Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder marked by deficits in executive control affecting millions of children and adults. Impaired motivation is also a frequent complaint; indeed, alterations in the dopamine system, the primary neurobiological substrate of motivation, have long been implicated in ADHD. Nonetheless, the way in which motivational deficits impact cognitive processes, such as learning and decision making, in individuals with ADHD is not yet fully understood. This knowledge gap represents a significant hurdle in translation: new approaches are required to develop curative interventions free of the adverse side effects of existing therapies for ADHD. Here, we propose several studies to advance understanding of motivational processes in ADHD using computational modeling; we further test the efficacy of a novel intervention for impaired motivation, cognitive neurostimulation (CN), on learning in ADHD. CN is a non- invasive neurobehavioral training technique that has been shown to reliably and sustainably stimulate dopaminergic circuitry through self-generated motivational imagery.
In Aim 1, we will use eye tracking and attentional drift diffusion modeling (aDDM) to characterize attention-motivation interactions in ADHD using a model-based versus model-free reinforcement learning (RL) paradigm that has previously been shown to have substantial explanatory and predictive power in in psychiatric disorders. Given that model-free strategies have been associated with disrupted dopamine neurotransmission, we predict that attention deficit severity will be reflected in noisier decision processes and a relative reliance on model-free (i.e., trial-by-trial) relative to model- based (i.e., goal-oriented) learning.
In Aim 2, we will test the effect of CN on the system characterized in Aim 1, testing how volitional enhancement of motivational state impacts attention-motivation interactions (as characterized by aDDM model parameters). We predict that CN will enhance motivation through stimulation of the dopamine system, improving RL and increasing attentional gain. Finally, in Aim 3, we will use an exploratory approach to test whether aDDM model parameters can predict individual response to CN. We predict that, because of its unique ability to describe attention-motivation interactions, these parameters will, compared to other behavioral metrics, better reflect underlying neurobiological processes in ADHD and their ability to be modulated by CN. Together, these findings will advance our understanding of the neural and behavioral mechanisms shaping disrupted motivation in ADHD while also providing new insight into the efficacy of a potential intervention for behavior change in this population. The proposed research will also demonstrate how an interdisciplinary, neuroscience-based approach might inform intervention design for individualized behavior change, a process that has implications for optimal wellbeing across the spectrum of mental health.
Attention-deficit/hyperactivity disorder (ADHD) is characterized not only by deficits in executive control, but also by impaired motivation; motivational deficits significantly impact both quality of life and the potential efficacy of therapeutic intervention, since motivation is critical for successful learned behavior change. By combining computational approaches with a non-invasive, real-time fMRI neurofeedback intervention, we propose to model how self-activation of motivational neural circuitry impacts reinforcement learning in ADHD and to test whether this model of attention-motivation interaction can predict intervention response. We expect that these studies will provide new insight into the neurobehavioral basis of the complex interactions between attention, motivation, and cognition in ADHD while also demonstrating how computational models may be used to inform novel, efficacious interventions for psychopathology.
