Study of Brain, Reward, and Kids' Eating (BRAKE)

Children from rural communities are at greater risk for obesity than children from more urban communities. However, some children are resilient to obesity despite greater exposure to obesogenic influences in rural communities (e.g., fewer community-level physical activity or healthy eating resources). Identifying factors that promote this resiliency could inform obesity prevention. Eating habits are learned through reinforcement (e.g., hedonic, familial environment), the process through which environmental food cues become valued and influence behavior. Therefore, understanding individual differences in reinforcement learning is essential to uncovering the causes of obesity. Preclinical models have identified two reinforcement learning phenotypes that may have translational importance for understanding excess consumption in humans: 1) goal-tracking-environmental cues have predictive value; and 2) sign-tracking-environmental cues have predictive and hedonic value (i.e., incentive salience). Sign-tracking is associated with poorer attentional control, greater impulsivity, and lower prefrontal cortex (PFC) engagement in response to reward cues. This parallels neurocognitive deficits observed in pediatric obesity (i.e., worse impulsivity, lower PFC food cue reactivity). The proposed research aims to determine if reinforcement learning phenotype (i.e., sign- and goal-tracking) is 1) associated with adiposity due to its influence on neural food cue reactivity and 2) associated with reward-driven overconsumption and meal intake due to its influence on eating behaviors. The investigators hypothesize that goal-tracking will promote resiliency to obesity due to: 1) reduced attribution of incentive salience and greater PFC engagement to food cues; and 2) reduced reward-driven overconsumption. Finally, the investigators hypothesize reinforcement learning phenotype will be associated due to its influence on eating behaviors associated with overconsumption (e.g., larger bites, faster bite rat and eating sped). To test this hypothesis, the investigators will enroll 76, 8-9-year-old children, half with healthy weight and half with obesity based on Centers for Disease Control definitions. Methods will include computer tasks to assess reinforcement learning, dual x-ray absorptiometry to assess adiposity, and neural food cue reactivity from functional near-infrared spectroscopy (fNIRS).

Children will rate foods on taste, health, and desire to eat. The order in which they rate the food characteristics is randomly assigned and counter-balanced across participants

Functional near infrared spectroscopy (fNIRS) will measure brain activity through oxy- and deoxyhemoglobin in response to images of high and low energy dense foods.

Intake in grams during a snack buffet using a standard eating in the absence of hunger paradigm (i.e., non-homeostatic intake)

Intake in kcal during a snack buffet using a standard eating in the absence of hunger paradigm (i.e., non-homeostatic intake)

The 2-stage reinforcement learning task has a first stage two arm bandit with deterministic stage progression and a second stage one arm bandit. Reward distributions between the two second-stage states gradually drift throughout the task. Half the trials will be 'bonus' trials. Performance will be assessed using a computational model in addition to looking at trial-to-trial decisions.

The value-modulated attentional capture task uses two phases - a training phase during which high and low reward conditions are learned and a test phase during which participants complete a task that no longer depends upon the previously learned reward contingencies. During the test phase, stimuli from the training phase are used as distractors. Attentional capture is measured by comparing performance on trials that have distractors previously associated with high reward to those with distractors previously associated with low reward.

A digital recording of the child eating a standard meal will be saved. We have developed a behavior coding protocol to measure child meal microstructure (e.g., bites, bite size, meal duration). We have also validated a computational model to assess cumulative intake curves from video coded bite data.

Functional near infrared spectroscopy will measure brain activity through oxy- and deoxyhemoglobin while children rate food images on health, taste and wanting

Functional near infrared spectroscopy will measure brain activity through oxy- and deoxyhemoglobin while children choose which of two foods they would like to eat

The extent to which previously reward distractors capture attention will be assessed with eye tracking

Eye-tracking will be measured to determine if attention is drawn to the tastier food item prior to making a food choice

A digital recording of the child during the eating in the absence of hunger protocol will be saved. We will use the video to characterize the amount of attention children give toward the food items when they are not hungry and code behaviors associated with self-control

Ratings of how much a child likes a food is measured on a computerized visual scale. The scale is from 1 (Hate It) to 5 (Love It).

Child fullness will be measured using a pictorial fullness scale termed a Freddy Fullness scale. Prior to the visit, children will be instructed to fast for at least 3hrs. They are then instructed on use of the scale by trained research personnel. Following this, children will be asked to report current fullness on the scale, which measures a fullness range of 0 to 150 mm. This will be done before and after each meal and taste test. The scale will also be used before and after the fNIRS session.

An ActiGraph watch will be given to the parent with instruction. The child will be wearing the wrist watch for one week in the baseline time period (other than when showering, bathing or in a pool) and will hand it back at their next visit.

An ActiGraph watch will be given to the parent with instruction. The child will be wearing the wrist watch for one week in the baseline time period (other than when showering, bathing or in a pool).

Self-report questionnaire for child: The Children's Anxiety Meter Scale measures state anxiety before and after the fNRIS session

Self-report questionnaire for child: The Perceived Stress Scale asks the child about feelings over the last month to assess perceived stress

Self-report questionnaire for child: The loss of control eating questionnaire asks children if they have recently experienced and episode of loss of control eating

Parental report questionnaire: information describing the cultural, social and financial characteristics of the family.

Parental report questionnaire: Child Puberty and Tanner Questionnaire is score in the following way: Male genitals are scored on a scale of 1 to 5 maturity, female breasts on a scale of 1 to 5 and both males and females on a scale of 1 to 5 for pubic hair quality and extension. Higher values indicate more pubertal development

Parental report questionnaire: Child Feeding Questionnaire (CFQ) is scored on a scale of 1-5 with lower values being better (score is the average of items for each subscale).

Parental report questionnaire: The Child Eating Behavior Questionnaire (CEBQ) was designed to assess children's eating scale styles. It is a parent-report measure comprised of 35 items, each rated on a five-point likert scale that ranges from never to always. It is made up of eight scales: Food responsiveness, Emotional over-eating, Enjoyment of food, Desire to drink, Satiety responsiveness, Slowness in eating, Emotional under-eating, and Food fussiness.

Parental report questionnaire: The Child Behavior questionnaire (CBQ) is an assessment of temperament.Children are assessed on 15 primary temperament characteristics using a 7 point Likert scale.

Parental report report questionnaire: The Behavior Rating Inventory of Executive Function (BRIEF-2) assesses executive function and self-regulation. Questions are answered on a 3-point scale (never, sometimes, often). This rating is scored by taking a sum of all items and referencing and age- and sex-normalized tables to get T-scores and percentiles; Higher T-scores indicate less Executive Function.

Parental report questionnaire: The Binge Eating Scale asks parents about children's eating behaviors related to binge eating and overeating. Response are from 1 - 4 which scales from the least to most severe

Parental report questionnaire: The Child Sleep Habits Questionnaire contains 8 questions describing a child's sleep habits. The answers are on a 3 point scale with an opportunity to note if the answer indicates a problem.

Parental report questionnaire: The External Food Cue Responsiveness Scale Questionnaire asks nine questions related to external food cues, answered in a 1 (never) -5 (always) scale. Higher scores indicate an increased responsiveness to external food cues.

Parental report questionnaire: The Family Food Behavior Survey asks parents about the food and feeding behaviors in the home and has the following sub scales: and provides subscale scores for the following behaviors: Maternal Control, Maternal Presences, Child Choice, and Organization. Questions are scored from 0 - Never True, 1 - Rarely True, 2 - Sometimes, 3 - Often True, 4 - Always True

Parental report questionnaire: The Sensitivity to Punishment and Reward Questionnaire asks about child behaviors and has the follow sub scales: Fear/Shyness, Anxiety, Conflict Avoidance, Sensory Reward, Drive, Responsiveness to Social Approval, Impulsivity/Fun Seeking. The original 4 subscales (2004; 34 item subscales): Sensitivity to Punishment, Impulsivity/Fun Seeking, Drive, and Reward Responsiveness. Parents respond from 1 - Strongly Disagree, 2 - Disagree, 3 - Neither Agree nor Disagree, 4 - Agree, 5 - Strongly Agree

Self-report questionnaire for the parent: The Three Factor Eating Questionnaire asks the parents about their own eating behaviors and has the following subscales - Cognitive Control of Eating Behaviors, Disinhibition of Control, and Susceptibility to Hunger.

Self-report questionnaire for the parent: The Parent Weight Loss Behavior Questionnaire asks parents about their use of healthy and unhealthy weight loss behaviors

The child will be given the Wechsler Abbreviated Scale of Intelligence to estimate cognitive ability. It is a battery of four subtests: Vocabulary (31-item), Block Design (13-item), Similarities (24-item) and Matrix Reasoning (30-item). Each of the 4 subtests is scored by taking a sum of all items and referencing and age- and sex- normed tables to get standardized scores; these standardized scores are then added to get the 3 subscale scores, their associated IQ scores percentiles; Higher scores indicate a higher IQ.

The Dellis-Kaplan Executive Function System Trail Making Test, Design Fluency, and Verbal Fluency will be used to assess motor processing speed, cognitive processing speed, fluency, and switching

The Flanker is a measure of interference. Scoring is based on a combination of accuracy and reaction time. A 2-vector scoring method is employed that uses accuracy and reaction time, where each of these "vectors" ranges in value between 0 and 5, and the computed score, combining each vector score, ranges in value from 0-10. For any given individual, accuracy is considered first. If accuracy levels for the participant are less than or equal to 80%, the final "total" computed score is equal to the accuracy score. If accuracy levels for the participant reach more than 80%, the reaction time score and accuracy score are combined. Higher scores indicate higher levels of ability to attend to relevant stimuli and inhibit attention from irrelevant stimuli.

The List Sorting Working memory test assesses working memory. The List Sorting test requires immediate recall and sequencing of different visually and orally presented stimuli (i.e., "working memory"). Pictures of different foods and animals are displayed with accompanying audio recording and written text (e.g., "elephant"), and the participant is asked to say the items back in size order from smallest to largest, first within a single dimension (either animals or foods, called 1-List) and then on two dimensions (foods, then animals, called 2-List). The test takes approximately seven minutes to administer. List Sorting is scored by summing the total number of items correctly recalled and sequenced on 1-List and 2-List, which can range from 0-26.Higher scores on each of these indicate higher levels of working memory within the normative standard being applied.

The Dimensional Change Card Sort Test is used to measure cognitive flexibility. Two target pictures are presented that vary along two dimensions (e.g., shape and color). Scoring is based on a combination of accuracy and reaction time. A 2-vector scoring method is employed that uses accuracy and reaction time, scores ranging from 0-10. For any given individual, accuracy is considered first. If accuracy levels for the participant are less than or equal to 80%, the final "total" computed score is equal to the accuracy score. If accuracy levels for the participant reach more than 80%, the reaction time score and accuracy score are combined. Higher scores indicate higher levels of cognitive flexibility.

The parent who primarily makes food related decisions in the house will accompany the child to the visit and will have their height and weight measured. They will also report on the height and weight of the other parent

Child Inclusion Criteria: In order to be enrolled, children must be of good health based on parental self-report. Have no neurodevelopmental disorder (e.g., attention deficit hyperactivity disorder - ADHD) or learning disabilities (e.g., dyslexia). Have no allergies to the foods or ingredients used in the study. Not be taking any medications known to influence body weight, taste, food intake, behavior, or blood flow. Be 8-9 years-old at enrollment. speaks English. Parent Inclusion Criteria: The parent who has the most knowledge of the child's eating behavior, sleep and behavior must be available to attend the visits with their child. This would be decided among the parents. Exclusion Criteria: They are not within the age requirements (< than 8 years old or > than 9 years-old at baseline). If they are taking cold or allergy medication, or other medications known to influence cognitive function, taste, appetite, or blood flow. don't speak English. are colorblind. has a learning disability, ADHD, language delays, autism or other neurological or psychological conditions. has a pre-existing medical condition such as type I or type II diabetes, rheumatoid arthritis, Cushing's syndrome, Down's syndrome, severe lactose intolerance, Prader-Willi syndrome, HIV, cancer, renal failure, or cerebral palsy. is allergic to foods or ingredients used in the study. has had an X-ray in the month prior to Visits 1. If so, they will be scheduled at a later date. Parent Exclusion Criteria: the parent is unable to attend the study visits

All protocols, methods, and data obtained from this project will be made publicly available following the National Institutes of Health's FAIR principles on Open Science Framework or other sites for data sharing. The final dataset will be published with a persistent identifier to ensure that the dataset will be Findable even if the hosting platforms change. Rich meta-data will be published to ensure the data are Accessible. Where possible, meta-data will use formal and searchable language and terms based on common ontologies so that the data are Interoperable. In order to ensure Reusability, meta-data will include detailed information about the protocols and methods following scientific community standards. Data intended for broader use will be free of identifiers that would permit linkages to individual research participants and variables that could lead to deductive disclosure of individual subjects.

Pearce AL, Adise S, Roberts NJ, White C, Geier CF, Keller KL. Individual differences in the influence of taste and health impact successful dietary self-control: A mouse tracking food choice study in children. Physiol Behav. 2020 Sep 1;223:112990. doi: 10.1016/j.physbeh.2020.112990. Epub 2020 Jun 4.

Pearce AL, Cevallos MC, Romano O, Daoud E, Keller KL. Child meal microstructure and eating behaviors: A systematic review. Appetite. 2022 Jan 1;168:105752. doi: 10.1016/j.appet.2021.105752. Epub 2021 Oct 16.

Fuchs BA, Roberts NJ, Adise S, Pearce AL, Geier CF, White C, Oravecz Z, Keller KL. Decision-Making Processes Related to Perseveration Are Indirectly Associated With Weight Status in Children Through Laboratory-Assessed Energy Intake. Front Psychol. 2021 Aug 18;12:652595. doi: 10.3389/fpsyg.2021.652595. eCollection 2021.

Rangel A. Regulation of dietary choice by the decision-making circuitry. Nat Neurosci. 2013 Dec;16(12):1717-24. doi: 10.1038/nn.3561. Epub 2013 Nov 22.

van Meer F, Charbonnier L, Smeets PA. Food Decision-Making: Effects of Weight Status and Age. Curr Diab Rep. 2016 Sep;16(9):84. doi: 10.1007/s11892-016-0773-z.

Colaizzi JM, Flagel SB, Joyner MA, Gearhardt AN, Stewart JL, Paulus MP. Mapping sign-tracking and goal-tracking onto human behaviors. Neurosci Biobehav Rev. 2020 Apr;111:84-94. doi: 10.1016/j.neubiorev.2020.01.018. Epub 2020 Jan 20.