Project Summary/Abstract
Autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) are common
neurodevelopmental disorders which exhibit enormous variability in their developmental trajectories. ASD and
ADHD also frequently co-occur, such that ASD is associated with elevated ADHD symptoms and vice versa.
Notably, such co-occurring ASD and ADHD symptoms are associated with greater impairment, as well as
reduced treatment responsiveness. However, the convergent and divergent neural underpinnings of ASD and
ADHD remain poorly understood, impeding the personalization of current treatments and the development of
more targeted ones. Furthermore, it is not yet possible to predict how an individual’s symptoms will change over
development. Yet, such predictions could be advantageous for treatment planning. The current project will
improve our understanding of the shared and distinct neural mechanisms underlying ASD and ADHD, as well as
our ability to predict how an individual’s symptoms may evolve over time. Specifically, this study will use magnetic
resonance imaging (MRI) to investigate the functional and structural properties of the brain in ASD and ADHD
by comparing the following groups: ASD, ADHD, comorbid ASD+ADHD, and neurotypical controls. Analyses will
be completed in both a lifespan sample (ages 5-65; N>2,700) and a pediatric sample (ages 9-10; N>4,900).
Functional connectivity will be calculated from resting-state functional MRI scans, structural connectivity from
diffusion tensor imaging (DTI) scans, and structural morphometry measures from T1-weighted structural MRI
scans. This multimodal neuroimaging data will also be used with baseline symptom severity to predict trajectories
of ASD, ADHD, and internalizing (e.g., anxious, depressive) symptoms between late childhood (ages 9-11) and
early adolescence (ages 11-13) in a longitudinal sample (N>700). Ridge regression analyses conducted within
each diagnostic group will reveal whether such brain-based information significantly improves predictive ability
compared to symptom severity alone. These analyses will be conducted both within groups defined by traditional
diagnostic categories and within transdiagnostic brain-based subgroups to determine the potential utility of such
subgroups in increasing predictive accuracy; these subtypes will be created using similarity network fusion on
subjects’ multimodal neuroimaging data, followed by spectral clustering. As a whole, this project will allow the
applicant to receive extensive training in cutting-edge neuroimaging methods, machine learning approaches
(ridge regression and spectral clustering), and conducting translational research. Most importantly, findings from
this research will improve our understanding of the shared and distinct mechanisms of ASD and ADHD, which
may ultimately lead to more tailored treatments. Furthermore, the proposed research may significantly improve
our ability to predict how an individual’s symptoms will change over time. This could have a direct impact on
individual treatment planning, as well as the design and implementation of future treatment studies.