Using behavioral economics to understand reinforcement mechanisms of loss-of-control eating: An ecological momentary assessment approach.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice J Psychopathol Clin Sci . Author manuscript; available in PMC: 2026 Apr 10. Published in final edited form as: J Psychopathol Clin Sci. 2025 Apr 10;134(7):832–843. doi: 10.1037/abn0000996 Search in PMC Search in PubMed View in NLM Catalog Add to search Using Behavioral Economics to Understand Reinforcement Mechanisms of Loss-of-Control Eating: An Ecological Momentary Assessment Approach Emily K Burr Emily K Burr 1 Department of Psychology, University of Central Florida Find articles by Emily K Burr 1 , Lidia Z Meshesha Lidia Z Meshesha 1 Department of Psychology, University of Central Florida Find articles by Lidia Z Meshesha 1 , Robert D Dvorak Robert D Dvorak 1 Department of Psychology, University of Central Florida Find articles by Robert D Dvorak 1 , Quinn Allen Quinn Allen 1 Department of Psychology, University of Central Florida Find articles by Quinn Allen 1 , Tatiana Magri Tatiana Magri 1 Department of Psychology, University of Central Florida Find articles by Tatiana Magri 1 , Callie L Wang Callie L Wang 1 Department of Psychology, University of Central Florida Find articles by Callie L Wang 1 , Emma R Hayden Emma R Hayden 1 Department of Psychology, University of Central Florida Find articles by Emma R Hayden 1 , Nadia E Rodriguez Nadia E Rodriguez 1 Department of Psychology, University of Central Florida Find articles by Nadia E Rodriguez 1 , Angelina V Leary Angelina V Leary 1 Department of Psychology, University of Central Florida Find articles by Angelina V Leary 1 , Madison Maynard Madison Maynard 1 Department of Psychology, University of Central Florida Find articles by Madison Maynard 1 , Stephen A Wonderlich Stephen A Wonderlich 2 Center for Biobehavioral Research, Sanford Research 3 Department of Psychiatry, University of North Dakota School of Medicine and Health Sciences Find articles by Stephen A Wonderlich 2, 3 , Glen Forester Glen Forester 2 Center for Biobehavioral Research, Sanford Research 3 Department of Psychiatry, University of North Dakota School of Medicine and Health Sciences Find articles by Glen Forester 2, 3 , Lauren M Schaefer Lauren M Schaefer 2 Center for Biobehavioral Research, Sanford Research 3 Department of Psychiatry, University of North Dakota School of Medicine and Health Sciences Find articles by Lauren M Schaefer 2, 3 Author information Article notes Copyright and License information 1 Department of Psychology, University of Central Florida 2 Center for Biobehavioral Research, Sanford Research 3 Department of Psychiatry, University of North Dakota School of Medicine and Health Sciences Issue date 2025 Oct. PMC Copyright notice PMCID: PMC12353950  NIHMSID: NIHMS2071457  PMID: 40208741 The publisher's version of this article is available at J Psychopathol Clin Sci Abstract Loss-of-control eating (LOCE) is the subjective inability to stop eating once one has started or to refrain from food consumption. State-level affect, food craving, and reward dysfunction have all been implicated as vulnerabilities to recurrent LOCE, mostly studied in the context of binge eating (i.e., LOCE with objective overeating). Hypothetical purchase tasks are a behavioral economic approach to assessing the reward value of a given behavior or commodity, which have typically been used in substance use literature. The current study tested a momentary mediation model in which positive and negative affect at Time 1 was hypothesized to predict Time 2 food demand (assessed using three variables from an ambulatory food purchase task), in turn leading to LOCE at Time 3 by way of Time 2 craving (affect → food demand → craving → LOCE). This model was assessed using a ten-day ecological momentary assessment protocol in 75 community adults with recurrent LOCE (87% female; 71% white). At the within-subjects (i.e., momentary) level, LOCE was predicted by prior food craving. Food reward value metrics additionally mediated the LOCE antecedent of negative affect, but not positive affect. Interestingly, between-subjects, the relationship between craving and LOCE was unexpectedly negative and only negative affect was associated with subsequent LOCE, by way of time two craving but not time two food reward value. Clinical implications and future directions are discussed. Keywords: loss-of-control eating, binge eating, reward, craving, affect General Scientific Summary: Aberrations in reward processing and maladaptive responses to affect have both been implicated in risk for loss-of-control eating. However, research has largely neglected to look at the temporal associations of these phenomena, particularly through a behavioral economic lens, which allows for real-time quantifying of food demand. This study fills that gap and provides nuanced evidence for the role of food demand and craving in the more established predictive nature of affect leading to uncontrolled eating. Loss of control eating (LOCE) involves the consumption of food paired with a perceived sense that one is unable to stop or resist eating. LOCE may occur within objective binge-eating episodes when paired with the consumption of an objectively large amount of food ( American Psychiatric Association, 2022 ), but may also occur outside such episodes. Indeed, subjective binge episodes, characterized by LOCE with the perception one has overeaten (despite an absence of true objective overeating) are perhaps more prevalent than objective binge episodes ( Brownstone & Bardone-Cone, 2021 ; Goossens et al., 2009 ). Although the current edition of the Diagnostic and Statistics Manual (DSM-5-TR; American Psychiatric Association, 2022 ) focuses exclusively on objective binge-eating episodes to define both bulimia nervosa and binge-eating disorder, the International Classification of Diseases (ICD-11) does not make such distinctions, focusing more squarely on the experience of LOCE rather than the amount of food consumed to define these disorders ( World Health Organization, 2018 ). Regardless of portion size, recurrent LOCE is associated with adverse psychological and physical outcomes, including mood and anxiety disorders ( Araujo et al., 2010 ; Latner et al., 2007 ; Rosenbaum & White, 2015 ), and metabolic illness ( Byrne et al., 2024 ). LOCE is also associated with many psychological vulnerabilities for mental illness, including disruptions in emotion regulation ( Burr et al., 2021 ; Stevenson et al., 2021 ), inhibitory control ( Ramalho et al., 2023 ), and distress tolerance ( Anestis et al., 2007 ; Burr et al., 2021 ). In addition, reward dysfunction may be implicated in the onset and maintenance of uncontrolled eating. Reward dysfunction has been most commonly evaluated in relation to objective binge eating, with results indicating that binge eating pathology is associated with greater valuation of food rewards ( Stojek & MacKillop, 2017 ) and a tendency to prefer smaller but more rapidly delivered rewards over larger but delayed rewards (i.e., increased delay discounting; Amlung et al., 2019 ; Lempert et al., 2019 ; Stojek & MacKillop, 2017 ). Additionally, neuroimaging findings reflect differences in reward pathways between individuals with binge eating and healthy controls ( Leenaerts et al., 2022 ). Recent findings indicate that frequency of LOCE may be associated with sensitivity to reward ( Burr, Dvorak, et al., 2024 ; Wierenga et al., 2014 ), suggesting the potential role of reward dysfunction in LOCE. However, reward models of LOCE outside of objective binge eating are under-evaluated and the role of food reward value has been largely ignored in this population. Overvaluation of a commodity may be influenced by psychological factors such as affective states or craving. Behavioral economics seeks to integrate principles of psychology and economics to understand behavior and decision making ( Rachlin et al., 2018 ), often by evaluating how the degree of reinforcement or reward derived from a commodity may be predictive of problematic health outcomes ( Kaplan et al., 2018 ). This framework has been primarily applied in the context of substance use, to beneficial effect (Buscemi et al., 2015). One prominent behavioral economics construct that has been extensively utilized to evaluate reward value in this context is demand. Demand aims to quantify an individual’s valuation for the acquisition and subsequent consumption of a specific commodity (i.e., the commodity’s reward value). The extant literature has predominantly examined demand or reward value in the context of commonly used substances including, but not limited to, cigarettes, alcohol, and illicit drugs ( Gex et al., 2022 ; Payne et al., 2020 ; Reed et al., 2020 ). Specifically, studies have revealed significant associations between alcohol demand, negative mood states, and increased craving for alcohol ( Ellis & Grekin, 2021 ; Meshesha et al., 2020 ; Tripp et al., 2015 ), which may translate to increased subsequent substance use and related problems ( Murphy et al., 2013 ; Soltis et al., 2017 ). Notably, recent research has examined reward value in the context of alternative commodities that may also produce elevated levels of reward such as the consumption of highly palatable foods ( Balantekin et al., 2020 ; Stojek & MacKillop, 2017 ). Thus, food reward value refers to an individual’s level of willingness to allocate valuable resources (e.g., time, money) to obtaining and subsequently consuming specific foods. To assess demand or reward value for food, researchers utilize a Food Purchase Task (FPT) derived from prior hypothetical purchase task structures such as the Cigarette Purchase Task ( Jacobs & Bickel, 1999 ) and Alcohol Purchase Task ( Murphy & MacKillop, 2006 ). The FPT requires respondents to indicate the amount of food they would purchase and consume across a series of incremental hypothetical prices. Responses on the FPT are organized into five unique indices of food demand: intensity (i.e., the number of units of the commodity that would be purchased or consumed when the commodity is priced at $0.00), O max (i.e., the maximum total amount that one is willing to spend on the commodity), breakpoint (i.e., the unit price in which consumption ends), P max (i.e., the price point associated with maximum expenditure), and elasticity (i.e., the sensitivity of consumption to increases in price) ( Kaplan et al., 2018 ; Kiselica et al., 2016 ). Importantly, previous research investigating food demand has revealed significant associations between increased levels of food demand and increased risk for eating disorders characterized by binge eating ( Stojek & MacKillop, 2017 ). However, few studies have investigated food demand in relation to LOCE more broadly. Further, the methodological approaches employed in the existing literature are primarily cross-sectional in nature and thus, are limited in their ability to rigorously evaluate food reward value as a potential maintenance factor for LOCE by testing their proposed temporal associations ( Epstein et al., 2018 ; Stojek & MacKillop, 2017 ). Additional research aimed at assessment of food demand and pathological eating using more ecologically valid methodologies and examination of more diverse eating pathology are needed. Importantly, behavioral economic frameworks acknowledge that the reward value of a commodity may be associated with psychological factors such as affective states or craving (MacKillop et al., 2012), and this principle has been demonstrated in the addiction literature. Specifically, studies have revealed positive associations between alcohol demand, negative mood and positive mood states, and increased craving for alcohol ( Ellis & Grekin, 2021 ; Meshesha et al., 2020 ; Tripp et al., 2015 ), which may translate to increased subsequent substance use and related problems ( Murphy et al., 2013 ; Soltis et al., 2017 ). These relationships may also extend to other potentially hazardous health behaviors, such as eating pathology. For example, a recent model of binge-eating maintenance (the Affect, Reward, and Cognition or ARC model; Schaefer et al., 2023 ) posits that food reward value fluctuates across the course of the day, with increased valuation of highly palatable foods being potentiated by aversive affective states. The model further proposes that once these dysfunctional reward processes are activated, individuals will be more likely to engage in problematic eating behavior (e.g., LOCE). Studies using ecological momentary assessment (EMA) and daily diary methods, which involve the repeated assessment of participants’ experiences multiple times per day in naturalistic settings, consistently demonstrate that increased negative affect is a significant predictor of subsequent uncontrolled eating, including LOCE ( Burr et al., 2023 ; Haedt-Matt & Keel, 2011 ; Schaefer et al., 2020 ; Stevenson et al., 2018 ). There is also evidence that positive affect may contribute to LOCE and food reward value via reward expectancies ( Burr et al., 2023 ). Further, prior research on reward processes in alcohol and drug use has also found that increased positive affect is a reliable influence on use urges ( Schlauch et al., 2013 ). However, no study has used EMA to test the hypothesis that momentary affect predicts later increases in the reward value of food, or that momentary elevations in the reward value of food increase risk for LOCE. Food craving may also mediate the momentary relationship between aversive affective states, food reward value, and risk for LOCE. Negative affect is associated with subsequent food craving ( Leenaerts et al., 2023 ; Sagui-Henson et al., 2021 ; Smith et al., 2021 ; Waters et al., 2001 ), and food craving has been found to mediate the association between affect and binge eating ( Leenaerts et al., 2023 ). Additionally, food craving may be associated with variations in food demand. A recent study found that when using the Food Purchase Task (FPT), higher food reward was related to higher food craving and more binge eating problems ( García-Pérez et al., 2023 ). As noted above, affect has been linked to subsequent substance use reward but has never been examined in the context of food reward. Presumably, variations in food reward may provide the link between affect and craving to promote motivated behavior. However, no study has yet evaluated the momentary prospective associations between affect, food reward value, craving, and LOCE. Therefore, the aim of the present study was to examine the prospective associations between positive and negative affect (t-2), food demand (t-1), food craving (t-1) and LOCE (t) using EMA protocol conducted among individuals with recurrent LOCE episodes. The likelihood of an eating episode to be considered LOCE is hypothesized to be greater when craving is elevated at prior assessment, due to increased momentary food demand elicited by elevated negative affect, in line with the affect regulation and ARC models of binge eating. In addition, positive affect has been shown to be a vulnerability to LOCE in the presence of reward dysfunction and thus, positive affect is thought to be predictive of LOCE when food demand is elevated temporally. Hypotheses were assessed using an EMA protocol in a general population sample that reported recurrent LOCE episodes. Use of EMA will allow for establishing causal relationships between variables in ecological settings. Prior investigations of reward have often relied on participant’s retrospective accounts of reward or laboratory studies ( Zvorsky et al., 2019 ). Retrospective self-reports, while useful in many instances, are subject to recall bias and accuracy of data may not be fully reliable ( Van den Bergh & Walentynowicz, 2016 ), particularly for constructs that fluctuate (e.g., affect, demand, or craving, Shiffman et al., 2008 ). Additionally, laboratory findings are of utility in establishing internally valid findings, but these findings must be replicated in the real world for generalizability. Utilization of EMA allows for improved understanding of variability of reward valuation not typically obtained in prior investigations. Finally, although affective and reward mechanisms are being explored in binge eating literature, consideration of food demand is not well represented, and LOCE regardless of food quantity even less so. Given that LOCE is implicated in deleterious outcomes, this is an important area of exploration. Therefore, these findings help to fill a gap in the extant literature that can be applied to both eating disorder diagnosis and general disordered eating. Method Procedure This project consisted of three phases run across 2023 - 2024: a screening phase via online survey, a virtual visit to confirm eligibility and orient participants to the EMA protocol, and a 10-day EMA participation period. This study was approved via the IRB at the institution at which it took place. Participants were compensated for their involvement via online gift card and treated in accordance with APA Ethical Guidelines ( American Psychological Association, 2017 ) IRB standards per CITI institutional guidelines. Online Screening First, an online screener was used to recruit participants, following their agreement to participate on a virtual consent form. This survey was advertised via Facebook and Instagram, limited to accounts for individuals 18 and older and with geolocation set in the United States. Individuals were considered eligible to be invited to a virtual lab visit if they reported they were 18 years of age or older, resided in the United States, and reported a minimum of two LOCE episodes over the past two weeks. Participants who were eligible and provided their contact information were randomly selected to be invited to participate in the virtual lab visit. Of 1683 individuals who completed the survey, 142 were invited to the laboratory visit. Virtual Laboratory Visit Participants were consented by both signing a consent form they review online and with a verbal summary of all project procedures and terms with a member of research personnel. Following consent, modules to assess eating disorders, mood disorders, and psychotic disorders were administered from the Structured Clinical Interview for DSM-5, Research Version (SCID-5-RV). Participants were excluded based on past-month active suicidality or active psychosis, or if they did not report at least two LOCE episodes in the past two weeks during the interview. 107 participants engaged in the virtual lab visit, of which 78 were considered eligible to participate. Eligible participants were then oriented to the EMA protocol and compensated for their visit. Zero eligible individuals dropped out or opted not to engage in the EMA study phase. Ecological Momentary Assessment Protocol The EMA protocol lasted ten days. Semi-random EMA signals were delivered to participants’ smartphones via the iNSIGHT platform ( Dvorak et al., 2023 ) five times daily with one signal per three hour window between 8:00 and 23:00 in a participant’s time zone. Participants were also instructed to self-initiate pre-eating surveys right before eating and post-eating surveys immediately following eating episodes. Signaled survey completion was used to calculate compliance rates. Participants The final sample consisted of 78 individuals, 87.18% of which were female. The sample was predominantly White (71.79%), 10.26% Black/ African American, 8.97% Asian, 9.00% American Indian/ Alaskan Native, and 1.28% Other. A small proportion of the sample identified as multi-racial (7.69%). Among the multi-racial group, 1.28% identified as White and Black or African American, 2.56% identified as White and American Indian or Alaskan Native, and 3.85% identified as White and Asian. Per structured assessment, 53.28% of the sample did not meet criteria for an eating disorder, 37.18% met criteria for binge eating disorder, and 9.00% met criteria for bulimia nervosa. Table 1 highlights all descriptive statistics. Table 1. Descriptive Statistics Descriptive Statistics % Mean SD Range Demographic Variables Age -- 44.54 11.74 18 - 66 Sex (female) 87.18% -- -- Race White/ Caucasian 71.79% -- -- -- Black/African American 10.26% -- -- -- Asian 8.97% -- -- -- American Indian/ Alaskan Native 9.00% -- -- -- Native Hawaiian/ Pacific Islander 0.00% -- -- -- Other 1.28% -- -- -- Multi-racial 7.69% Ethnicity (Hispanic) 6.41% -- -- LOCES 4.64 3.05 0.79 1.25 – 5.00 SCID-5-RV Eating Disorder Diagnosis Binge Eating Disorder 37.18% -- -- -- Bulimia Nervosa 9.00% -- -- -- No Eating Disorder Diagnosis 53.28% Between-Subjects Momentary Predictors Positive Affect -- 1.91 1.33 0 - 6 Negative Affect -- 1.46 1.39 0 - 6 Intensity -- 3.91 1.40 0 - 50 Breakpoint -- 2.29 1.99 0 - 10 O max -- 6.47 12.01 0 – 100 Craving -- 2.311 1.91 0 - 7 LOCE (yes) 7.58 -- -- -- Open in a new tab Measures Online Screening Demographic Information. Age, sex assigned at birth, race, ethnicity, body mass index, and US state of residence were reported by the participant in the online screener survey. Gender identity and sexual orientation were also recorded if the participant felt comfortable doing so, in accordance with the American Psychological Association’s 2016 resolution on Data About Sexual Orientation and Gender Identity ( American Psychological Association, 2016 ). LOCE frequency. LOCE frequency was assessed with two items assessing number of LOCE episodes per week over the last two weeks. These have been used in prior studies to screen for LOCE in EMA participants ( Burr et al., 2022 ). Participants had to endorse at least one LOCE per week on average to be eligible to be invited to the virtual visit. Loss of Control Eating Scale (LOCES). The LOCES ( Latner et al., 2014 ) is a 24-item validated assessment of subjective uncontrolled eating over the past four weeks. Participants respond to items using a 1 (“never”) to 5 (“very often”) Likert-type scale, with greater average scale scores indicating higher LOCE pathology. Internal consistency of the LOCES was excellent in this sample, Cronbach’s α = .96. Virtual Laboratory Visit Structured Clinical interview for Diagnostic and Statistical Manual of Mental Disorders-5 Research Version (SCID-5-RV). The SCID-5-RV ( First et al., 2014 ) is a validated semi-structured clinician-administered interview designed to assess for psychological disorders based on DSM-5 criteria. The Psychosis and Mood Disorders Current Major Depressive Disorder modules were both administered to assess for current psychosis and elevated suicide risk. The Eating Disorders module was administered to determine frequency of past three-month LOCE and probe disordered eating history. Ecological Momentary Assessment Protocol Momentary Affect. Current positive and negative affect were reported via eight items selected from the Positive Affect and Negative Affect Schedule (PANAS) and mood circumplex. The PANAS and mood circumplex items ( Watson & Clark, 1999 ) has been used in prior EMA studies to assess momentary affect in eating pathology ( Burr et al., 2023 ; Dvorak et al., 2024 ; Stevenson et al., 2018 ). Participants provided ratings of how much they were currently experiencing each affective state on a 0 (“not at all”) to 6 (“extremely”) scale. Items “happy” and “joyful” were combined and averaged to create the positive affect variable. These items had good internal consistency (Cronbach’s α = 0.91) with an ICC of 0.60. Ratings of items “anxious”, “nervous”, “sad”, “depressed”, “guilty” and “disappointed” were combined and averaged to create a global negative affect variable. Negative affect items showed good internal consistency (Cronbach’s α = 0.94, ICC = 0.80). Food Purchase Task. An ambulatory Food Purchase Task (FPT) was developed by this study team to assess momentary levels of food reward value in the random and pre-eating surveys. The ambulatory FPT was developed based on state versions of the Alcohol Purchase Task ( Amlung et al., 2015 ), which seek to assess three indices typically captured via standard purchase tasks. More specifically, the ambulatory FPT sought to assess momentary intensity, breakpoint, and O max for participants’ preferred snack foods. At each assessment, participants were prompted to select one of six highly palatable food options that they “want the most right now” (i.e., Reese’s Peanut Butter Cups, Nacho Cheese Doritos, Chips Ahoy Cookies, Mini Oreos, Little Debbie Zebra Cakes, Milk Chocolate M&Ms, Pringles Chips). They were next shown an image of a single 20-gram serving of the selected snack food (e.g., two Chips Ahoy cookies) and were asked to respond to three items. Intensity was assessed with the item “considering your selected snack food, right now, how many servings would you eat if each serving was free?” (ICC = 0.44). Breakpoint was assessed with the item “considering your selected snack food, right now, what is the maximum amount you would pay for a single serving?” ($0 - $50; ICC = 0.66), and O max was assessed with the item “considering your selected snack food, right now, what is the maximum total amount that you would spend to eat as much of this snack as you want?” ($0 - $100; ICC = 0.67). Food Craving. Craving (ICC = 0.39) was assessed with a single Likert-type item (“How much are you craving food right now?”). Participants rated current craving from 0 (“not at all”) to 6 (“extremely”) in both the random and pre-eating surveys. Craving as a construct can be successfully measured via a single momentary item, consistent with prior EMA research assessing craving in eating pathology ( Schaefer, Forester, Burr, et al., 2023 ). Momentary Loss of Control Eating. When administered in the random surveys, participants were first asked “have you eaten since your last random survey?” (“yes”/”no”) and then asked if they had completed their post-eating survey. If no, participants were asked follow-up questions regarding the eating episode, including “Did you have a sense of having lost control over your eating (at the time that you were eating)?” (“yes”/”no”). This item was adapted from the Eating Disorder Examination Questionnaire ( Fairburn et al., 2014 ), and has been used in prior studies to assess momentary or daily-level LOCE ( Burr et al., 2023 ; Stevenson et al., 2018 ). Participants were then asked to rate “how much did you feel you lost control over your eating?” and rated LOC 0 – 10. When questions were administered in the post-eating survey, LOCE probing was identical, but participants were not asked if they have eaten prior. The ICC of LOCE endorsements was 0.08. Data Preparation and Analysis The analysis included lagged associations between prior affect (t −2 ), current food demand and craving (t −1 ), and future loss of control eating (t). To do this, affect was lagged backward one assessment period while LOCE was pushed forward by one assessment period. Thus, there was always a missing observation at the first assessment for the day for affect and the last assessment for the day for LOCE. See Figure 1 for a depiction of lagged variables. There was also missing data throughout the dataset due to incomplete and/or missed items. Data missingness for all variables ranged from <1 to 29 percent. This included data from all three assessment types (random, pre-eating, and post-eating). All assessments included measures of affect (~1% missing). Random and pre-eating included behavioral economics measures (these were not assessed after eating; 29% missing). LOCE was assessed at random and post-eating assessments, but not pre-eating. We used multiple imputation from 100 imputed datasets to replace missing values. All model variables and covariates across all levels were used in the imputation process. The full model included 6048 momentary observations across 78 individuals. Correlations between the food reward indices were examined to determine if these should be included in a single model or in three individual models to account for multicollinearity. Figure 1. Open in a new tab Data Preparation Variable Lagging A two-level mediation model was specified in M plus v 8.11 using a Bayesian estimator. At the within-subjects level, t −2 positive affect and negative affect were specified as exogenous variables. Food demand metrics (intensity, breakpoint, and O max ) were conceptualized as initial mediators at t −1 and craving at t −1 was a sequential mediator. LOCE likelihood at time three was specified as the endogenous variable. Between-subjects, the same pathways were specified to compare mean associations across participants. At level one, day of week and time of day were controlled for as model covariates. The variance in all variables was parsed across levels by centering within-person (group mean) at the momentary level and between-person (grand mean) at the person level. The analysis uses pooled estimates across 100 imputed datasets to calculate model parameters. Data and code can be found on Open Science Framework: ( Burr, Meshesha, et al., 2024 ) https://osf.io/65ezw/?view_only=d0f12eead02f43b5a0505112398ebee2 . Results Compliance and Bivariate Statistics There was a total of 6048 observations available for analysis. Of these, participants received 3,900 random signals and completed 3,038 assessments (77.90% compliance). Participants self-initiated 1,571 pre-eating assessment ( m = 2.14 per day) and 1,439 post-eating assessments ( m = 1.85 per day) with 1,131 LOCE episodes reported (32.41% of eating episodes). There was no evidence of decrease in compliance across time ( r = −.02, p = .621). Compliance was uncorrelated with age ( r = −.06, p = .600) and modestly correlated with male biological sex ( r = .21, p = .062). The correlations varied considerably across indices. Breakpoint was highly correlated with intensity ( r within = .37, r between = .67). The correlation between Omax and intensity was quite low ( r within = .10, r between = .10) and the correlation between Omax and breakpoint was moderate ( r within = .22, r between = .14). Given this overall pattern of correlations, we opted to include all variables in the model. Primary Analysis Completing surveys on Tuesday was significantly associated with lower endorsements on breakpoint and O max metrics. Craving was higher on Mondays, Wednesdays, and Thursdays. Time of day had minimal but significant associations with LOCE ( B = 0.001, 95% BCI = −0.001, 0.000) and craving ( B = <0.001, 95% BCI = −0.001, 0.000), indicating craving and LOCE tended to occur later in the day. Age and sex were controlled for as level two model covariates, although they did not evidence significant associations with any mediator or endogenous variables. Direct associations between both positive and negative affect and LOCE were not significant in the model, indicating full mediation effects from positive and negative affect (t −2 ) through craving and BE metrics (t −1 ) to LOCE (t). Within-Subjects Effects See Figure 2 for a diagram of all model pathways. At the within-subjects level, t −1 reward metrics of breakpoint ( B = 0.186, 95% BCI = 0.089, 0.278), intensity ( B = 0.295, 95% BCI =−0.186, 0.404), and O max ( B = 1.117, 95% BCI = 0.599, 1.625) were significantly predicted by t −2 negative affect. Craving was associated with breakpoint ( B = 0.039, 95% BCI = 0.017, 0.061) and intensity ( B = 0.150, 95% BCI = 0.133, 0.169) but not O max . Craving significantly predicted higher likelihood of future LOCE ( B = 0.086, 95% BCI = 0.059, 0.114). Figure 2. Open in a new tab Model Direct Paths Note. Solid line indicates a statistically significant pathway between variables; dotted line indicates a non-significant pathway between variables. Regarding indirect and total effects, positive affect was only significantly associated with LOCE by way of craving (PA ➔ Craving ➔ LOCE), not mediated by any food demand variable. Conversely, negative affect was positively associated with LOCE both with a single mediation effect of craving (NA ➔ Craving ➔ LOCE) and by way of breakpoint (NA ➔ Breakpoint ➔ Craving ➔ LOCE) and intensity (NA ➔ Intensity ➔ Craving ➔ LOCE). See Table 2 for pathways and coefficients. This model accounted for 5% of the within-subjects variance in LOCE (4.6% of total variance in LOCE). Table 2. Within-Subjects Indirect Effects from Affect to LOCE Model Parameters Estimate SD 95% BCI Positive Affect Indirect Effects PA ➔ Craving ➔ LOCE * 0.004 0.002 >0.000, 0.009 PA ➔ Intensity ➔ Craving ➔ LOCE <0.001 0.001 −0.001, 0.002 PA ➔ Breakpoint ➔ Craving ➔ LOCE <0.001 <0.001 0.000, 0.001 PA ➔ O max ➔ Craving ➔ LOCE <0.001 <0.001 0.000, 0.000 Negative Affect Indirect Effects NA ➔ Craving ➔ LOCE * 0.009 0.001 0.004, 0.015 NA ➔ Intensity ➔ Craving ➔ LOCE * 0.004 0.001 0.002, 0.006 NA ➔ Breakpoint ➔ Craving ➔ LOCE * 0.001 <0.001 >0.000, 0.001 NA ➔ O max ➔ Craving ➔ LOCE <0.001 <0.001 0.000, 0.001 Total Effects Negative Affect ➔ LOCE * 0.014 0.003 0.008, 0.021 Positive Affect ➔ LOCE * 0.005 0.002 0.001, 0.010 Open in a new tab Note. PA = positive affect; NA = negative affect; BCI = Bayesian Credibility Interval * p < .05 Between-Subjects Effects See Figure 2 for a diagram of all model pathways. At the between-subjects level, average t-1 demand metrics of breakpoint ( B = 0.614, 95% BCI = 0.146, 1.083), and intensity ( B = 0.731, 95% BCI = 0.357, 1.107), were significantly predicted by negative affect. O max was not predicted by either positive or negative affect. Average craving was predicted by mean negative affect only ( B = 0.335, 95% BCI = 0.135, 1.534). Although mean craving was significantly associated with higher mean likelihood of LOCE, this association was negative, contrary to expectation ( B = −0.143, 95% BCI = 10.227, −0.063). As was found with momentary positive affect within-subjects, mean positive affect was unassociated with food demand or LOCE. This model accounted for 40% of the between-subject variance in LOCE (3.2% of total LOCE variance). See Table 3 for pathways and coefficients. Mean positive affect was not significantly associated with any of the other mean-level variables. Mean negative affect was only associated with mean LOCE by way of craving (NA ➔ Craving ➔ LOCE), although this association was negative, due to an unexpected negative relationship between mean craving and LOCE. Unlike the within-subjects results, mean food demand variables did not significantly mediate any pathway of the model. Table 3. Between-Subjects Indirect Effects from Affect to LOCE Model Parameters Estimate SD 95% BCI Positive Affect Indirect Effects PA ➔ Craving ➔ LOCE −0.026 0.002 −0.074, 0.013 PA ➔ Intensity ➔ Craving ➔ LOCE 0.002 0.007 −0.010, 0.020 PA ➔ Breakpoint ➔ Craving ➔ LOCE 0.001 0.005 −0.008, 0.015 PA ➔ O max ➔ Craving ➔ LOCE <0.001 0.003 −0.006, 0.007 Negative Affect Indirect Effects NA ➔ Craving ➔ LOCE * −0.057 0.022 −0.108, −0.021 NA ➔ Intensity ➔ Craving ➔ LOCE −0.017 0.013 −0.048, 0.002 NA ➔ Breakpoint ➔ Craving ➔ LOCE 0.002 0.008 −0.012, 0.020 NA ➔ O max ➔ Craving ➔ LOCE <0.001 0.003 −0.006, 0.008 Total Effects Negative Affect ➔ LOCE * −0.072 0.024 −0.127, −0.033 Positive Affect ➔ LOCE −0.021 0.022 −0.070, 0.019 Open in a new tab Note. PA = positive affect; NA = negative affect; BCI = Bayesian Credibility Interval; * p < .05 Posthoc Power Analysis The a priori power analysis utilized a minimum detectable effect (MDE) approach ( Snijders, 2005 ). We initially examined a range of ICCs (.10 to .80) and found that at n = 80 between subject observations and 4800 total assessments across participants, we would be sufficiently powered to detect effects of f 2 = .01 at the lowest ICC, f 2 = .08 at the highest ICC, and f 2 = .10 at the between-subjects level (effective sample size n = 77). While this was an accurate estimate of the upper ICC, our lower raw ICC was .39 (.34 in the final model). Though there are concerns about the use of posthoc power analyses that should be considered ( Althouse, 2021 ; Goodman & Berlin, 1994 ), we nonetheless conducted a posthoc power analysis to examine the minimum detectable effect using the revised ICCs. This analysis indicated that we were sufficiently powered to detect f 2 values of .03 (lowest ICC) to .07 (highest ICC). This indicated that three of our observed within-subject significant effects (NA➔Break Point: f 2 = .06 [MDE = .07 based on ICC of .67]; PA➔Craving: f 2 = .01 [MDE = .03 based on ICC of .34]; Break Point➔Craving: f 2 = .02 [MDE = .03 based on ICC of .34]) did not meet the MDE threshold for the observed ICC. Despite statistical significance, these effects were underpowered, suggesting a higher likelihood of Type I error or an overestimation of effect sizes. Thus, the direct paths from NA to Break Point, PA to Craving, and Break Point to Craving, as well the indirect paths from NA to Craving (via Break Point), NA to LOCE (via Break Point and Craving), and PA to LOCE (via Craving) while statistically significant, should be interpreted with a healthy dose of skepticism until replication of these findings. Discussion The present study sought to examine the temporal relationships between affect, food reward value, food craving, and LOCE using EMA. Specifically, we hypothesized that higher negative affect and positive affect would lead to elevated food demand metrics, which in turn would elevate craving and predict LOCE at a future time point. These hypotheses were partially supported. In analyses examining the momentary prospective associations between variables (i.e., the within-subjects level), negative affect predicted all three demand metrics (intensity, breakpoint, O max ). Further, higher breakpoint and intensity were associated with higher concurrent food craving, and higher food craving predicted increased subsequent risk for LOCE. Moreover, the indirect pathways from negative affect to LOCE were significant via intensity and breakpoint. Notably, food craving was not associated with O max from the concurrent timepoint and therefore the indirect relationship between negative affect and LOCE via O max was not supported. In hypothetical purchase task literature, O max has been identified as a variable that is inconsistent in predicting consumption ( Zvorsky et al., 2019 ). Results of the within-subjects model provide preliminary support for portions of the ARC model ( Schaefer et al., 2023 ), particularly the prospective association between increasing aversive affect and subsequent increases in food reward value, which relate to risk for later problematic eating patterns. However, future research is needed to test the direct effects of food reward value indices on risk for binge eating. Interestingly, between-subjects negative affect was only associated with LOCE via craving, bypassing any mean food demand metrics. Additionally, this small association was negative, due to an inverse relationship between mean craving and LOCE. Negative affect was positively associated with food craving, indicating individuals with higher mean negative emotional state also have high higher mean craving, as anticipated. It may be that for a sample of individuals who are all characterized by recurrent LOCE, these processes are only interpretable or salient at the momentary level. It is also possible that there may be differences attributable to duration of LOCE pathology. Incentive sensitization theory (IST) regards reinforcing mechanisms of reward processes for recurrent compulsive behavior, including binge eating, based on neurological findings ( Bodell & Racine, 2023 ; Werle et al., 2021 ). According to IST, “liking” (enjoying) and “wanting” (craving) are separate neurobiological pathways, with “liking” being more relevant for prediction of uncontrolled eating early in the pattern of recurrent binge episodes and “wanting” (i.e., craving) asserting greater influence in maintaining the pathology over time ( Bodell & Racine, 2023 ). Although IST has not been directly applied to uncontrolled eating outside of objective binges, it is therefore possible that there may be more individuals in the sample earlier in the experience of LOCE reinforcement. Future empirical inquiry is needed to confirm rationale for this unexpected finding or to assess whether this negative association is an aberration. In the within-subjects model, positive affect also evidenced significant links with LOCE but only by way of craving. Food demand endorsements were not predicted by previous positive affective state. At the between-subjects level, mean positive affect did not significantly predict any subsequent variables. Findings are generally in line with prior literature that suggests behavioral processes, such as motive strength, may create a significant link between positive affect and LOCE ( Burr et al., 2023 ). The literature regarding associations between positive affect and LOCE is mixed ( Dingemans et al., 2009 ; Selby et al., 2012 ; Stevenson et al., 2021 ; Wolff et al., 2000 ), despite reward processes being strongly implicated in increasing LOCE likelihood ( Schaefer, Forester, Dvorak, et al., 2023 ). It may be that the most direct perceived rewarding aspect of LOCE is not to influence positive affect but rather to alleviate negative affect, in line with the affect regulation theory ( Haedt-Matt & Keel, 2011 ) and that positive affect is only a vulnerable for LOCE under certain circumstances. All participants were expected to experience LOCE episodes due to recruitment targeting of those with recurrent LOCE. Therefore, average craving may not lead to highly variable outcomes in LOCE when participants are recruited specifically due to LOCE presence. It stands to reason that fluctuation within an individual’s craving experience is more likely to predict whether they engage in uncontrolled eating rather than the average craving they have. Craving is a highly individualistic experience: some individuals are conscious of craving food near constantly, while others have greater variability ( Davis, 2013 ; Ng & Davis, 2013 ). Therefore, one’s average craving compared to another is likely not as relevant to predicting food demand as fluctuations in their individual experience. The finding that negative affect, at the within-subjects level but not at the between-subjects level, predicted food craving and LOCE through demand can also be partially explained by the assessment of state versus trait demand. While measurements of demand remain generally stable ( Acuff & Murphy, 2017 ), they can be variable as a function of subjective states, environmental contexts, or interventions ( Aston & Cassidy, 2019 ). The current results found momentary negative affect influencing one’s willingness to consume more food when readily available (intensity) and willingness to purchase food at a higher price point prior to ceasing purchasing behavior (breakpoint). The FPT asked participants to report their demand “right now” and in contexts when an individual is high in negative affect, there was an increase in demand. Suggesting the association can be understood as a temporal dynamic between negative affect, demand, and craving rather than stable individual differences. While research using the FPT is still nascent, similar results were observed in studies using other commodity purchase tasks; a meta-analysis of experimental manipulation studies of alcohol and other purchase tasks found within-person increases in demand when individuals are experience negative affect and stress ( Acuff et al., 2020 ). Food demand ( García-Pérez et al., 2023 ) akin to alcohol demand ( Amlung et al., 2015 ; Marsden et al., 2023 ) is associated with higher cravings. Yet, prior studies have not examined the sequential dynamic between craving, demand, and other contextual factors. The current EMA data allowed for the examination of these sequential events in real time: negative affect at Time 1 predicted demand and food craving at Time 2 which predicted future LOCE at Time 3. These results begin to display the temporal precedence and processes involved in this behavioral outcome. This study is the first to provide conclusions regarding an ambulatory demand metric predicting LOCE, as well as adding to the scant research on LOCE to-date in terms of affective and craving predictors. Limitations and Constraints on Generality The present study is not without limitations. The model yielded 6048 observations across 78 individuals, which is an acceptable sample for EMA protocols ( Wrzus & Neubauer, 2023 ). For a multi-level model with serial mediation, small effect sizes would be anticipated given the parsing of variance. However, true strengths of infinitesimal effects in this sample may not be represented with the current sample size. The lagged effects here may actually be phenomena that occur closer in time. For instance, reward valuation and craving may be triggered closer to changes in affect. However, affect was assessed up to six hours before LOCE. Temporal precedence for some variables is established via current analyses, but at the cost of observation of variables that may co-occur (e.g., negative affect and craving may occur concurrently). Similarly, our effect sizes did not reach the level of “minimum detectable effects” for power of .80 given our observed ICCs. Posthoc power, while itself problematic, did suggest that some of our observed effects are underpowered and should be interpreted with caution. We would encourage future researchers to scrutinize these paths until replication of these associations has been confirmed. Methodologically, a potential limitation is that participants had finite options to choose among foods they were currently craving. Although highly palatable processed foods are typical foods of choice for LOCE ( Ayton et al., 2021 ; Shank et al., 2015 ), it is possible associations may have been stronger if participants had a food craving that was very different than those foods presented during the FPT. Finally, EMA is not a data collection procedure with 100% compliance. It is possible that participants may have had undocumented eating episodes. Study protocols attempted to mitigate this gap by allowing for retroactive reporting of eating episodes in random assessments if participants had not self-initiated pre-eating and post-eating surveys. However, compliance was considered adequate for an EMA study (77.90%; Shiffman et al., 2008 ). The limited sample size prohibits analysis of cross-racial and cross-gender comparisons. The sample was predominantly white (71%) and female (87%). Future research should oversample racial and ethnic minorities to assess for differences. The current sample was collected nationally, but still relied on recruitment of individuals via social media, meaning stratification of location was not achievable to equally represent all states. The age range was also restricted to only include adults, which also excludes any individuals under 18. Disordered eating is often onset in adolescence ( Rohde et al., 2015 ). Ergo, findings may represent a different developmental period both in terms of age and in terms of course of disordered eating in the current sample and generalizing to young individuals should be done with caution. Conclusion This study provides novel insight into the nature of food demand and conceptualizing reward in LOCE. This manuscript represents the first to the authors’ knowledge to assess a FPT in an ambulatory design, probing real-time variables that influence likelihood of uncontrolled eating. Additionally, findings provide support for the supposition that negative affect may be predictive of craving at a future time point. Food craving has conversely also been associated with future negative affect in binge eating disorder ( Schaefer, Forester, Burr, et al., 2023 ), suggesting that there may be a reciprocal relationship between the two given these findings in conjunction. Over the course of within-day momentary assessments, individuals were more likely to engage in LOCE when negative affect was higher by way of select food demand variables and craving. However, positive affect was only a vulnerability to LOCE by way of craving and bypassed food demand. Findings suggest that a behavioral economic approach to LOCE, conceptualizing uncontrolled eating as a behavior with a “trade off” or cost may only be relevant conceptually to participants in the context of negative affect. This finding is theoretically consistent with prior literature that links increases in negative affect and expectancies of reduced negative affect to LOCE. Therefore, these findings imply that expected “reward” of uncontrolled eating is alleviation of negative emotions rather than increasing positive affective states. Overall, findings suggest that negative affective, food reward value, and food craving may operate as predictors of subsequent LOCE episodes. Although numerous existing interventions target affect in the context of eating disorders (e.g., dialectical behavior therapy, integrative cognitive-affective therapy), continued research is needed to clarify the potential benefit of intervening on food reward processes and craving to help curb the likelihood of uncontrolled eating. Future research should use also laboratory paradigms to assess affect-demand-craving trajectories in closer proximity. Additionally, follow-up studies may use clinical populations to probe whether there are differences in these relationships from the current community-based sample. All told, findings provide a first step in a promising conceptualization of LOCE and lay preliminary evidence requiring much additional research to draw robust conclusions. Supplementary Material 3 NIHMS2071457-supplement-3.pdf (123.4KB, pdf) Acknowledgements: EKB is supported by funding from the National Institute of Mental Health (NIMH, F31MH135713-01A1). This research was supported by funding from the Academy for Eating Disorders to EKB. LZM is supported by the National Institute on Alcohol Abuse and Alcoholism (NIAAA, K23 AA028269) This research was supported by funding from the National Institute of General Medical Sciences (NIGMS, P20GM134969) to LMS, GF, and SAW. References Acuff SF, Amlung M, Dennhardt AA, MacKillop J, & Murphy JG (2020). Experimental manipulations of behavioral economic demand for addictive commodities: A meta-analysis. Addiction, 115(5), 817–831. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Acuff SF, & Murphy JG (2017). Further examination of the temporal stability of alcohol demand. Behavioural Processes, 141, 33–41. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Althouse AD (2021). Post hoc power: not empowering, just misleading. Journal of Surgical Research, 259, A3–A6. [ DOI ] [ PubMed ] [ Google Scholar ] American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th, text ed.). [ Google Scholar ] American Psychological Association. (2016). Resolution on Data about Sexual Orientation and Gender Identity. American Psychological Association. (2017). Ethical principles of psychologists and code of conduct (2002, amended effective June 1, 2010, and January 1, 2017. Amlung M, Marsden E, Holshausen K, Morris V, Patel H, Vedelago L, Naish KR, Reed DD, & McCabe RE (2019). Delay discounting as a transdiagnostic process in psychiatric disorders: A meta-analysis. JAMA Psychiatry, 76(11), 1176–1186. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Amlung M, McCarty KN, Morris DH, Tsai C-L, & McCarthy DM (2015). Increased behavioral economic demand and craving for alcohol following a laboratory alcohol challenge. Addiction, 110(9), 1421–1428. 10.1111/add.12897 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Anestis MD, Selby EA, Fink EL, & Joiner TE (2007). The multifaceted role of distress tolerance in dysregulated eating behaviors. International Journal of Eating Disorders, 40(8), 718–726. 10.1002/eat.20471 [ DOI ] [ PubMed ] [ Google Scholar ] Araujo DMR, Santos GFDS, & Nardi AE (2010). Binge eating disorder and depression: A systematic review. World Journal of Biological Psychiatry, 11(2 PART 2), 199–207. 10.3109/15622970802563171 [ DOI ] [ PubMed ] [ Google Scholar ] Aston ER, & Cassidy RN (2019). Behavioral economic demand assessments in the addictions. Current Opinion in Psychology, 30, 42–47. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Ayton A, Ibrahim A, Dugan J, Galvin E, & Wright OW (2021). Ultra-processed foods and binge eating: A retrospective observational study. Nutrition, 84, 111023. [ DOI ] [ PubMed ] [ Google Scholar ] Balantekin KN, Ziegler AM, Crandall AK, & Temple JL (2020). The relationships between eating disorder pathology and relative reinforcing value of food, delay discounting, and related constructs in adolescents. Appetite, 148, 104576. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Bodell LP, & Racine SE (2023). A mechanistic staging model of reward processing alterations in individuals with binge-type eating disorders. International Journal of Eating Disorders, 56(3), 516–522. [ DOI ] [ PubMed ] [ Google Scholar ] Brownstone LM, & Bardone-Cone AM (2021). Subjective binge eating: A marker of disordered eating and broader psychological distress. Eating and Weight Disorders-Studies on Anorexia, Bulimia and Obesity, 26, 2201–2209. [ DOI ] [ PubMed ] [ Google Scholar ] Burr EK, Dvorak RD, De Leon AN, Klaver SJ, Peterson R, Hayden ER, Maynard MH, Almeyda S, & Medina A (2024). Understanding Loss of Control Eating Through the Lens of Expectancies and Reinforcement Sensitization. Psychiatry, 1–16. 10.1080/00332747.2024.2403664 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Burr EK, Dvorak RD, De Leon AN, Leary AV, Peterson R, Schaefer LM, & Wonderlich SA (2023). The role of eating expectancies and eating motives in the association between mood and loss-of-control eating: A national sample daily diary study. Appetite, 180, 106322. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Burr EK, Dvorak RD, Peterson R, & De Leon AN (2022). Greater negative affect reduction expectancies moderate the interactive relationship between emotion regulation difficulties and distress tolerance in predicting loss-of-control eating. British Journal of Health Psychology, n/a (n/a). 10.1111/bjhp.12611 [ DOI ] [ PubMed ] [ Google Scholar ] Burr EK, Dvorak RD, Stevenson BL, Schaefer LM, & Wonderlich SA (2021). Ability to tolerate distress moderates the indirect relationship between emotion regulation difficulties and loss-of-control over eating via affective lability. Eating Behaviors, 43, 101561. 10.1016/j.eatbeh.2021.101561 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Burr EK, Meshesha LZ, Dvorak RD, Allen Q, Magri T, Wang C, Hayden ER, Rodriguez NE, Leary AV, Maynard M, Schaefer LM, Forester G, & Wonderlich SA (2024, December 9). Using Behavioral Economics to Understand Reinforcement Mechanisms of Loss-of-Control Eating: An Ecological Momentary Assessment Approach. Retrieved from osf.io/65ezw [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Byrne ME, Shank LM, Lavender JM, Higgins-Neyland MK, Rice A, Sweeney RS, Norton C, Haigney M, Yanovski JA, & Tanofsky-Kraff M (2024). A systematic review and meta-analysis of loss-of-control eating in relation to cardiometabolic health components and inflammatory markers. Obesity Reviews, n/a (n/a), e13805. 10.1111/obr.13805 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Davis C. (2013). Compulsive overeating as an addictive behavior: overlap between food addiction and binge eating disorder. Current Obesity Reports, 2, 171–178. [ Google Scholar ] Dingemans AE, Martijn C, van Furth EF, & Jansen ATM (2009). Expectations, mood, and eating behavior in binge eating disorder. Beware of the bright side. Appetite, 53(2), 166–173. 10.1016/j.appet.2009.06.002 [ DOI ] [ PubMed ] [ Google Scholar ] Dvorak RD, Finch M, Allen Q, Toth E, Burr EK, & De Leon AN (2023). iNSIGHT: Investigation using Naturalistic Sampling for Individual and Group Health Tracking. [ Google Scholar ] Dvorak RD, Paulson D, Dunn ME, Burr EK, Peterson R, Maynard M, De Leon AN, Klaver SJ, Leary AV, Hayden ER, Allen Q, & Toth E (2024). Effects of medical cannabis use on physical and psychiatric symptoms across the day among older adults. Psychiatry Research, 339, 116055. 10.1016/j.psychres.2024.116055 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Ellis JD, & Grekin ER (2021). The effects of anger on alcohol demand and subjective alcohol craving. Alcoholism: Clinical and Experimental Research, 45(2), 436–445. [ DOI ] [ PubMed ] [ Google Scholar ] Epstein LH, Paluch RA, Carr KA, Temple JL, Bickel WK, & MacKillop J (2018). Reinforcing value and hypothetical behavioral economic demand for food and their relation to BMI. Eating Behaviors, 29, 120–127. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Fairburn C, Cooper Z, & O’Conner M (2014). Eating Disorder Examination (17th ed.). [ Google Scholar ] First MB, Williams JBW, Karg RS, & Spitzer RL (2014). Structured clinical interview for DSM-5 disorders–research version (SCID-5-RV). Arlington: American Psychiatric Assocation. [ Google Scholar ] García-Pérez Á, García-Fernández G, Krotter A, González-Roz A, Martínez-Loredo V, & Secades-Villa R (2023). Validation of the Food Purchase Task (FPT) in a clinical sample of smokers with overweight and obesity. Appetite, 185, 106549. [ DOI ] [ PubMed ] [ Google Scholar ] Gex KS, Acuff SF, Campbell KW, Mun E, Dennhardt AA, Borsari B, Martens MP, & Murphy JG (2022). Change in alcohol demand following a brief intervention predicts change in alcohol use: A latent growth curve analysis. Alcoholism: Clinical and Experimental Research, 46(8), 1525–1538. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Goodman SN, & Berlin JA (1994). The use of predicted confidence intervals when planning experiments and the misuse of power when interpreting results. Annals of Internal Medicine, 121(3), 200–206. [ DOI ] [ PubMed ] [ Google Scholar ] Goossens L, Soenens B, & Braet C (2009). Prevalence and characteristics of binge eating in an adolescent community sample. Journal of Clinical Child & Adolescent Psychology, 38(3), 342–353. [ DOI ] [ PubMed ] [ Google Scholar ] Haedt-Matt AA, & Keel PK (2011). Revisiting the affect regulation model of binge eating: A meta-analysis of studies using ecological momentary assessment. Psychological Bulletin, 137(4), 660–681. 10.1037/a0023660 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Jacobs EA, & Bickel WK (1999). Modeling drug consumption in the clinic using simulation procedures: demand for heroin and cigarettes in opioid-dependent outpatients. Experimental and Clinical Psychopharmacology, 7(4), 412. [ DOI ] [ PubMed ] [ Google Scholar ] Kaplan BA, Foster RNS, Reed DD, Amlung M, Murphy JG, & MacKillop J (2018). Understanding alcohol motivation using the alcohol purchase task: A methodological systematic review. Drug and Alcohol Dependence, 191, 117–140. [ DOI ] [ PubMed ] [ Google Scholar ] Kiselica AM, Webber TA, & Bornovalova MA (2016). Validity of the alcohol purchase task: a meta-analysis. Addiction, 111(5), 806–816. 10.1111/add.13254 [ DOI ] [ PubMed ] [ Google Scholar ] Latner JD, Hildebrandt T, Rosewall JK, Chisholm AM, & Hayashi K (2007). Loss of control over eating reflects eating disturbances and general psychopathology. Behaviour Research and Therapy, 45(9), 2203–2211. 10.1016/j.brat.2006.12.002 [ DOI ] [ PubMed ] [ Google Scholar ] Latner JD, Mond JM, Kelly MC, Haynes SN, & Hay PJ (2014). The loss of control over eating scale: Development and psychometric evaluation. International Journal of Eating Disorders, 47(6), 647–659. 10.1002/eat.22296 [ DOI ] [ PubMed ] [ Google Scholar ] Leenaerts N, Jongen D, Ceccarini J, Van Oudenhove L, & Vrieze E (2022). The neurobiological reward system and binge eating: A critical systematic review of neuroimaging studies. International Journal of Eating Disorders, 55(11), 1421–1458. 10.1002/eat.23776 [ DOI ] [ PubMed ] [ Google Scholar ] Leenaerts N, Vaessen T, Sunaert S, Ceccarini J, & Vrieze E (2023). How negative affect does and does not lead to binge eating—The importance of craving and negative urgency in bulimia nervosa. Journal of Psychopathology and Clinical Science. 10.1037/abn0000830 [ DOI ] [ PubMed ] [ Google Scholar ] Lempert KM, Steinglass JE, Pinto A, Kable JW, & Simpson HB (2019). Can delay discounting deliver on the promise of RDoC? Psychological Medicine, 49(2), 190–199. [ DOI ] [ PubMed ] [ Google Scholar ] Marsden E, Murphy JG, MacKillop J, & Amlung M (2023). Alcohol cues increase behavioral economic demand and craving for alcohol in nontreatment-seeking and treatment-seeking heavy drinkers. Alcohol: Clinical and Experimental Research, 47(11), 2149–2160. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Meshesha LZ, Aston ER, Teeters JB, Blevins CE, Battle CL, Marsh E, Feltus S, Stein MD, & Abrantes AM (2020). Evaluating alcohol demand, craving, and depressive symptoms among women in alcohol treatment. Addictive Behaviors, 109, 106475. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Murphy JG, & MacKillop J (2006). Relative reinforcing efficacy of alcohol among college student drinkers. Experimental and Clinical Psychopharmacology, 14(2), 219. [ DOI ] [ PubMed ] [ Google Scholar ] Murphy JG, Yurasek AM, Dennhardt AA, Skidmore JR, McDevitt-Murphy ME, MacKillop J, & Martens MP (2013). Symptoms of depression and PTSD are associated with elevated alcohol demand. Drug and Alcohol Dependence, 127(1–3), 129–136. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Ng L, & Davis C (2013). Cravings and food consumption in binge eating disorder. Eating Behaviors, 14(4), 472–475. 10.1016/j.eatbeh.2013.08.011 [ DOI ] [ PubMed ] [ Google Scholar ] Organization, W. H. (2018). International classification of diseases for mortality and morbidity statistics (11th Revision). Payne J, Manning M, Fleming C, & Pham H-T (2020). The price elasticity of demand for illicit drugs: A systematic review. Trends and Issues in Crime and Criminal Justice, 606, 1–19. [ Google Scholar ] Rachlin H, Green L, Vanderveldt A, & Fisher EB (2018). Behavioral medicine’s roots in behaviorism: Concepts and applications. Principles and Concepts of Behavioral Medicine: A Global Handbook, 241–275. [ Google Scholar ] Ramalho SM, Conceição E, Tavares AC, Freitas AL, Machado BC, & Gonçalves S (2023). Loss of control over eating, inhibitory control, and reward sensitivity in children and adolescents: a systematic review. Nutrients, 15(12), 2673. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Reed DD, Naudé GP, Salzer AR, Peper M, Monroe-Gulick AL, Gelino BW, Harsin JD, Foster RNS, Nighbor TD, & Kaplan BA (2020). Behavioral economic measurement of cigarette demand: A descriptive review of published approaches to the cigarette purchase task. Experimental and Clinical Psychopharmacology, 28(6), 688. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Rohde P, Stice E, & Marti CN (2015). Development and predictive effects of eating disorder risk factors during adolescence: Implications for prevention efforts. International Journal of Eating Disorders, 48(2), 187–198. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Rosenbaum DL, & White KS (2015). The relation of anxiety, depression, and stress to binge eating behavior. Journal of Health Psychology, 20(6), 887–898. 10.1177/1359105315580212 [ DOI ] [ PubMed ] [ Google Scholar ] Sagui-Henson SJ, Radin RM, Jhaveri K, Brewer JA, Cohn M, Hartogensis W, & Mason AE (2021). Negative Mood and Food Craving Strength Among Women with Overweight: Implications for Targeting Mechanisms Using a Mindful Eating Intervention. Mindfulness, 12, 2997–3010. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schaefer LM, Forester G, Burr EK, Laam L, Crosby RD, Peterson CB, Crow SJ, Engel SG, Dvorak RD, & Wonderlich SA (2023). Examining the role of craving in affect regulation models of binge eating: Evidence from an ecological momentary assessment study. Journal of Psychopathology and Clinical Science. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schaefer LM, Forester G, Dvorak RD, Steinglass J, & Wonderlich SA (2023). Integrating aspects of affect, reward, and cognition to develop more comprehensive models of binge-eating pathology. International Journal of Eating Disorders, 56(8), 1502–1510. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schaefer LM, Smith KE, Anderson LM, Cao L, Crosby RD, Engel SG, Crow SJ, Peterson CB, & Wonderlich SA (2020). The role of affect in the maintenance of binge-eating disorder: Evidence from an ecological momentary assessment study. In Journal of Abnormal Psychology (Vol. 129, Issue 4, pp. 387–396). American Psychological Association. 10.1037/abn0000517 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schlauch RC, Gwynn-Shapiro D, Stasiewicz PR, Molnar DS, & Lang AR (2013). Affect and craving: Positive and negative affect are differentially associated with approach and avoidance inclinations. Addictive Behaviors, 38(4), 1970–1979. 10.1016/j.addbeh.2012.12.003 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Selby EA, Doyle P, Crosby RD, Wonderlich SA, Engel SG, Mitchell JD, & Le Grange D (2012). Momentary emotion surrounding bulimic behaviors in women with bulimia nervosa and borderline personality disorder. Journal of Psychiatric Research, 46(11), 1492–1500. 10.1016/j.jpsychires.2012.08.014 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Shank LM, Tanofsky-Kraff M, Nelson EE, Shomaker LB, Ranzenhofer LM, Hannallah LM, Field SE, Vannucci A, Bongiorno DM, Brady SM, Condarco T, Demidowich A, Kelly NR, Cassidy O, Simmons WK, Engel SG, Pine DS, & Yanovski JA (2015). Attentional bias to food cues in youth with loss of control eating. Appetite, 87, 68–75. 10.1016/j.appet.2014.11.027 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Shiffman S, Stone AA, & Hufford MR (2008). Ecological momentary assessment. Annual Review of Clinical Psychology, 4, 1–32. 10.1146/annurev.clinpsy.3.022806.091415 [ DOI ] [ PubMed ] [ Google Scholar ] Smith KE, Mason TB, Schaefer LM, Anderson LM, Critchley K, Crosby RD, Engel SG, Crow SJ, Wonderlich SA, & Peterson CB (2021). Dynamic Stress Responses and Real-Time Symptoms in Binge-Eating Disorder. Annals of Behavioral Medicine, 55(8), 758–768. 10.1093/abm/kaaa061 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Snijders TAB (2005). Power and sample size in multilevel modeling. Encyclopedia of Statistics in Behavioral Science, 3(157), 1573. [ Google Scholar ] Soltis KE, McDevitt-Murphy ME, & Murphy JG (2017). Alcohol demand, future orientation, and craving mediate the relation between depressive and stress symptoms and alcohol problems. Alcoholism: Clinical and Experimental Research, 41(6), 1191–1200. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Stevenson BL, Dvorak RD, Wonderlich SA, Crosby RD, & Gordon KH (2018). Emotions before and after loss of control eating. Eating Disorders, 26(6), 505–522. 10.1080/10640266.2018.1453634 [ DOI ] [ PubMed ] [ Google Scholar ] Stevenson BL, Wilborn D, Kramer MP, & Dvorak RD (2021). Real-time changes in emotion regulation and loss of control eating. Journal of Health Psychology. 10.1177/1359105318823242 [ DOI ] [ PubMed ] [ Google Scholar ] Stojek MMK, & MacKillop J (2017). Relative reinforcing value of food and delayed reward discounting in obesity and disordered eating: A systematic review. Clinical Psychology Review, 55, 1–11. [ DOI ] [ PubMed ] [ Google Scholar ] Tripp JC, Meshesha LZ, Teeters JB, Pickover AM, McDevitt-Murphy ME, & Murphy JG (2015). Alcohol craving and demand mediate the relation between posttraumatic stress symptoms and alcohol-related consequences. Experimental and Clinical Psychopharmacology, 23(5), 324. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Van den Bergh O, & Walentynowicz M (2016). Accuracy and bias in retrospective symptom reporting. Current Opinion in Psychiatry, 29(5), 302–308. [ DOI ] [ PubMed ] [ Google Scholar ] Waters A, Hill A, & Waller G (2001). Bulimics’ responses to food cravings: is binge-eating a product of hunger or emotional state? Behaviour Research and Therapy, 39(8), 877–886. 10.1016/S0005-7967(00)00059-0 [ DOI ] [ PubMed ] [ Google Scholar ] Watson D, & Clark LA (1999). The PANAS-X: Manual for the positive and negative affect schedule-expanded form. [ Google Scholar ] Werle D, Schroeder PA, Wolz I, & Svaldi J (2021). Incentive sensitization in binge behaviors: A mini review on electrophysiological evidence. Addictive Behaviors Reports, 13, 100344. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Wierenga CE, Ely A, Bischoff-Grethe A, Bailer UF, Simmons AN, & Kaye WH (2014). Are extremes of consumption in eating disorders related to an altered balance between reward and inhibition? Frontiers in Behavioral Neuroscience, 8, 410. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Wolff GE, Crosby RD, Roberts JA, & Wittrock DA (2000). Differences in daily stress, mood, coping, and eating behavior in binge eating and nonbinge eating college women. Addictive Behaviors, 25(2), 205–216. 10.1016/S0306-4603(99)00049-0 [ DOI ] [ PubMed ] [ Google Scholar ] Wong MM, Craun EA, Bravo AJ, & Pearson MR (2019). Insomnia symptoms, cannabis protective behavioral strategies, and hazardous cannabis use among U.S. college students. In Experimental and Clinical Psychopharmacology (Vol. 27, pp. 309–317). American Psychological Association. 10.1037/pha0000273 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Wrzus C, & Neubauer AB (2023). Ecological momentary assessment: A meta-analysis on designs, samples, and compliance across research fields. Assessment, 30(3), 825–846. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Zvorsky I, Nighbor TD, Kurti AN, DeSarno M, Naudé G, Reed DD, & Higgins ST (2019). Sensitivity of hypothetical purchase task indices when studying substance use: A systematic literature review. Preventive Medicine, 128, 105789. 10.1016/j.ypmed.2019.105789 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials 3 NIHMS2071457-supplement-3.pdf (123.4KB, pdf) ACTIONS View on publisher site PDF (592.6 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top