Neural indicators of food cue reactivity, regulation, and valuation and their associations with body composition and daily eating behavior

Abstract

Exposure to food cues activates the brain’s reward system and undermines efforts to regulate impulses to eat. During explicit regulation, lateral prefrontal cortex activates and modulates activity in reward regions and decreases food cravings. However, it is unclear whether: (1) there are between-person differences in recruitment of regions associated with reward processing, subjective valuation, and regulation during food cue exposure—absent instructions to regulate; and (2) individual differences in activation patterns predict body composition and daily eating behaviors. In this preregistered study, we pooled data from several fMRI samples to examine whether regions associated with reward, valuation, and regulation, respectively, as well as whole-brain pattern expression indexing these processes, were recruited during food cue exposure and associated with body composition and real-world eating behavior. Cross-validated regression models indicated that univariate and multivariate measures of reward and valuation were associated with individual differences in BMI and enactment of daily food cravings. Specification curve analyses further revealed reliable associations between univariate and multivariate neural indicators of reactivity, regulation, and valuation, and all outcome measures. These findings highlight the utility of these methods to elucidate brain-behavior associations and suggest that multiple processes are implicated in proximal and distal markers of self-regulatory success and failure.

Neural Substrates of Food Valuation and Its Relationship With BMI and Healthy Eating in Higher BMI Individuals

“…When examining differences in the self-reported valuation of healthy and unhealthy foods, our results indicate that self-reported valuation of unhealthy foods is inversely related to real world healthy eating, but we found no relationship for the valuation of either food type with BMI. This finding echoes prior work demonstrating a relationship between unhealthy eating behavior and reactivity to unhealthy food cues (Tetley et al, 2009), and suggests that valuation may be an important factor underlying this relationship (Cosme and Lopez, 2020). Moreover, this highlights the importance of regulation strategies that target the devaluation of unhealthy foods, whether by reappraisal or other down-regulation strategies (Giuliani et al, 2014).…”

Revisiting the Neural Architecture of Adolescent Decision-Making: Univariate and Multivariate Evidence for System-Based Models

“…Although it is not a given that psychological modularity necessitates neural modularity, this assumption has been preserved in many neuroscientific implementations of system-based theories (Shulman et al, 2016;Strang et al, 2013), despite evidence in adults that multivariate patterns reflect meaningful information about decision-making (Hampton & O'Doherty, 2007). Our univariate and multivariate results, somewhat surprisingly, respectively provide support for both modularity and population coding (Cosme & Lopez, 2020) specifically, results revealed that univariate NAcc and lPFC activity was associated with decision behavior, and also that pattern expression of a multivariate cognitive control (but not value) signature predicted decisions. The former (evidence of modularity) is surprising, given the limited support for neural modularity that prior studies have found (Erickson, 2001), whereas the latter (population coding) is notable because no prior studies, to our knowledge, have found evidence of such in the context of brain modeling decision behavior (i.e., using multivariate metrics to model behavior/cognition).…”

“…In other words, the multivariate patterns code for a given psychological process whereas the univariate activity of the modules controls the intensity of the process. Indeed, such multidimensional coding schemes appear to support decision behaviors in monkeys (Zhang, Chen, & Monosov, 2019), and similar findings from human samples in other domains (eating behavior) further hint at the neural plausibility of a modular-population hybrid scheme (Cosme & Lopez, 2020). Further work could also examine whether there is a qualitative shift between coding schemes across development (Gee et al, 2013).…”

“…An alternative to brain mapping is brain modeling (Kragel et al, 2018), which uses neuroscientific data to predict cognitions and behavior (i.e., any kind of neural metric predicting behavior). Having recently grown in popularity, brain modeling approaches have seen broad applications, some of which involve within-person modeling, including prediction of food craving (Cosme & Lopez, 2020;Cosme, Ludwig, & Berkman, 2019), emotion regulation tendencies (Doré, Weber, & Ochsner, 2017), negative affect (Chang, Gianaros, Manuck, & Krishnan, 2015), chronic pain (Wager et al, 2013) and vision (Gardner & Liu, 2019;Liu, Cable, & Gardner, 2018). While brain mapping has generated key discoveries in neuroscience (e.g., Kanwisher, 2017), it can be problematic for evaluating system-based theories.…”

Age-related change in task-evoked amygdala—prefrontal circuitry: a multiverse approach with an accelerated longitudinal cohort aged 4-22 years

“…Despite not being completely comprehensive, this approach still allowed for thorough investigation into the robustness of results. For all multiverse analyses, we constructed specification curves by ranking models by their beta estimates (ascending) for parameters of interest for interpretation and visualization [92][93][94][95][96] . Because specification choices were not preregistered, we did not conduct formal null hypothesis testing of specification curves.…”