Adiposity covaries with signatures of asymmetric feedback learning during adaptive decisions

Abstract

<jats:title>Abstract</jats:title> <jats:p>Unhealthy weight gain relates, in part, to how people make decisions based on prior experience. Here we conducted post hoc analysis on an archival data set to evaluate whether individual differences in adiposity, an anthropometric construct encompassing a spectrum of body types, from lean to obese, associate with signatures of asymmetric feedback learning during value-based decision-making. In a sample of neurologically healthy adults (N = 433), ventral striatal responses to rewards, measured using fMRI, were not directly associated with adiposity, but rather moderated its relationship with feedback-driven learning in the Iowa Gambling Task tested outside the scanner. Using a biologically-inspired model of basal ganglia-dependent decision processes, we found this moderating effect of reward reactivity to be explained by an asymmetrical use of feedback to drive learning; that is, with more plasticity for gains than for losses, stronger reward reactivity leads to decisions that minimize exploration for maximizing long-term outcomes. Follow-up analysis confirmed that individual differences in adiposity correlated with signatures of asymmetric use of feedback cues during learning, suggesting that reward reactivity may especially relate to adiposity, and possibly obesity risk, when gains impact future decisions more than losses.</jats:p>

D2 dopamine receptor expression, sensitivity to rewards, and reinforcement learning in a complex value-based decision-making task

“…Of primary importance for future work determining the mechanism by which sensitivity to feedback signals interacts might reward reactivity. Our reinforcement learning model, as well as general intuition, shows clearly that these two factors should interact, yet we failed to find this in our data (but see Verstynen et al, 2020). One possibility could be the need to find a better or more specific, marker of phasic dopamine responses.…”

“…Participants were scanned on a 3 T Trio TIM whole-body scanner (Siemens, Erlangen, Germany) using a 12-channel phased-array head coil (FOV) = 200 × 200 mm, matrix = 64 × 64, repetition time (TR) = 2000 ms, echo time (TE) = 29 ms and flip angle (FA) = 90◦, for more information see (Verstynen et al, 2020). While in the MRI scanner, participants completed a computerized reward task paradigm (for preprocessing information see (Verstynen et al, 2020). After preprocessing, linear contrast images, reflecting relative BOLD signal changes (i.e.…”

“…win blocks versus loss blocks), were estimated for each participant using general linear model (GLM) estimation. The mean BOLD contrast parameter estimates were extracted from a predefined VS ROI (Verstynen et al, 2020; Gianaros et al, 2011). For more information on the estimation process and creating the a priori ROI mask, see (Verstynen et al, 2020), and Gianaros et al, 2011, respectively.…”

Health neuroscience 2.0: integration with social, cognitive and affective neuroscience

“…Illustrative examples from the current special issue that capitalize on this perspective include those that use state-of-the-science analytical approaches (e.g. machine learning to identify multivariate patterns of neural activity) to predict a host of the most pressing health issues facing society today: cardiovascular disease ( Gianaros et al , 2020 ), obesity ( Stice et al , 2019 ; Cosme et al , 2020 ; Donofry et al , 2020 ; Verstynen et al , 2020 ) and physical pain ( Reddan et al , 2020 ). Similar approaches help us understand links between brain patterns of activity to emotional content and systemic inflammation, a key biological mediator linking psychological experience and health ( Alvarez et al , 2020 ), or to health messages and population-level sharing of the information ( Dore et al ., in press ).…”

“…As both healthy life expectancy and life expectancy increase worldwide ( World Health Organization, 2020 ), there is also an urgent need to conduct research at later stages of life. Existing longitudinal brain datasets are publicly available for exploration (examples from the current special issue include MIDUS ( Alvarez et al, 2020 ) and AHAB-II ( Gianaros et al , 2020 ; Verstynen et al , 2020 )), but an additional way that social, cognitive and affective neuroscientists can contribute to the lifespan perspective is by extending their paradigms to new age groups (especially midlife and later life).…”