This project will investigate behavioral and neural mechanisms of learning from prediction and choice using a developmental cognitive neuroscience approach. We hypothesize that the developing child is intrinsically motivated to seek information and recruits neural circuits related to reward and reinforcement during the learning process. This project has two Specific Aims: 1) to investigate whether predictive information is intrinsically rewarding, and 2) to investigate the reward value of choice and its effects on learning as the child gains autonomy and behavioral control across development. Using a combination of cutting-edge techniques (i.e., eye tracking, behavioral, fMRI) across four experiments with infants, toddlers, and children, we will probe early learning behavior and its neural bases in the developing brain. We will establish the use of eye-blink rate (EBR) to index both reward anticipation and recruitment of striatal dopaminergic circuitry during learning in infants and children. In Experiments 1.1 and 1.2, we will present participants with stimuli that parametrically increase in reward value or in predictiveness. Changes in EBR and striatal activity will be measured and related to children's preferences, orienting responses, and memory accuracy for rewarding and predictive stimuli. In Experiments 2.1 and 2.2, children will participate in a novel choice vs. no-choice learning task, in which we will examine attention and memory, as well as changes in EBR and striatal activation, for animations that children select themselves, compared to those that are selected for them. We expect that the intrinsic reward of choice will enhance children's preferences, memory, and striatal activation for chosen information, and will increase as the child becomes more autonomous and capable of controlling behavior. Through the Research and Training Plan, the applicant will deepen her theoretical and conceptual knowledge of developmental, cognitive and neural mechanisms of human learning, while advancing her skills in eye-tracking, neuroimaging, and computational methods. The project will refine current ideas about the roles of prediction and endogenous choice in learning during typical development, which will inform our understanding of striatal dopamine networks and the cognitive processes in which they are implicated. This work will lay the foundation for asking principled questions about risk-factors and biomarkers in populations with clinical and developmental disruptions to reward and prediction-related circuitry, such as children with ADHD, learning disabilities, schizophrenia, and addiction.

Public Health Relevance

Disruptions to reward and prediction-related neural circuitry are associated with a number of clinical and developmental disorders including ADHD, learning disabilities, schizophrenia, and addiction. This project will investigate behavioral and neural mechanisms of learning from prediction and choice in early development, considering whether infants and children are intrinsically motivated to seek information and engage reward- related striatal regions during learning. Our refined understanding of the roles of prediction and choice in learning during typical development will lay the foundation for asking principled questions about risk-factors and biomarkers for atypical developmental trajectories beginning early in postnatal life.

Agency
National Institute of Health (NIH)
Institute
National Institute of Mental Health (NIMH)
Type
Postdoctoral Individual National Research Service Award (F32)
Project #
5F32MH108278-03
Application #
9385007
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Desmond, Nancy L
Project Start
2015-12-01
Project End
2018-11-30
Budget Start
2017-12-01
Budget End
2018-11-30
Support Year
3
Fiscal Year
2018
Total Cost
Indirect Cost
Name
Brown University
Department
Social Sciences
Type
Schools of Arts and Sciences
DUNS #
001785542
City
Providence
State
RI
Country
United States
Zip Code
02912
Tummeltshammer, Kristen; Amso, Dima; French, Robert M et al. (2017) Across space and time: infants learn from backward and forward visual statistics. Dev Sci 20:
Tummeltshammer, Kristen; Amso, Dima (2017) Top-down contextual knowledge guides visual attention in infancy. Dev Sci :