CoChat
Sign Up
Cited by 64Innovative Human-Technology InteractionMobile Health and mHealth ApplicationsDigital Mental Health Interventions

Food and Mood: Just-in-Time Support for Emotional Eating

Carroll, Erin A., Czerwinski, Mary, Roseway, Asta, Kapoor, Ashish, Johns, Paul, Rowan, Kael, Schraefel, M.C.·2013·2013 Humaine Association Conference on Affective Computing and Intelligent Interaction
DOI 10.1109/acii.2013.48
Open this paper in CoChat — free account

Abstract

Behavior modification in health is difficult, as habitual behaviors are extremely well-learned, by definition. This research is focused on building a persuasive system for behavior modification around emotional eating. In this paper, we make strides towards building a just-in-time support system for emotional eating in three user studies. The first two studies involved participants using a custom mobile phone application for tracking emotions, food, and receiving interventions. We found lots of individual differences in emotional eating behaviors and that most participants wanted personalized interventions, rather than a pre-determined intervention. Finally, we also designed a novel, wearable sensor system for detecting emotions using a machine learning approach. This system consisted of physiological sensors which were placed into women's brassieres. We tested the sensing system and found positive results for emotion detection in this mobile, wearable system.

AI snapshot

Written from the paper’s full text.

TL;DR

This paper presents three user studies toward a just-in-time support system for emotional eating: tracking emotional eating patterns, testing a deep-breathing intervention, and detecting emotions with wearable physiological sensors. It finds individual differences and a desire for personalized interventions, and demonstrates above-chance wearable emotion classification (arousal 75.00%, valence 72.62%).

Key findings

  • In Study 1 (N=12, 2 males), most participants had eating events in all four quadrants of the Circumplex model; 6 of 12 tended to eat predominantly when stressed (upper-left quadrant: negative & pumped), and few eating events occurred in the calm/serene lower-right quadrant.
  • In Study 2 (N=11, 1 male), all participants reported becoming more aware of their eating behaviors and 87.5% reported becoming more aware of their emotions, but only 37.5% reported their eating behaviors changed after Study 1. Intervention trigger accuracy was M=2.80 (SD=1.30); deep breathing effectiveness was M=3.0 (SD=1.87) for reducing stress and M=2.20 (SD=1.30) for preventing emotional eating; 3 of 11 participants were enthusiastic about deep breathing, while others wanted alternatives and personalization.
  • In Study 3, four women wore a bra-based EKG/EDA sensor system for about 4–6 hours a day over approximately four days; using 40 features and Gaussian Process Regression with leave-one-example-out, the system classified arousal at 75.00% and valence at 72.62% accuracy, significantly better than chance and at par with other affect recognition systems.

Methods

  • Study 1 had 12 self-identified emotional eaters use the EmoTree mobile phone application to self-report emotions hourly and log food over four days, with a pre-experiment Emotional Eating Scale survey. Study 2 used the same participants except one withdrawal (N=11) over four days, adding a deep-breathing intervention triggered by self-reported stress/anxiety and pre/post surveys.
  • Study 3 had four women wear a custom wearable system with GRASP boards capturing EKG, EDA, 3-axis accelerometer, and 2-axis gyroscope; EmoTree self-reports served as labeled ground truth. Sensor data were filtered, normalized, and converted into 40 features (8 signals × 5 bands), then classified with Gaussian Process Regression using leave-one-example-out.
  • Analysis included per-person scatterplots of emotional ratings and food-associated ratings, survey means, and machine learning classification accuracy.

Limitations

  • The paper states that Study 3’s brassiere form factor only allowed women to participate; participants had to recharge boards every 3–4 hours and found the prototype tedious; EDA was collected from a non-standard under-breast location; and classification used log files rather than real-time sensor data. The authors also note the need for a robust real-world system addressing battery life, comfort, and suitability for both men and women.
  • The paper states that the stress/anxiety trigger was not highly accurate (M=2.80) and deep breathing was only moderately effective at reducing stress and less effective at preventing emotional eating; most participants wanted personalized interventions rather than a single pre-determined one.
  • Evident from the text: samples were very small (N=12, N=11, N=4), study durations were short (four days), emotion labels relied on self-report, and no control condition is reported, limiting generalizability.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1109/acii.2013.48

Cites

23 of 23 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 18 cited sources checked · 0 supported · 0 flagged

  • Findingown data
    § A. Analysis and Results · p. 3

    In Study 1, pre-survey results on the Emotional Eating Scale showed that the emotions marked as causing a moderate to strong urge to eat included: on edge, nervous, upset, worried, nothing to do, bored, irritated, restless, and discouraged.

    “Our pre-survey results on the Emotional Eating Scale showed that the emotions marked as causing a moderate to strong urge to eat included: on edge, nervous, upset, worried, nothing to do, bored, irritated, restless, and discouraged.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [20] [20] A. M. Koball, M. R. Meers, A. Storfer-Isser, S. E. Domoff, and D. R. Musher-Eizenman, 'Eating When Bored: Revision of the Emotional Eating Scale With a Focus on Boredom,' Health Psychology , vol. 31, no. 4, pp. 521-524, 2012. - source

      Not checked in the free run.

  • Findingown data
    § A. Analysis & Results · p. 4

    In Study 2 Pre-Survey, all participants reported that they became more aware of their eating behaviors.

    “In the Pre-Survey, all participants reported that they became more aware of their eating behaviors, and 87.5% reported that they became more aware of their emotions.”
    Quote in the paperNot a substantial claim
  • Findingnone given
    § A. Emotion Sensing System · p. 5

    EDA collected from the bra cup, just under the breast, was still a useful signal.

    “It is worth nothing that collecting EDA from this part of the body is non-standard; however, EDA was still a useful signal.”
    Quote in the paperSubstantial claim
  • Findingown data
    § A. Analysis & Results · p. 4

    In Study 2 Pre-Survey, 87.5% of participants reported that they became more aware of their emotions.

    “In the Pre-Survey, all participants reported that they became more aware of their eating behaviors, and 87.5% reported that they became more aware of their emotions.”
    Quote in the paperSubstantial claim
  • Findingown data
    § A. Analysis & Results · p. 4

    Qualitative feedback indicated that logging in EmoTree was an excellent awareness tool.

    “Qualitative feedback also indicated that logging in EmoTree was an excellent awareness tool:”
    Quote in the paperSubstantial claim
  • May combine two claims. 12
  • Classified as probably not a substantial claim. 14

  • May combine two claims.

    Findingown data
    § A. Analysis & Results · p. 4

    Personalization is desirable for intervention, both for individual differences in eating patterns and for personal preferences about interventions.

    “We learned that personalization is desirable for intervention, not only in regards to individual differences in eating patterns but also for personal preferences about interventions.”
    • Also stated in: § Food and Mood: Just-in-Time Support for Emotional Eating · p. 1 · plain
      “We found lots of individual differences in emotional eating behaviors and that most participants wanted personalized interventions, rather than a pre-determined intervention.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § A. Analysis and Results · p. 3

    Across all participants, few eating events happened when users were in a calm/serene state, so very few eating events were in the lower-right quadrant (positive & relaxed).

    “We also observed across all participants that few eating events happened when the users were in a calm/serene state, so very few eating events were in the lower-right quadrant (i.e., positive & relaxed).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data
    § A. Emotion Sensing System · p. 5

    It was very tedious for participants to wear the prototyped sensing system, as the boards had to be recharged every 3-4 hours, which resulted in participants having to finagle with their wardrobe throughout the day.

    “It was very tedious for participants to wear our prototyped sensing system, as the boards had to be recharged every 3-4 hours, which resulted in participants having to finagle with their wardrobe throughout the day.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data
    § A. Emotion Sensing System · p. 5

    The average conductivity between the sensor and connector was approximately 12 ohms.

    “The average conductivity between the sensor and connector was approximately 12 ohms.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data
    § A. Analysis & Results · p. 4

    Three out of 11 participants were enthusiastic about the deep breathing activity and tried to incorporate it throughout their day, but the others wanted alternatives.

    “For instance, 3 out of 11 participants were enthusiastic about the deep breathing activity and tried to incorporate it throughout their day, but the others wanted alternatives.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Findingown data
    § A. Analysis & Results · p. 4

    Participants suggested alternative interventions including presenting something funny, filling out a gratitude questionnaire, meditation, suggestions to take a walk, brain teasers/games, asking if they are really hungry, reading or writing, and calling a friend.

    “The alternative interventions that they suggested, included: presenting something funny to alleviate boredom or stress; having them fill out a gratitude questionnaire (i.e., reminding them what they have to be thankful for); meditation; suggestions to take a walk; brain teasers/games; asking if they are really hungry; reading or writing; and calling a friend.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § A. Analysis and Results · p. 3

    Six participants tended to eat when they were predominantly stressed, with the majority of eating events occurring in the upper-left quadrant (negative & pumped).

    “However, we did find that six participants tended to eat when they were predominantly stressed with the majority of eating events occurring in the upper-left quadrant (i.e., negative & pumped).”
    • Also stated in: § VIII. CONTRIBUTIONS AND FUTURE WORK · p. 6 · plain
      “We found in Study 1 from user logging that 6 out of 12 participants primarily ate when they were stressed.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data
    § A. Emotion Sensing System · p. 6

    For both arousal and valence, the recognition accuracy is significantly better than chance.

    “We observe that for both arousal and valence the recognition accuracy is significantly better than chance.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Self-reported emotions are an accurate ground truth for emotion detection.

  • May combine two claims.

    Interpretationown data
    § A. Analysis & Results · p. 4

    While logging made participants more aware, most needed something extra to incent real change.

    “However, only 37.5% reported that their eating behaviors changed as a result of Study 1, so while logging made participants more aware, it seemed that most needed something extra to incent real change.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationown data
    § A. Analysis & Results · p. 4

    From these results, personalizing the intervention would be the best solution.

    “From these results, personalizing the intervention would be the best solution.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § B. Intervention Technique · p. 3

    Stress has been identified as a contributor to obesity and to emotional eating.

    “First, stress has been identified as a contributor to obesity and to emotional eating [7,21].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [7] [7] J. F. Davis, D. L. Choi, D. J. Clegg, and S. C. Benoit, 'Signaling through the ghrelin receptor modulates hippocampal function and metal behavior in mice,' Physiology & Behavior , vol. 103, no. 1, pp. 39-43, 2008. -

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [21] [21] D. J. Wallis and D. M. Hetherington, 'Stress and eating: The effects of ego-threat and cognitive demand on food intake in restrained and emotional stress,' Appetite , vol. 43, no. 1, pp. 39-46, 2004. - source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationown data
    § A. Analysis & Results · p. 4

    Offering a menu of these interventions could break the food focus and prove to be a useful approach, consonant with intervention work in CBT.

    “However, offering a menu of these interventions could break the food focus and prove to be a useful approach, which is consonant with intervention work in CBT.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationown data
    § A. Analysis & Results · p. 4

    The stress/anxiety trigger accuracy result was not surprising given that participants ate for a variety of reasons, not just when stressed.

    “This result was not surprising given that we knew from Study 1 that our participants ate for a variety of reasons, not just when stressed.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning
    § B. Intervention Technique · p. 3

    The breathing exercise may break the focus on stress, allowing further cognitive processing of what one may be about to do, thus offering the opportunity to choose a different, and more positive, action.

    “by targeting stressful moments that may be precursors to an emotional eating episode, the breathing exercise may break the focus on stress, allowing for further cognitive processing of what one may be about to do, thus offering the opportunity to choose a different, and more positive, action.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation
    § A. Analysis and Results · p. 3

    These findings support research on emotional eating as reported elsewhere.

    “These findings support research on emotional eating as reported elsewhere [21].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] [21] D. J. Wallis and D. M. Hetherington, 'Stress and eating: The effects of ego-threat and cognitive demand on food intake in restrained and emotional stress,' Appetite , vol. 43, no. 1, pp. 39-46, 2004. - source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § A. Emotion Sensing System · p. 5

    For testing the classifiers, the authors used a leave-one-example-out methodology.

    “For testing the classifiers, we used a leave-one-example-out methodology.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § C. Supporting Behavior Change · p. 2

    Cognitive Behavioral Therapy (CBT) is an increasingly well-regarded approach to support eating behavior change.

    “An increasingly well-regarded approach to support eating behavior change is Cognitive Behavioral Therapy (CBT) [9].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [9] [9] S. G. Hoffman, A. T. Sawyer, and A. Fang, 'The empirical status of the 'new wave' of CBT,' Psychiatr Clin North Am , vol. 33, no. 3, pp. 701-710, 2010. -

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Premisecitation
    § C. Supporting Behavior Change · p. 2

    Monitoring helps patients be more aware of what is happening in the moment so that they can begin to make changes to behavior that may have seemed automatic or beyond their control.

    “The rationale for the criticality of monitoring is that 'it helps patients to be more aware of what is happening in the moment so that they can begin to make changes to behavior that may have seemed automatic or beyond their control' [12].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [12] [12] R. Murphy, S. Straebler, Z. Cooper, and C. G. Fairburn, 'Cognitive behavioral therapy for eating disorders,' Psychiatr Clin North Am , vol. 33, no. 3, pp. 611-627, 2010. - source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisereasoning
    § C. Supporting Behavior Change · p. 2

    A common approach to supporting discovery of antecedents to maladaptive behaviors, such as cues for emotional eating, is to have participants in CBT identify triggers.

    “A common approach to supporting discovery of these antecedents to maladaptive behaviors, such as one's cues for emotional eating, is to have participants in CBT do work to identify triggers.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisereasoning
    § C. Supporting Behavior Change · p. 2

    An important aspect of CBT is bringing automatic or thoughtless actions into consciousness for deliberate engagement/change.

    “An important aspect of CBT is bringing the automatic or thoughtless actions into consciousness for deliberate engagement/change.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § VIII. CONTRIBUTIONS AND FUTURE WORK · p. 6

    The wearable sensors in the brassiere form factor only allowed women to participate in Study 3.

    “While the wearable sensors in the brassiere form factor only allowed women to participate in Study 3, we have now moved towards using the Affectiva Q 3 sensor bracelets for this collection.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § B. Eating for Non-Homeostatic Reasons · p. 1

    Both physiological responses and psychosocial responses are malleable.

    “In other words, both our physiological responses (the release of ghrelin to cue stomach grumbling at particular times [7]) and psychosocial responses (eat in the presence of food or in response to stress) are malleable.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [7] [7] J. F. Davis, D. L. Choi, D. J. Clegg, and S. C. Benoit, 'Signaling through the ghrelin receptor modulates hippocampal function and metal behavior in mice,' Physiology & Behavior , vol. 103, no. 1, pp. 39-43, 2008. -

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [8] [8] C. H. Gilhooly, S. K. Das, J. K. Golden, M. A. McCrory, G. E. Dallal, E. Saltzman, F. M. Kramer, and S. B. Roberts, 'Food cravings and Energy Regulation: The characteristics of craved foods and their relationship with eating behaviors and weight change during 6 months of dietary energy restriction,' International Journal of Obesity , vol. 31, no. 12, pp. 1849-1859, 2007. - source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation
    § D. Implicit Emotion Detection · p. 2

    McDuff et al. found that the electrodermal activity (EDA) signal was the most beneficial for emotion detection.

    “They found that the electrodermal activity (EDA) signal was the most beneficial, which is a measure of the eccrine sweat glands [14,15].”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [13] [13] D. McDuff, A. Karlson, A. Kapoor, A. Roseway, and M. Cszerwinski, 'AffectAura: An Intelligent System for Emotional Memory,' in Proceedings of ACM CHI 2012 , pp. 849-858, 2012. - source

      Not checked in the free run.

    • cites [14] [14] W. Boucsein, Electrodermal Activity . New York: Springer-Verlag, 1992. - source

      Not checked in the free run.

    • cites [15] [15] J. A. Russell, 'A circumplex model of affect,' Journal of Personality and Social Psychology , vol. 39, no. 6, pp. 1161-1178, 1980. - source

      Not checked in the free run.

23 of 23

  • The paper relies on this without support.

    Fixed solutions for persuasive technology cannot by themselves effect long-term behavioral change.

    “However, one cannot rely solely upon fixed solutions for persuasive technology to effect longterm behavioral change.”

  • The paper relies on this without support.

    Self-reported emotions are an accurate ground truth for emotion detection.

    “They also used the mobile application, EmoTree, in order to self-report their emotions, which was used as labeled data in our machine learning classifier (ground truth).”
2 of 2

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Emotion self-report tool is assigned to the wrong figure panel

    Minor

    The text identifies Figure 1A as the emotion self-report tool, but the Figure 1 caption identifies panel A as the main page.

1 of 1

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. mandryk, 2007, A fuzzy physiological approach for continuously modeling emotion during interaction with play environments, International Journal of Human-Computer Interaction, 6, 329
  2. chang, 2011, From affect to mental health: A speech analysis library for unobtrusive monitoring on mobile phones, 2nd International Workshop on Sensing Applications on Mobile Phones (PhoneSense)

Cited by

Sorted by citations

Related

Authors
Carroll, E. A., Czerwinski, M., Roseway, A., Kapoor, A., Johns, P., Rowan, K., & Schraefel, M. C.
Affiliations
  • University of Rochester (US)
  • Microsoft (United States) (US)
  • University of Southampton (GB)
ORCID
Venue
2013 Humaine Association Conference on Affective Computing and Intelligent Interaction
Publisher
IEEE
Published
2013-09
Pages
252-257
ISSN
2156-8103 (ISSN-L)
DOI
10.1109/acii.2013.48
OpenAlex
W2037131392
Cited by
64
References
23
Citations by year
2025: 3 · 2024: 1 · 2023: 3 · 2022: 3 · 2021: 2 · 2020: 7 · 2019: 8 · 2018: 6 · 2017: 8 · 2016: 6 · 2015: 9 · 2014: 8
FWCI
6.05
Type
conference paper
Language
en
Open access
closed
Locations
  • International Conference on Affective Computing and Intelligent Interaction and workshops (publishedVersion)
  • ePrints Soton (University of Southampton) (acceptedVersion)
Topics
  • Physical Sciences › Computer Science › Human-Computer Interaction › Innovative Human-Technology Interaction
  • Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
  • Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
Keywords
emotional eating, persuasive technology, behavior modification, emotion monitoring, mobile application, personalized interventions, individual differences, wearable sensors, physiological sensors, emotion recognition, women

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review