Research within the Section on Social and Cognitive Developmental Neuroscience

National Institute Of Mental Health

Investigators

Linked publications, trials & patents

Abstract

We have focused on four primary projects in the past year, comprising: (i) data collection for the first 7 Tesla longitudinal neuroimaging study relating neurodevelopmental traits (impulsivity, compulsivity, social responsiveness) with ultra-high field assessments of brain structure, function and neurochemicals, and longitudinal assessments of mental health outcomes; in the last year we completed 51 new participant assessments, and 81 neuroimaging scans with new and follow-up participants; (ii) developing a simulation framework to evaluate best practices for longitudinal data analysis, which will inform analytic strategies for Project 1; (iii) conducting multiple projects using data from pre-existing 3 Tesla datasets to inform hypotheses for Project 1, including largescale mega-analytic reviews quantifying relationships between brain function and neurodevelopmental traits, as well as a large longitudinal assessment of the neurochemical glutamate and outcome trajectories in ADHD; and (iv) evaluating a large cohort of twins from Virginia Commonwealth University’s Mid-Atlantic Twin Registry to characterize genetic, environmental, and stochastic influences on neurodevelopmental traits and their co-occurrence with related conditions and traits. For Project 1, we are recruiting children and adults who vary in impulsivity, compulsivity, and social difficulties, and we follow them over time to test whether changes in brain chemistry, structure, and connectivity are related to developmental outcomes. Earlier work indicates three brain signatures associated with these traits: differences in glutamate levels, altered communication between cortical and subcortical brain regions, and structural differences within small subcortical nuclei within cortico‑striato‑thalamo‑cortical circuits. The 7 Tesla (7T) scanner is an ultra‑high‑field MRI; Tesla is the unit of magnet strength, and 7T is nearly twice as strong as conventional 3T neuroimaging (magnetic field strength scales by the square root of the field strength). This higher field gives substantially sharper images and clearer spectroscopy, which improves separation of neurochemicals like glutamate and GABA and allows us to study small subcortical nuclei at a much finer spatial resolution. The longitudinal design will allow us to link differences in neural development to changes in mental health symptoms and daily functioning. We are actively recruiting and building the dataset; preliminary findings link differences in GABA levels to ADHD symptoms, and ongoing analyses are examining neurodevelopmental traits in relation to resting‑state fMRI activity and structural metrics within the relevant subcortical regions. For Project 2, we coordinated an international effort aimed at investigating the longitudinal interplay between brain, behavior, cognition and condition (ASD vs. neurotypicals) from childhood into young adulthood. One of the challenges of longitudinal studies is the lack of a gold standard related to the nature of the underlying relationship between brain and behavior. Thus, applying different statistical models to longitudinal data, considering that different models have different underlying assumptions, will allow us to test which models work best with specific underlying assumptions. By working together with other groups at NIH and four external sites, we have created large simulated longitudinal datasets, which have been openly shared with the wider research community. Research groups are invited to explore statistical models that capture the associations between brain and behavior in the simulated datasets. The project proposal was announced during the Organization for Human Brain Mapping (OHBM) meeting in 2023, the simulated data was made available during OHBM 2024, the ground truth (code used to generate the simulated data) was released in Spring of 2025 along with a published manuscript detailing the design of the study and dataset characteristics. All the code and the data are freely available via GitHub and the Open Science Framework. In addition, a second manuscript analyzing the data has been submitted to Aperture Neuro. For Project 3, we are leveraging existing 3T datasets to generate and refine hypotheses for the ultra-high field 7T study. One publication examined social difficulties and ADHD traits in more than 12,000 participants aggregated across multiple large 3T neuroimaging datasets. The analyses revealed altered striatocortical and thalamocortical connectivity in association with neurodevelopmental traits. In follow up work using four waves from the Adolescent Brain Cognitive Development Study, associations with ADHD traits were robust to controls for general psychopathology and were stable over development in 3T data; this manuscript is planned for submission in late 2025. In parallel, a manuscript now at an advanced stage of peer review uses 3T MRS data collected within the NIH Intramural Research Program, showing that reductions in impulsivity related symptoms and remission of ADHD coincide with a normalization of glutamate levels. Together, these 3T findings are shaping region of interest selection and longitudinal hypotheses for the 7T study in Project 1. For Project 4, in collaboration with Virginia Commonwealth University’s Mid-Atlantic Twin Registry (MATR), we screened a large community sample of twins for neurodevelopmental traits and cooccurring mental health symptoms. In the past year, we analyzed data from 815 twin pairs recruited through our collaborative project with VCU. Cotwin, within pair analyses among monozygotic (MZ) twins, which control for shared genetics and family-wide environment, indicate that differences in neurodevelopmental traits covary with differences across a wider set of mental health and behavioral outcomes, including sleep. These patterns have been replicated in an independent sample of approximately 800 twin pairs from the Adolescent Brain Cognitive Development (ABCD) Study. We also applied twin modeling to estimate the genetic and environmental contributions to traits such as sleep disturbance and social responsiveness, as well as their covariation. This allows us to assess whether the association between these traits are driven primarily by shared genetic factors or common environmental influences. We next plan to recruit a subset of VCU twin pairs into Project 1 for 7T neuroimaging, prioritizing pairs discordant for the neurodevelopmental traits of interest to maximize sensitivity to within pair associations between brain measures and behavior. In addition to these projects, we have simultaneously been utilizing large datasets, including the Adolescent Brain Cognitive Development (ABCD) study and the Generation R study to examine several research questions. These studies include i). maternal viral infection risk and development of autistic traits; ii). the role of plasticizers in the environment and downstream brain development; iii). neuroimaging approaches to differentiate the course of anorexia nervosa in adolescent girls; iv). social responsiveness in children with and without a diagnosis of ASD: cognitive and behavioral correlates.