Impact Ambivalence: How People with Eating Disorders Get Trapped in the Perpetual Cycle of Digital Food Content Engagement
Extracted source text · arXiv 2311.05920
%%
%% This is file `sample-manuscript.tex',
%% generated with the docstrip utility.
%%
%% The original source files were:
%%
%% samples.dtx (with options: `manuscript')
%%
%% IMPORTANT NOTICE:
%%
%% For the copyright see the source file.
%%
%% Any modified versions of this file must be renamed
%% with new filenames distinct from sample-manuscript.tex.
%%
%% For distribution of the original source see the terms
%% for copying and modification in the file samples.dtx.
%%
%% This generated file may be distributed as long as the
%% original source files, as listed above, are part of the
%% same distribution. (The sources need not necessarily be
%% in the same archive or directory.)
%%
%% Commands for TeXCount
%TC:macro \cite [option:text,text]
%TC:macro \citep [option:text,text]
%TC:macro \citet [option:text,text]
%TC:envir table 0 1
%TC:envir table* 0 1
%TC:envir tabular [ignore] word
%TC:envir displaymath 0 word
%TC:envir math 0 word
%TC:envir comment 0 0
%%
%%
%% The first command in your LaTeX source must be the \documentclass command.
%%%% Small single column format, used for CIE, CSUR, DTRAP, JACM, JDIQ, JEA, JERIC, JETC, PACMCGIT, TAAS, TACCESS, TACO, TALG, TALLIP (formerly TALIP), TCPS, TDSCI, TEAC, TECS, TELO, THRI, TIIS, TIOT, TISSEC, TIST, TKDD, TMIS, TOCE, TOCHI, TOCL, TOCS, TOCT, TODAES, TODS, TOIS, TOIT, TOMACS, TOMM (formerly TOMCCAP), TOMPECS, TOMS, TOPC, TOPLAS, TOPS, TOS, TOSEM, TOSN, TQC, TRETS, TSAS, TSC, TSLP, TWEB.
% \documentclass[acmsmall]{acmart}
%%%% Large single column format, used for IMWUT, JOCCH, PACMPL, POMACS, TAP, PACMHCI
% \documentclass[acmlarge,screen]{acmart}
%%%% Large double column format, used for TOG
% \documentclass[acmtog, authorversion]{acmart}
%%%% Generic manuscript mode, required for submission
%%%% and peer review
%\documentclass[sigconf, review, anonymous, 10pt]{acmart}
%\documentclass[manuscript, review, anonymous]{acmart}
\documentclass[sigconf]{acmart}
%\documentclass[acmsmall, review, anonymous, 10pt]{acmart}
%\documentclass[manuscript,review]{acmart}).
%\documentclass[sigconf]{acmart}
\renewcommand\footnotetextcopyrightpermission[1]{}
\settopmatter{printfolios=true,printacmref=false}
%% Fonts used in the template cannot be substituted; margin
%% adjustments are not allowed.
%%
%% \BibTeX command to typeset BibTeX logo in the docs
\AtBeginDocument{%
\providecommand\BibTeX{{%
\normalfont B\kern-0.5em{\scshape i\kern-0.25em b}\kern-0.8em\TeX}}}
% Bib\Tex}}}
%% Rights management information. This information is sent to you
%% when you complete the rights form. These commands have SAMPLE
%% values in them; it is your responsibility as an author to replace
%% the commands and values with those provided to you when you
%% complete the rights form.
%\setcopyright{acmcopyright}
\copyrightyear{2025}
\acmYear{2025}
%\acmDOI{XXXXXXX.XXXXXXX}
\copyrightyear{2025}
\acmYear{2025}
\makeatletter % customized
\def\@ACM@copyright@check@cc{} % customized
\makeatother % customized
% \setcopyright{cc}
% \setcctype{by}
\acmConference[]{}{}
\acmBooktitle{}\acmDOI{}
\acmISBN{}
\setcopyright{none}
%% These commands are for a PROCEEDINGS abstract or paper.
%\acmConference[]{}{}{}
%
% Uncomment \acmBooktitle if th title of the proceedings is different
% from ``Proceedings of ...''!
%
%\acmBooktitle{}
%\acmPrice{15.00}
%\acmISBN{978-1-4503-XXXX-X/18/06}
%%
%% Submission ID.
%% Use this when submitting an article to a sponsored event. You'll
%% receive a unique submission ID from the organizers
%% of the event, and this ID should be used as the parameter to this command.
%%\acmSubmissionID{123-A56-BU3}
%%
%% For managing citations, it is recommended to use bibliography
%% files in BibTeX format.
%%
%% You can then either use BibTeX with the ACM-Reference-Format style,
%% or BibLaTeX with the acmnumeric or acmauthoryear sytles, that include
%% support for advanced citation of software artefact from the
%% biblatex-software package, also separately available on CTAN.
%%
%% Look at the sample-*-biblatex.tex files for templates showcasing
%% the biblatex styles.
%%
%%
%% The majority of ACM publications use numbered citations and
%% references. The command \citestyle{authoryear} switches to the
%% "author year" style.
%%
%% If you are preparing content for an event
%% sponsored by ACM SIGGRAPH, you must use the "author year" style of
%% citations and references.
%% Uncommenting
%% the next command will enable that style.
%%\citestyle{acmauthoryear}
\usepackage{xcolor}
\usepackage{caption}
\usepackage{subcaption}
\usepackage{multirow}
\usepackage{makecell}
\usepackage{array}
\usepackage{longtable}
\usepackage{booktabs}
\usepackage{graphicx}
\usepackage{lscape}
\usepackage{ragged2e}
\usepackage{soul}
\usepackage{tablefootnote}
\usepackage{enumitem}
\usepackage{tabularx}
\usepackage{tikz}
\usepackage{tcolorbox}
\colorlet{LightYellow}{orange!30!}
\colorlet{LightRed}{red!20!}
\colorlet{LightBlue}{blue!20!}
\colorlet{LightViolet}{violet!20!}
\newcommand*{\circled}[2][]{\tikz[baseline=(C.base)]{
\node[inner sep=0pt] (C) {\vphantom{1g}#2};
\node[draw, circle, inner sep=1pt, yshift=0.7pt]
at (C.center) {\vphantom{1g}};}}
\colorlet{RED}{red}
\newcommand{\hr}[1]{{\textcolor{black}{
#1
}}}
\newcommand{\sj}[1]{{\textcolor{orange}{
\it [SJ: #1]
}}}
\newcommand{\sujin}[1]{{\textcolor{black}{
#1
}}}
\newcommand{\subin}[1]{{\textcolor{black}{
#1
}}}
\newcolumntype{P}[1]{>{\raggedright\arraybackslash}p{#1}}
%%
%% end of the preamble, start of the body of the document source.
\begin{document}
\title[Impact Ambivalence]{Impact Ambivalence: How People with Eating Disorders Get Trapped in the Perpetual Cycle of Digital Food Content Engagement}
%
% Digital Food Content and Eating Disorders: Understanding Interaction Patterns and Their Impacts
% Unpacking Digital Food Content Engagement in Binge Eating Disorder and Bulimia
%%
%% The "author" command and its associated commands are used to define
%% the authors and their affiliations.
%% Of note is the shared affiliation of the first two authors, and the
%% "authornote" and "authornotemark" commands
%% used to denote shared contribution to the research.
\author{Ryuhaerang Choi}
\affiliation{%
\institution{KAIST}
\city{Daejeon}
\country{Republic of Korea}
}
\email{ryuhaerang.choi@kaist.ac.kr}
\author{Subin Park}
\affiliation{%
\institution{KAIST}
\city{Daejeon}
\country{Republic of Korea}
}
\email{subin.park@kaist.ac.kr}
\author{Sujin Han}
\affiliation{%
\institution{KAIST}
\city{Daejeon}
\country{Republic of Korea}
}
\email{sujinhan@kaist.ac.kr}
\author{Jennifer G. Kim}
\affiliation{%
\institution{Georgia Institute of Technology}
\city{Atlanta, Georgia}
\country{USA}
}
\email{jennifer.kim@cc.gatech.edu}
\author{Sung-Ju Lee}
\affiliation{%
\institution{KAIST}
\city{Daejeon}
\country{Republic of Korea}
}
\email{profsj@kaist.ac.kr}
%%
%% By default, the full list of authors will be used in the page
%% headers. Often, this list is too long, and will overlap
%% other information printed in the page headers. This command allows
%% the author to define a more concise list
%% of authors' names for this purpose.
\renewcommand{\shortauthors}{Ryuhaerang Choi et al.}
%\settopmatter{printfolios=true}
%%
%% The abstract is a short summary of the work to be presented in the
%% article.
\begin{abstract}
Digital food content could impact viewers' dietary health, \hr{with individuals with eating disorders being particularly sensitive to it.}
However, a comprehensive understanding of why and how these individuals interact with such content is lacking. To fill this void, we conducted \hr{exploratory (N=23) and in-depth studies (N=22)} with individuals with eating disorders to understand their motivations and practices of consuming digital food content. \hr{We reveal that participants engaged with digital food content for both disorder‑driven and recovery‑supporting motivations, leading to conflicting outcomes. This \emph{impact ambivalence}, the coexistence of recovery‑supporting benefits and disorder‑exacerbating risks, sustained a cycle of quitting, prompted by awareness of harm, and returning, motivated by anticipated benefits. We interpret these dynamics within dual systems theory and highlight how recognizing such ambivalence can inform the design of interventions that foster healthier digital food content engagement and mitigate post-engagement harmful effects.} % highlight how recognizing such ambivalence can inform interventions that foster healthier engagement.}
%Our study reveals that participants anticipate positive effects from food media to overcome their condition, but in practice, it often exacerbates their disorder. We also discovered that many participants experienced a cycle of quitting and returning to digital food content consumption. Based on these findings, we articulate cognitive mechanisms underlying their digital food content engagement by incorporating \emph{ambivalence} into dual systems theory and discuss design implications to support digital food content engagement and moderate post-engagement impacts. %and multimedia platforms to support vulnerable individuals.
\end{abstract}
% We also discovered that many participants experienced a cycle of quitting and returning to digital food content consumption, including food delivery apps.
%Food content on digital media is ubiquitous. Individuals with eating disorders are more sensitive to such visual and auditory stimuli of digital food content,
%While the effect of digital food content on dietary health has been researched, the potential negative impact on individuals with eating disorders has received little attention. To address this gap, we developed \system{}, a system that monitors and intervenes in response to food content on smartphones and personal computers. Specifically, \system{} passively hides food content and actively intervenes when users attempt to access it. We conducted a three-week user study with individuals with binge eating disorder or bulimia (n=22) to investigate the perceived impact of food content and the effects of \system{} on exposure and symptoms. Our results demonstrate the potential of \system{} to mitigate the negative impact of food content on individuals with eating disorders. We also provide design implications and insights for future research in this area.
%%
%% The code below is generated by the tool at http://dl.acm.org/ccs.cfm.
%% Please copy and paste the code instead of the example below.
%%
\begin{comment}
\begin{CCSXML}
<ccs2012>
<concept>
<concept_id>10003120.10003121</concept_id>
<concept_desc>Human-centered computing~Human computer interaction (HCI)</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003120.10003121.10003122</concept_id>
<concept_desc>Human-centered computing~HCI design and evaluation methods</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003120.10003121.10003122.10003334</concept_id>
<concept_desc>Human-centered computing~User studies</concept_desc>
<concept_significance>500</concept_significance>
</concept>
</ccs2012>
\end{CCSXML}
\ccsdesc[500]{Human-centered computing~Human computer interaction (HCI)}
\ccsdesc[500]{Human-centered computing~HCI design and evaluation methods}
\ccsdesc[500]{Human-centered computing~User studies}
%%
%% Keywords. The author(s) should pick words that accurately describe
%% the work being presented. Separate the keywords with commas.
\keywords{food content, eating disorder, intervention, bulimia, binge eating disorder}
\end{comment}
\maketitle
\input{01_Introduction.tex}
\input{02_Related_Work.tex}
\input{03_Exploratory_Study.tex}
\input{04_In-Depth_Study.tex}
\input{05_Results.tex}
\input{06_Discussion.tex}
\input{07_Conclusion.tex}
%%
%% The acknowledgments section is defined using the "acks" environment
%% (and NOT an unnumbered section). This ensures the proper
%% identification of the section in the article metadata, and the
%% consistent spelling of the heading.
\newpage
%TC:ignore
%%
%% The next two lines define the bibliography style to be used, and
%% the bibliography file.
\bibliographystyle{ACM-Reference-Format}
\bibliography{main}
%TC:endignore
%%
%% If your work has an appendix, this is the place to put it.
% \appendix
% \input{09_Supplementary.tex}
\end{document}
\endinput
%%
%% End of file `sample-authordraft.tex'.
\section{Introduction}
\textbf{Caution}: \emph{This paper discusses eating disorders and includes media that could potentially be a trigger to those dealing with eating disorders. Please use discretion when reading and disseminating this paper.}
%(1) The growing popularity of digital food content
Digital media's exponential growth has made food content ubiquitous and diversified. On social media, the popularity of food videos (e.g.,~eating and cooking broadcasts) and images has significantly increased~\cite{kang2020popularity, mejova2016fetishizing, de2005tv, adema2000vicarious}. On TikTok, the hashtag \#TikTokFood has been \hr{posted over 5M~times~\cite{tiktokfood}} and viewed over 43~billion times~\cite{tiktokwrapped2021}, and Instagram has over \hr{313~million} posts tagged with \#foodporn~\cite{InstaFoodporn}.
\hr{The online food market---especially app-based delivery that trades on visually enticing food imagery---has kept surging: delivery accounted for 21\% of global consumer food service spending in 2024, up from 9\% in 2019~\cite{foodservice_delivery}.} %The online food market, especially food delivery services often showcasing visually enticing images, has also seen tremendous growth, more than threefold, since 2017, rising as a global market worth more than \$150~billion~\cite{onlineFoodDelivery, global2021}.
%(2) Researches have examine bad impacts of digital food content on eating habits
A growing body of research has examined the impact of increased exposure to various digital food content on eating habits and dietary health~\cite{petit2016can, kinard2016insta, ofli2017saki, vermeir2020visual, pope2015viewers, choi2024foodcensor, wu2024digging, wang2024viral}.
The trend of watching and posting food images and videos with catchy hashtags has been associated with inadequate eating habits~\cite{yan2019digital, qutteina2019media}, unhealthy food choices~\cite{peng2018feast, basso2018taste}, and overeating~\cite{bodenlos2013watching, halford2004effect} through various visual and social factors~\cite{seal2022can}. Browsing and ordering food using online shopping and food delivery services are also known to involve unhealthy food choices~\cite{yeo2017consumer}, unhealthy eating~\cite{eu2021consumers}, and obesity~\cite{stephens2020food}. Moreover, many digital food content features take-away food~\cite{kang2020popularity}, and combining eating content with online food delivery services could lead to unhealthy lifestyles.
%(3) Less investigation on how digital food content affects people with EDs in daily lives. => We try to shed a light on it.
Recent studies have highlighted that \emph{individuals with eating disorders} (EDs) engage with digital food content significantly more than the general population~\cite{bachner2018lives} and are particularly sensitive to such content~\cite{jin2018interactive, legenbauer2004anticipatory}.
When coupled with frequent unconscious exposure to digital food content, it could exacerbate ED symptoms, such as binge eating and obsessive food thoughts~\cite{choi2024foodcensor, kinkel2022food}.
Despite these findings, there remains a significant gap in understanding why individuals with EDs engage with digital food content and how such engagement ultimately contributes to their disordered eating in their daily lives. In this regard, we aim to shed light on the experience of people with disordered eating behaviors with digital food content by answering the following question: \textit{Why and how do people with EDs interact with digital food content?} %how do people with EDs perceive and interact with digital food content?
To answer this question, we conducted two rounds of studies with individuals with eating disorders:~one round of an exploratory study (N=23) and another round of an in-depth study (N=22).
The exploratory study assesses how likely individuals with EDs are to be concerned about associations between digital food content and disordered eating, and identifies which digital food content types are associated with disordered eating.
Notably, we refrained from bringing up digital food content, allowing participants to broach the subject voluntarily. Sixteen participants out of twenty-three voluntarily linked digital food content to their ED symptoms. Our analysis identified two representative digital food content positively associated with ED symptoms: food media~(e.g.,~images and videos) and food delivery apps. \hr{In particular, we found that food media played an \emph{ambivalent} role; sometimes used for managing eating-related urges, yet at other times acting as a trigger for binge eating. }
% We highlight ambivalence as a core element of user experience to be designed for and with in susta.
%To deepen our understanding of ED individuals' motivation and practices of consuming each type of digital food content, we conducted in-depth interviews with 22~individuals (including two from our exploratory study) with EDs.
\hr{To further understand the underlying dynamics of this ambivalence, we conducted in-depth interviews focusing on ED individuals' motivations and practices of consuming digital food content.} Going beyond prior research's emphasis on the detrimental impacts of digital food content on ED populations~\cite{kinkel2022food, qutteina2019media, yan2019digital}, \hr{our in-depth study examined how this ambivalence permeates participants' engagement from motivations that draw them to such content, to the impacts it produces, and the cycles of avoidance and relapse it sustains.} %our main finding highlights its dual role as both a positive self-care tool and a trigger for disordered eating.
%-- 이야기하고 싶은 것들 --%
%-- (1) 기존 attitudinal ambivalence와 잘 연결
%-- (2) 우리 연구에서 발견한 impact ambivalence
%-- (3) 이 impact ambivalence가 결국 anticipated impacts의 ambivalence를 불러일으키고, 그래서 attitudinal impact도 생김는 것
\hr{This tension echoes the concept of \emph{attitudinal ambivalence}, the coexistence of positive and negative evaluations toward the same object or behavior~\cite{thompson2014let, van2009agony, schneider2017mixed}. In our study, we first observed the coexistence of recovery-supporting and disorder-exacerbating impacts within the same digital food content engagement. We term this coexistence \emph{impact ambivalence}, extending attitudinal ambivalence to the consequence domain of behavior. Importantly, the recognition of such conflicting impacts shapes individuals' anticipations of future engagements, which then feed into \emph{attitudinal ambivalence}: they come to both desire engagement, expecting benefits, and resist it, anticipating harms. Such impact ambivalence not only explains individuals' recurring cycle of quitting and returning to digital food content, but also surfaces design opportunities often overlooked by binary harm-reduction perspectives. Recognizing this ambivalence enables the identification of moments where supportive and harmful effects intersect, informing interventions that guide users' decisions before engagement and mitigate adverse impacts afterward.
}
%(7) Summary of our research contributions
The main contributions of this paper are summarized as follows.
\begin{itemize}
\item We reveal the perceptions, motivations, impacts, and engagement patterns with various digital food content among individuals with eating disorders.
Extending beyond prior literature, which primarily focuses on its detrimental impacts on individuals with EDs, our findings provide deeper insights into why and how these individuals engage with such content, highlighting its overlooked recovery-supporting potential.
\item We articulate the cognitive mechanisms underlying how individuals with EDs interact with digital food content by integrating the concept of \emph{impact ambivalence} into dual systems theory (Section~\ref{sec:discussion:ambivalence}). While the double-sided impacts of digital food content in the general population indicate distinct positive (e.g.,~entertainment) and negative (e.g.,~triggering overeating) outcomes, \emph{impact ambivalence} reveals the coexistence of conflicting outcomes, \hr{specifically} recovery-supporting benefits and ED-exacerbating risks. This perspective offers deeper insights into the unique challenges faced by individuals with EDs.
\item Drawing on a thorough understanding of the cognitive mechanisms behind their content engagement, we propose intervention designs that promote informed and mindful interactions with digital food content \hr{and mitigate} its effects after engagement.
\end{itemize}
Our study complements prior research on digital food content by \hr{foregrounding the experiences of individuals with eating disorders, providing unique insights into their engagement patterns, and informing the design of digital interventions that better reflect their needs and struggles.}
\section{Related Work}
\subsection{Eating Disorders and Digital Content Use}
% ED & Digital Content
The intersection of digital content use and eating disorders has been a focal point in the HCI community~\cite{feuston2020conformity, pater2021charting, pater2019notjustgirls}. Studies have identified that extensive engagement with digital media, including pro-eating disorder content and imagery showcasing idealized body images, correlates with the onset and exacerbation of eating disorders~\cite{chancellor2017multimodal, gak2022distressing}. Research has also underpinned the critical role of digital environments, which often emphasize thinness and perfection as desirable body standards, in shaping individuals' perceptions of body image and eating behaviors~\cite{aparicio2019social}. In addition, such social media environments encouraged individuals to constantly compare individuals with influencers and celebrities, leading to negative body image and low self-esteem, driving individuals to engage in disordered eating behaviors to meet unrealistic ideals~\cite{pedalino2022instagram}. Individuals vulnerable to body dissatisfaction and disordered eating behaviors are particularly susceptible to these media messages, as they fuel the internalization of harmful beauty ideals~\cite{murraystuart2018contribution}. The consistent engagement in such content contributed to the persistence and intensification of eating disorder symptoms.
% ~\cite{milton2023see}
%Moreover, recent investigations have expanded our understanding of how individuals with eating disorders actively utilize and interact with various digital media, going beyond mere passive exposure. These studies explore the dynamic ways users with eating disorders engage with online communities and the active seeking or avoidance of certain types of media. Researchers have examined the effects of these interactions on self-perception, treatment, and recovery, as well as how they may perpetuate or alleviate disordered behaviors.
The interplay between digital content and eating disorders has underscored the need for a comprehensive investigation that considers the multifaceted nature of the impact of digital content on eating disorders~\cite{chancellor2016post, chancellor2016thyghgapp}. While a substantial body of research has explored the link between various digital content consumption and eating disorders, there is a notable gap in our understanding of the specific association between food-related content, which is closely connected to dietary health~\cite{petit2016can, kinard2016insta, ofli2017saki, vermeir2020visual}, and disordered eating behaviors.
\subsection{An Overarching Presence of Digital Food Content}
Digital food content comprises diverse multimedia materials that revolve around the food theme. This encompasses various content types, including captivating recipe videos, tantalizing food imagery, informative cooking tutorials, and engaging food blogs. Moreover, the proliferation of food delivery apps that present visually captivating food pictures has expanded digital food content, providing users with seamless access to various culinary offerings through their digital devices~\cite{maimaiti2018we}.
In today's digital age, the prevalence of digital food content is palpable across various online platforms~\cite{kang2020popularity, mejova2016fetishizing, weber2021s, tiktokwrapped2021}. Individuals constantly engage in various food-related content, from social media to video-sharing platforms, dedicated food websites, and mobile applications. This content not only caters to culinary enthusiasts seeking inspiration and knowledge~\cite{de2005tv, wang2022tiktok} but also to a broader audience drawn by the visual allure and cultural narratives associated with food~\cite{spence2016eating, adema2000vicarious, peng2018feast}. As a result, the pervasive presence of digital food content has transformed how people engage with and perceive food in their daily lives~\cite{yan2019digital}.
This omnipresence of digital food content underscores the need to explore its impact on eating behaviors, especially among individuals with eating disorders who are especially susceptible to such content~\cite{jin2018interactive, legenbauer2004anticipatory}. The subsequent section delves into the intricate dynamics of digital food content's influence on individuals' eating behaviors, shedding light on its constructive and detrimental effects.
\subsection{Double-sided Impact of Digital Food Content on Eating Behaviors}
Digital food content exerts a complex influence on individuals' eating behaviors, presenting a dual-sided impact characterized by both constructive and detrimental facets~\cite{strand2020mukbang, wang2024viral}. While many individuals engage with digital food content for inspiration, recipe ideas, and cultural connections to food~\cite{de2005tv}, the comprehensive impact of such media on individuals with eating disorders, who are much more sensitive to digital food content than the general population~\cite{schienle2009binge, lee2017impaired}, has received limited attention. The relationship between individuals' consumption of digital food content, including the engagement with eating broadcasts (\textit{Mukbang}), and the potential escalation of disordered eating behaviors has garnered recognition in prior research, primarily focusing on the general population~\cite{von2023associations, anari2023relationship}. However, these investigations have yielded varying insights and have yet to explicitly target individuals with eating disorders.
% Positive impact
Certain instances of digital food content consumption have been associated with fostering positive eating habits and healthier dietary choices~\cite{drivas2024whatieatinaday}. Informative cooking tutorials, recipe demonstrations, and nutrition education videos can empower individuals to experiment with wholesome meal options and develop improved culinary skills~\cite{wang2022tiktok, ventura2021socialfood}. These platforms potentially serve as educational tools, equipping users with knowledge to make informed decisions about their food intake. Such exposure can lead to a sense of agency in managing dietary preferences and restrictions, encouraging healthier relationships with food~\cite{donnar2017food}. Furthermore, for certain individuals, engaging with digital food media provides a sense of presence that helps alleviate their loneliness~\cite{anjani2020people}.
However, it is essential to acknowledge that these potential positive effects are not uniformly experienced, and their effectiveness can depend on individual and personal contexts~\cite{strand2020mukbang, seal2022can}.
% Detrimental Impact
Conversely, a growing body of research has highlighted the potentially detrimental impact of digital food content, especially when exacerbating unhealthy eating habits~\cite{pope2015viewers, spence2016eating}. The omnipresence of visually appealing, calorie-laden food imagery in digital content can escalate disordered eating behaviors~\cite{bodenlos2013watching, ventura2021socialfood}. Studies have pointed out that exposure to images and videos of excessive food consumption, particularly those featuring eating challenges or binge eating, can inadvertently trigger or reinforce unhealthy eating patterns~\cite{spence2019digital}. Such content inadvertently glorifies or normalizes overconsumption~\cite{kircaburun2021problematic}. Additionally, the visual appeal of enticing food displays and the sensory experience of hearing eating sounds may inadvertently intensify cravings and compulsive eating tendencies~\cite{spence2016eating, anjani2020people}. Recent research proposed censoring food visuals to mitigate these adverse effects and demonstrated that preventing unconscious consumption of digital food content can help manage eating disorder conditions in daily life~\cite{choi2024foodcensor}.
In addition, food delivery apps, despite their convenience, have been implicated in negatively influencing eating habits~\cite{portingale2023tonight}. These platforms often feature visually appealing images of indulgent and calorie-rich foods strategically designed to entice users~\cite{kapoor2018technology}. Such visual appeals can lead to impulsive and unplanned consumption online~\cite{liu2013website}. Moreover, frequent use of food delivery apps can normalize excessive consumption and encourage a sedentary lifestyle, further exacerbating the risk of weight gain and related health issues~\cite{stephens2020food, maimaiti2018we}. The targeted marketing tactics employed by these apps might amplify the temptation to order food impulsively, making it challenging for individuals, especially those susceptible to disordered eating tendencies, to maintain a balanced and mindful approach to their eating habits.
The current research landscape lacks a profound understanding of why and how individuals with eating disorders engage with digital food content—a realm encompassing spontaneous interactions and passive influences. Existing studies are confined to quantitative survey-based methods, such as examining the correlation between the presence of eating disorders and the frequency of viewing eating broadcast~\cite{kircaburun2021problematic, kircaburun2021compensatory, von2023associations, kim2021binge, anari2023relationship, bachner2018lives}. While these studies are meaningful in establishing links between eating disorders and digital food content, they leave the deeper psychological and behavioral dynamics largely unexplored, creating a significant gap in our understanding. %the intricacies of these interactions with digital food content have mainly been uncharted territory. %Delving into the unique ways individuals with eating disorders engage in digital food content, whether through active exploration or passive exposure, remains a vital void in the literature.
These studies suggested further research into the various aspects of digital food content and media types that potentially promote disordered eating~\cite{kircaburun2021compensatory, kim2021binge}.
The goal of our research is to shed light on these intricate dynamics and unravel the profound impact these interactions have on individuals' attitudes, behaviors, and overall well-being. We aim to provide insights into the mechanisms underpinning the sway of digital food content on this population, paving the way for effective and tailored interventions and support mechanisms on digital food content platforms.
\section{Exploratory Study} % Survey
\label{sec:exploratoryStudy}
Before delving deeply into the investigation of how digital food content affects individuals with EDs, we conducted an IRB-approved exploratory study to broadly assess the extent to which they connect with digital food content concerning their ED symptoms.
\subsection{Participants}
\begin{table*}[ht]
\resizebox{\textwidth}{!}{
\begin{tabular}{rrccccrccl}
\Xhline{2\arrayrulewidth}
\multicolumn{1}{c}{\textbf{P}} & \multicolumn{1}{c}{\textbf{Age}} & \textbf{Gender} & \textbf{\begin{tabular}[c]{@{}c@{}}ED\\ Population\end{tabular}} & \textbf{\begin{tabular}[c]{@{}c@{}}Diagnosis\\ (formal/self)\end{tabular}} & \textbf{\begin{tabular}[c]{@{}c@{}}ED\\ Duration\end{tabular}} & \multicolumn{1}{c}{\textbf{\begin{tabular}[c]{@{}c@{}}EDE-Q\\ Score\end{tabular}}} & \textbf{Interview} & \textbf{Survey} & \multicolumn{1}{c}{\textbf{Stated Digital Food Content}} \\ \hline\hline
1 & 18 & Female & BN & formal & 1 years 2 months & 4.83 & \multicolumn{1}{l}{} & Yes & \\
2 & 19 & Female & BED & self & 4 years & - & \multicolumn{1}{l}{} & Yes & \\
3 & 20 & Female & BN & formal & 5 years & 5.39 & Yes & \multicolumn{1}{l}{} & food content, food delivery apps \\
4 & 20 & Female & BED & self & 7 years & 5.48 & Yes & Yes & food content \\
5 & 21 & Female & BN & self & 1 year & - & \multicolumn{1}{l}{} & Yes & \\
6 & 23 & Female & BN & formal & 4 years & 4.68 & Yes & Yes & food delivery apps \\
7 & 23 & Female & BN & self & 4 years & 4.22 & Yes & Yes & food content \\
8 & 23 & Female & BN & formal & 6 years 5 months & 4.98 & Yes & Yes & food delivery apps \\
9 & 24 & Not to disclose & BN & self & 1 year & 4.32 & Yes & \multicolumn{1}{l}{} & food delivery apps \\
10 & 24 & Female & BN & formal & 4 years & 4.55 & Yes & Yes & food content, food delivery apps \\
11 & 24 & Female & BED & formal & 5 years & 3.97 & Yes & Yes & food content, food delivery apps \\
12 & 24 & Female & BED & self & 9 years & - & \multicolumn{1}{l}{} & Yes & food delivery apps \\
13 & 25 & Female & BN & formal & 2 years 2 months & 5.03 & Yes & \multicolumn{1}{l}{} & food content \\
14 & 25 & Non-binary & BN & self & 3 years & 4.16 & Yes & \multicolumn{1}{l}{} & food content \\
15 & 26 & Female & BN & self & 3 years & 3.52 & Yes & Yes & food content, food delivery apps \\
16 & 27 & Female & BN & formal & 13 years & - & \multicolumn{1}{l}{} & Yes & \\
17 & 29 & Male & BED & self & 3 years & - & \multicolumn{1}{l}{} & Yes & food content \\
18 & 30 & Female & BED & self & 2 months & - & \multicolumn{1}{l}{} & Yes & food content \\
19 & 30 & Not to disclose & BN & formal & 4 years & 3.98 & Yes & Yes & food delivery apps \\
20 & 30 & Female & BN & formal & 15 years & - & \multicolumn{1}{l}{} & Yes & \\
21 & 36 & Female & BN & formal & 17 years & 4.66 & Yes & \multicolumn{1}{l}{} & \\
22 & 38 & Female & BED & self & 6 years 11 months & - & \multicolumn{1}{l}{} & Yes & \\
23 & 43 & Female & BED & self & 20 years & - & \multicolumn{1}{l}{} & Yes & food content
\\ \Xhline{2\arrayrulewidth}
\end{tabular}
}
\caption{Exploratory study participants' demographic, eating disorder information, and stated digital food content. The listed items (i.e.,~food content and food delivery apps) in the stated digital food content are identified from open-ended responses in surveys and interviews. BED refers to Binge Eating Disorder and BN stands for Bulimia Nervosa. BED is characterized by recurrent episodes of excessive eating in a short timeframe~(e.g.,~2~hours) and feeling a loss of control over their eating behavior~\cite{dingemans2002binge}. BN involves a cycle of binge eating and compensatory behaviors, such as self-induced vomiting~\cite{fairburn1986diagnosis}. %Higher scores on the EDE-Q indicate problematic eating behaviors and attitudes. The EDE-Q scores of our study's interviewees (4.56±0.56) were generally higher than those of individuals diagnosed with ED~(4.02±1.28)~\cite{aardoom2012norms}.
}
\label{table:exploratory:participants}
\vspace{-0.5cm}
\end{table*}
We recruited 23~participants (aged 18-43, mean=26.2~years; 19~females, 1~male, 1~non-binary, 2~chose not to disclose) through an advertisement post on an online community for people with EDs under the permission of the moderators in South Korea~\cite{KoreanEatingDisorderSocialSupportCommunity}. The participant gender composition is predominantly female, reflecting universal ED demographics (87.4\%~of identified South Korean binge eating disorder patients being female~\cite{KoreaBEDProportion}, and threefold higher lifetime prevalence of binge eating disorder and bulimia nervosa in females in the US~\cite{hudson2007prevalence, galmiche2019prevalence}).
%We
Table~\ref{table:exploratory:participants} provides an overview of participant demographics, ED information, and types of digital food content mentioned by each participant. % in our Supplementary Materials.
Our study intentionally included individuals who were either formally diagnosed or self-diagnosed, as has been done in prior research~\cite{choi2024foodcensor, feuston2020conformity, choi2025private}. By doing so, we aimed to encompass a population that may not have sought professional treatment, providing insights into the often overlooked experiences of these individuals within a clinical context.
\subsection{Study Protocol}
The study involved an online survey to investigate how people with EDs associate digital food content with their ED symptoms. Additionally, we conducted remote semi-structured interviews to gain a deeper understanding. We employed the Eating Disorder Examination Questionnaire 6.0 (EDE-Q) to assess if our participants represented the eating disorder population~\cite{fairburn2008eating}.\footnote{Higher scores on the EDE-Q indicate problematic eating behaviors and attitudes. The EDE-Q scores of our study's interviewees (4.56±0.56) were generally higher than those of individuals diagnosed with ED~(4.02±1.28)~\cite{aardoom2012norms}.}
To address the social stigma associated with EDs~\cite{puhl2015stigma}, participants were given the option to opt out of any part of the study, which consists of a survey, an interview, and the EDE-Q, they find uncomfortable. %participate in the survey, interviews, or both, as well as respond to the EDE-Q.
% the general population (0.93±0.86), as well as
In surveys, we intentionally refrained from introducing digital food content to accurately assess how participants perceived its influence on their symptoms.
Instead, we asked about their computing device usage associated with disordered eating compared with normal meals. Specifically, survey questions included are: \textit{Are there any differences in your digital device usage before/during/after eating disorder episodes than without eating disorder episodes? If so, which device(s) do you use, and for what purpose?} This approach allowed us to identify unique device usages, including consumption of digital food content, associated with their disordered eating.
%\textit{Is there any digital device usage associated with the time \textbf{before/during/after} eating disorder episodes compared with the time \textbf{before/during/after} usual meals? If so, which device(s) do you use and for what purpose?}
Each Zoom interview, lasting approximately thirty minutes to one hour, was recorded and transcribed after the interview with consent. Given the stigma surrounding EDs~\cite{puhl2015stigma}, we allowed the participants to freely turn off their cameras during the interview. In the interviews, participants shared their experiences regarding how various digital device usages are associated with and result in disordered eating. We also inquired about any apps and content they perceived to affect their ED symptoms. For participants who did not respond to the survey, we also asked the survey questions during the interview. We did not bring up digital food content unless participants voluntarily mentioned it, which is consistent with our survey approach. Only the interviewees, not survey-only respondents, were rewarded approximately USD~23 for a one-hour interview. \hr{The study materials, including survey questions, interview questions and participants' responses, were administered in Korean and subsequently translated into English. All translations were reviewed by the authors to ensure accuracy and preserve the integrity of participants' original statements.}
\subsection{Analysis}
We followed the inductive thematic analysis~\cite{braun2012thematic}. The first author initiated the data familiarization process by transcribing the interviews and used the inductive approach to open-code the survey responses and interview transcripts. To comprehensively understand and validate the initial code, every author reviewed the data and emerging themes throughout regular discussions. Then, we developed two themes that capture digital food content types associated with their ED symptoms. Finally, we titled each theme and gave a description. Note that we report only themes that involve digital food content as other types of digital content, such as body images on social media, have been disclosed from previous studies that investigated triggering multimedia content for ED patients~\cite{feuston2020conformity, gak2022distressing}.
\subsection{Results}
%- 어떤 food content들을 언급했는지 --%
\hr{Sixteen out of twenty-three participants voluntarily linked digital food content to their disordered eating behaviors, revealing a nuanced relationship. Our analysis identified two primary types of content associated with such behaviors: food media and food delivery apps. Specifically, eleven participants mentioned food media, including recipe blogs, food photos, chewing or crunching ASMR videos,\footnote{An Autonomous Sensory Meridian Response (ASMR) refers to a tingling sensation in response to specific visual, auditory, or tactile stimuli. ASMR videos often feature soothing voices, tapping, eating, and other soothing sounds.} and eating broadcasts. Nine participants referred to food delivery apps. These findings reaffirm the connection between digital food content and unhealthy dietary habits~\cite {bodenlos2013watching, spence2019digital}.}
% conflict between expectation and real usage를 발견했고, 이게 deeper investigation을 이끎
\hr{Participants' narratives revealed that food media played an ambivalent role in their experiences with disordered eating, serving both as a self-regulation strategy and a trigger for binge episodes. Several participants described watching food-related videos in an effort to suppress the urge to binge eat or to gain emotional comfort. For example, P15 shared, ``\textit{I watched Mukbangs to try to stop myself from binge eating to get some vicarious satisfaction. I watch them to suppress the urge.}'' She also noted that she ``\textit{sometimes watches videos of others who binge and regret it, especially when feeling stressed. It gives me a sense of solidarity and reminds me that I'm not alone, and that recovery is still possible by seeing people in such videos who make up their minds again even after such relapses.}'' Likewise, P3 shared, ``\textit{I've tried watching slow eating videos while eating, to help myself pace and prevent binge episodes.}''}
\hr{Yet, the same form of media often triggered harmful outcomes as emphasized in the previous study~\cite{choi2024foodcensor}. Some participants, including P3 and P15, who described food media as a helpful coping tool, also reported that it sometimes led to binge eating. P7 explained, ``\textit{When I repeatedly view such things (suggested food media), I unconsciously feel the urge to eat it. It would be okay if I could eat it without feeling guilty about it, but if I, even once, feel guilty (because it would lead to weight gain), it piles up on and on, pushing me to binge eat. So, in the long run, it does have an impact.}''}
\hr{Through our exploratory study, we identified the \emph{ambivalent impact} of food media---both supporting and undermining recovery efforts within a single engagement. To further understand the underlying dynamics of this ambivalence, we conducted a follow-up in-depth interview study focusing on (1)~individuals' motivations for engaging with digital food content, (2)~the conflicting impacts such content has on their eating disorder experiences, and (3)~the strategies they adopt to avoid exposure to triggering content. This deeper investigation aims to inform the design of digital interventions that acknowledge both the harms and the personally meaningful benefits embedded in everyday interactions with food content among individuals with eating disorders.}
\section{In-depth Study}
\subsection{Participants}
\begin{table*}[ht]
\resizebox{\textwidth}{!}{
\begin{tabular}{rrlcccrl}
\Xhline{2\arrayrulewidth}
\multicolumn{1}{c}{\textbf{P}} & \multicolumn{1}{c}{\textbf{Age}} & \multicolumn{1}{c}{\textbf{Gender}} & \multicolumn{1}{c}{\textbf{\begin{tabular}[c]{@{}c@{}}ED\\ Population\end{tabular}}} & \textbf{\begin{tabular}[c]{@{}c@{}}Diagnosis\\ (formal/self)\end{tabular}} & \textbf{\begin{tabular}[c]{@{}c@{}}ED\\ Duration\end{tabular}} & \multicolumn{1}{c}{\textbf{\begin{tabular}[c]{@{}c@{}}EDE-Q \\ Score\end{tabular}}} & \multicolumn{1}{c}{\textbf{Stated Digital Food Content}} \\ \hline\hline
1 & 18 & Female & BN & self & 1 year & 4.86 & food media, food delivery apps \\
2 & 20 & Female & BN & self & 3 months & 3.84 & food media \\
3 & 20 & Female & BN & self & 1 year & 4.67 & food media, food delivery apps \\
4 & 21 & Female & BN & formal & 2 years & 4.38 & food media \\
5 & 22 & Female & BED & self & 3 year & 4.41 & food media \\
6 & 22 & Female & BN & formal & 3 years 1 month & 5.30 & food media \\
7 & 22 & Female & BN & formal & 10 years & 5.11 & food media \\
8 & 23 & Female & BN & self & 3 years & 2.72 & food media, food delivery apps \\
9 & 23 & Female & BN & formal & 4 years & 4.59 & food media, food delivery apps \\
10 & 24 & Female & BED & self & 2 years & 4.19 & food media \\
11 & 24 & Female & BN & formal & 6 year 5 months & 5.27 & food media \\
12 & 25 & Female & BN & self & 5 years & 4.82 & food media, food delivery apps \\
13 & 26 & Female & BN & self & 2 years & 4.71 & food media \\
14 & 26 & Female & BED & self & 2 years & 5.36 & food media, food delivery apps \\
15 & 29 & Female & BN & self & 1 year & 4.96 & food media \\
16 & 29 & Female & BED & self & 2 years & 4.47 & food media, food delivery apps \\
17 & 33 & Female & BN & formal & 7 years 2 months & 5.21 & food media, food delivery apps \\
18 & 33 & Female & BN & self & 15 years & 5.03 & food media, food delivery apps \\
19 & 34 & Female & BED & formal & 10 years & 5.03 & food media, food delivery apps \\
20 & 34 & Female & BN & formal & 15 years & 4.32 & food delivery apps \\
21 & 39 & Male & BED & self & 10 years & 2.74 & food media \\
22 & 41 & Female & BN & self & 20 years & 0.72 & food delivery apps \\ \Xhline{2\arrayrulewidth}
\end{tabular}}
\caption{In-depth study participants' demographics, ED information, and stated digital food content. The stated digital food content items are identified from open-ended interview responses. %BED refers to Binge Eating Disorder and BN stands for Bulimia Nervosa. BED is characterized by recurrent episodes of excessive eating in a short timeframe~(e.g.,~2~hours) and feeling a loss of control over their eating behavior~\cite{dingemans2002binge}. BN involves a cycle of binge eating and compensatory behaviors, such as self-induced vomiting~\cite{fairburn1986diagnosis}.
}
\label{table:in-depthstudy:participants_ede}
\vspace*{-0.3cm}
\end{table*}
We recruited 22~participants (aged~18-41, mean=26.7~years; 21~females and 1~male) through advertisement posts on the online social support communities for people with EDs under the permission of the moderators in South Korea~\cite{KoreanEatingDisorderSocialSupportCommunity, KakaoSocialSupportChatroom}. %We
Table~\ref{table:in-depthstudy:participants_ede} presents participants' demographics and ED diagnosis information. % in our Supplementary Materials.
Two in-depth study participants, P9 and P11 (P6 and P8 in the exploratory study, respectively), also participated in our exploratory interviews.
To be eligible for the study, participants must (1)~be over 18 years old and (2)~have an ED. Since many people do not seek treatment for their EDs, eligibility for this study, as in our exploratory study, was not contingent upon a diagnosis. Participants, however, had to identify themselves as having ED symptoms (e.g.,~restriction of food intake, binge eating, or compensatory behaviors). Again, by recruiting users who did not seek treatment or had discontinued it, we could shed light on the needs of users whose struggles are largely ignored in a clinical setting.
\subsection{Study Protocol}
All phases of our study were IRB-approved and conducted remotely due to the high social stigma of EDs~\cite{puhl2015stigma}. Participants responded to a survey via email before the interview. The survey included the EDE-Q and questions about their demographics and ED~symptoms.\footnote{The average EDE-Q score of the participants was 4.40±1.09, compared with averages of individuals diagnosed with ED~(4.02±1.28)~\cite{aardoom2012norms}.}
Following this, we conducted semi-structured interviews with all participants on Zoom, each lasting approximately thirty minutes to one hour. We allowed participants to turn off their cameras. Every interview was recorded and transcribed with consent from the participants. Our interview delved into the motivations and practices of digital food content consumption by asking questions about participants’ experiences. We encouraged participants to share their varied experiences with different types of food media, covering various food videos, pictures, and posts. The detailed interview protocol is in our Supplementary Materials. \hr{The study materials, including interview questions and participants' responses, were administered in Korean and subsequently translated into English. All translations were reviewed by the authors to ensure accuracy and preserve the integrity of participants' original statements.}
\subsection{Analysis}
We conducted the inductive thematic analysis~\cite{braun2012thematic} of the interview responses to investigate why and how individuals with EDs interact with digital food content and how such content relates to disordered eating behaviors. After transcribing the interviews, the first and second authors individually looked through all transcriptions to fully understand the interview content. They independently generated initial codes and, through iterative discussions on emerging themes, addressed inconsistencies, resolved disagreements, and ensured that the data strongly supported the identified themes. To finalize the analysis, all authors participated in comprehensive discussions. Once the analysis was completed, each theme was titled and described.
\section{In-Depth Study Results}
\label{sec:results}
% \subsection{Interaction with Digital Food Content}
% \label{res:interaction}
\hr{Building on our exploratory study's observation of impact ambivalence---the coexistence of recovery-supporting and ED-exacerbating effects within the same engagement---we structured our in-depth analysis to examine how this ambivalence manifests across participants' motivations, perceived impacts, and avoidance practices.}
\subsection{Patterns of Digital Food Content Consumption}
\label{res:patterns}
\begin{figure*}
\centering
\includegraphics[width=\textwidth]{fig/flowchart_dfc_patterns.png}
\vspace{-0.3cm}
\caption{A flowchart describing interaction patterns with digital food content. The OR symbol ($\oplus$) denotes that the process continues in multiple branches. The Summoning Junction symbol ($\otimes$) indicates that multiple branches converge into a single process. %Patterns of food media consumption.
}
\Description{Figure 1 is the flowchart that depicts patterns of digital food content engagement. The detailed explanation of these patterns is described in Section 5.1.}
\label{fig:patterns_dfc}
% \vspace{-0.5cm}
\end{figure*}
\hr{Before delving into specific motivations, impacts, and avoidance practices, we outline the overarching patterns of participants’ engagement with digital food content. This overview provides context for how these different aspects unfold over time and interact with one another (Figure~\ref{fig:patterns_dfc}).}
\hr{Across the interviews, participants described their engagement as a recurring sequence:}
\begin{enumerate}
\item \hr{\textbf{Engage with Digital Food Content:} Participants began consuming digital food content, often by chance or with specific motivations, such as seeking comfort, managing urges, or browsing out of habit.}
\item \hr{\textbf{Recovery‑Supporting Impact:} Early in their engagement, they felt the content satisfied their motivations and did not recognize harmful effects.}
\item \hr{\textbf{Recognize ED‑Exacerbating Impact:} Over time, many noticed that the same content often worsened their ED symptoms.}
\item \hr{\textbf{Try to Avoid Digital Food Content:} This realization led participants to take deliberate steps to limit or stop exposure, such as changing settings, unfollowing channels, or uninstalling apps.}
\item \hr{\textbf{Cease Avoidance (Un)Intentionally:} Despite avoidance efforts, participants frequently resumed engagement, due to loss of self‑control, incidental exposure, or expectations of gaining benefits without harm, re‑entering the cycle.}
\end{enumerate}
\hr{We detail each stage in the following sections, examining the underlying motivations (Section~\ref{res:motivation}), the ambivalent impacts experienced (Section~\ref{res:dfc}), and the strategies and challenges of avoidance (Section~\ref{res:efforts}).}
\subsection{\hr{Recovery-Driven and Disorder-Driven} Motivations for Digital Food Content Consumption} % Motivations for Digital Food Content Consumption
\label{res:motivation}
\begin{table*}[t]
\resizebox{\textwidth}{!}{
\begin{tabular}{lll}
\Xhline{2\arrayrulewidth}
\textbf{Digital Food Content} & \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{Motivation}} & \textbf{Stated Subcategories of Digital Food Content} \\ \hline \hline
\multirow{4}{*}{Food Media} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M1} Expectations for Vicarious Satisfaction & \begin{tabular}[c]{@{}l@{}}eating broadcast, tasting video, \\ new food products showcasing video/post\end{tabular} \\ \cline{2-3}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M2} Expectations for Comfort from Shared Experience & binge eating broadcast/post \\ \cline{2-3}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M3} Seeking Motivation for Change & binge eating broadcast \\ \cline{2-3}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M4} Expectations for Relief by Cooking along with the Media & cooking video \\ \hline
\multirow{3}{*}{Food Delivery Apps} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M5} Habitual Window Shopping & \multicolumn{1}{c}{\multirow{3}{*}{}} \\ \cline{2-2}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M6} Food Media-Triggered Use & \multicolumn{1}{c}{} \\ \cline{2-2}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M7} Orders for Disordered Eating & \multicolumn{1}{c}{} \\ \Xhline{2\arrayrulewidth}
\end{tabular}
}
\caption{Motivations for digital food content consumption. The stated subcategories of digital food content are identified from open-ended interview responses. Note that these subcategories are not always distinctly separate from one another. Detailed descriptions of each subcategory are in our Supplementary Materials.}
\label{table:motivation}
\end{table*}
%Our interview results reaffirm the observations made in our exploratory study and deepen our understanding of why individuals with EDs engage in digital food content. We uncovered various motivations for digital food content consumption practices. Table~\ref{table:motivation} outlines the themes about motivations that emerged from the in-depth study.
\hr{Participants described a variety of motivations for engaging with digital food content. Table~\ref{table:motivation} outlines these motivations and their associated content subcategories. Overall, motivations for food media included both those linked to restrictive eating behaviors and those aimed at supporting recovery, whereas motivations for food delivery apps tended to align with disorder-exacerbating behaviors.}
We identified four motivations for seeking \textit{food media}. Many participants mentioned \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M1}expectations for vicarious satisfaction (N=7)}, \hr{often rooted in restrictive eating, a common eating disorder symptom characterized by severe limitation of food intake. They described intentionally avoiding actual food consumption, fearing weight gain, while watching others eat as a substitute. This mirrors the general population's use of eating broadcasts for vicarious satisfaction~\cite{anjani2020people}, but in our participants' case, the motivation was tied to dietary restriction rather than casual entertainment or sensory enjoyment.}
For example, P21 said, ``\textit{If I overeat whenever I feel like eating, I'll gain weight and become obese. So, [instead of eating,] I watched it [food media] to feel vicarious satisfaction.}'' % P21 "이제 먹고 싶은데 과식 하면은 이제 살찌고 비만이 되니까 이제 대리만족을 느끼기 위해서 봤어요"
%Some mentioned ``\textit{the expectation of vicarious satisfaction}'' (P4) as a trigger to search for food videos or subscribe to eating broadcast channels on YouTube.% P4: 보통 그런 대리 만족을 목적으로 처음에는 보는데 이제 보다가 보면 이 음식 먹고 싶다. 맛있겠다. 이런 생각이 들기도 하고
Some searched for videos filmed by people with EDs, such as binge eating broadcasts, with \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M2}expectations for comfort from shared experiences (N=4)}. Through those videos, they sought ``\textit{relief from shared experiences and feel solidarity}'' (P12). %P12 said, ``\textit{It makes me feel like we're going through the same pain. % P12 이제 그런 똑같은 고통을 겪고 있다는 동질감이 들어서 가끔씩은 안도감들고 싶을 때 불안할 때 그런 (폭식증 먹방) 영상을 보기도 하고
Every participant who remarked on expectations for relief from shared experiences noted that such media calm anxiety and stigma about having an ED. While previous research showed that the general population often watches eating broadcasts for a sense of social connectedness and attachment to the hosts~\cite{anjani2020people}, our finding reveals a different dimension: individuals with EDs are sometimes drawn to videos where the host engages in disordered eating. For them, this content offers not just a connection but a sense of comfort and validation in their shared struggles. This highlights a distinct psychological motivation closely tied to their experiences with their condition.
\textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M3}Seeking motivation for change (N=2)} was another motivator for watching such videos. To quote P2, ``\textit{[Looking at people with EDs binge eat food] made me realize that such behavior looks very bad to other people. This made me reflect on myself. ... Hence, I seek them to motivate myself to overcome my eating disorder.}'' % P2: 폭식 먹방 같은 거 보면은 되게 사실 되게 별로 보기 좋은 모습은 아니라서 이렇게 남들 눈에는 이렇게 보이는구나라고 해서 약간 자기 반성하게 되기도 하고 그랬어요. ... 폭식 먹방에 있는 약간 스스로 거울을 보게 되는 느낌이라서. 그런 걸 보면은 이제 그런 걸 보면서 참으려고
Additionally, a few stated \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M4}expectations for relief by cooking along with the media (N=2)}. For instance, they engaged with cooking videos to ``\textit{alleviate the urge to binge eat and feel relief by cooking along with the videos}'' (P8). % P8 (요리 영상을 보고 직접) 요리를 하면서 좀 더 이렇게 요리를 하는 과정에서부터 먹기까지 시간이 오래 걸리잖아요. 그 과정에서 이제 먹는 충동에 대한 욕구도 점점 이렇게 사라지고 그 과정을 전부 밟으면서 제거 행동에 대한 생각이 점점 줄어드는 경우가 있어요.
Interestingly, unlike the general population who often seek practical information such as recipes or nutritional details from cooking videos~\cite{wang2022tiktok}, some participants with EDs engage with this content for emotional relief and \hr{regulation of binge-eating urges,} %cravings management,
highlighting a unique psychological motivation.
For the use of \textit{food delivery apps}, we identified three motivations. Notably, more than half of the participants who associated food delivery apps with their ED behaviors mentioned \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M5}habitual window shopping (N=6)} as a motivation for using food delivery apps. They ``\textit{browse food delivery apps when bored, but do not order anything}'' (P4). %P16 said, ``\textit{it [food delivery app] is right there [on my phone], so I unconsciously open the app.}'' % 배달 앱을 이제 깔고 사용하다 보니까 이게 하루에 한두 번씩은 계속 보게 돼요. (P16)
\hr{While seemingly casual, participants linked this habitual browsing to their disordered eating cycles, describing it as a precursor that often escalated into binge episodes. P17 described, ``\textit{I often check food delivery apps. When I feel [binge] urges, I open the app, put desired items in the cart, and if the order isn't accepted right away, I cancel it, thinking `I shouldn't binge eat it,' but then I end up ordering it again because I still want it. Even without urges, I browse out of boredom, see pictures, and feel triggered to want the food.}''} % P17 (멋진표범) "저는 배달 앱을 상당히 자주 보는 편이에요. [폭식] 자극을 받으면 배달 앱을 켜고, 먹고 싶은 걸 장바구니에 담아놓고 결제까지 눌렀다가, 주문이 즉시 수락되지 않는 경우에는 그 틈에 취소하고 "그래 먹지 말아야지" 하다가. 결국 먹고싶으니까 다시 주문하는 이런 경우가 굉장히 많아요. [폭식] 자극이 없을 때도 심심해서 배달 앱 보고 [음식] 그림 보면서 자극받아서 먹고싶어 지는 경우도 있어요."
It is also notable that more than half of them mentioned \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M6}food media-triggered use (N=6)}. P6 described, ``\textit{After watching a video of somebody finishing a half gallon of mint chocolate ice cream, I ordered it (on a food delivery app). Often, I end up ordering and eating the same thing after watching it [eating broadcasts].}'' % (산책하는달팽이 P6) 유튜브에서 어떤 분이 이제 민트 초코 아이스크림을 한 통을 이제 막 다 드시는 그런 영상이 추천에 떠서 이제 보게 됐는데 ... 되게 먹으면서 기쁘게 먹는 것 같은 근데 이제 그 영상을 보고 나서 제가 이제 하루하고 한 반나절인가 지나서 이제 아이스크림을 똑같이 시켰어요. ...약간 그걸 따라서 (시켜서) 먹게 되는 그런 것도 있는 것 같아요.% H trigger: 멋진호랑이(8), 멋진표범(17), 온화한플라밍고(18), 산책하는여우(13), 산책하는달팽이(6), 당당한호랑이(16)
This suggests that combining different types of digital food content, specifically food media and food delivery apps, can together influence immediate behaviors among individuals with EDs. While food media on social media alone has been recognized as a powerful tool to encourage food product purchases~\cite{ventura2021socialfood}, its integration with food delivery apps appears to facilitate this effect. This dynamic interplay underscores the complex role of digital food content in shaping user behaviors, such as food choices and eating practices. %, which may contribute to disordered eating patterns.}
Some participants said they use food delivery apps only to make \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightYellow]{M7}orders for disordered eating (N=4)}, but never for normal meals. They even said they often make multiple orders for disordered eating to get enough food to binge eat and purge. P7 said, ``\textit{When I binge eat, though I'm very full, I make multiple orders}''. % P7 폭식을 해야겠다는 마음 먹었을 때는 엄청 배부른데도 … 또 시키고 또 시키고 이렇게 되는 게 맞는 것 같아요.
\hr{This motivation reflects a direct and purposeful link to their ongoing eating disorder behaviors}
% [Habitual window shopping]멋진 표범 (P17), 느긋한달팽이(P4), 활기찬 개구리(P1), 수줍은펭귄(P15), 당당한호랑이(P16), 자유로운표범 (P12)
% [Orders for disordered eating] 산책하는달팽이 (P6), 아름다운고래 (P11), 활기찬고양이 (P20), 자유로운고래 (P7)
To summarize, \hr{motivations for food media ranged from those rooted in restrictive eating behaviors to recovery-supporting aims. This mix of restrictive-eating-driven and recovery-supporting motivations illustrates the ambivalence underlying food media engagement. On the other hand, motivations for food delivery apps were largely disorder-driven, that is, arising from and reinforcing existing patterns of disordered eating behaviors.}%participants engaged in various food media, primarily as a self-care strategyto quell the urge to (binge) eat to overcome EDs. On the other hand, their use of food delivery apps exhibited a range of motivations, none of which were geared toward overcoming EDs.
% Digital Food Content ED-Promoting or Recovery-Empowering
\subsection{Digital Food Content Disorder-Exacerbating or Recovery-Supporting}
\label{res:dfc}
\begin{table*}[t]
\resizebox{\textwidth}{!}{
\begin{tabular}{lll}
\Xhline{2\arrayrulewidth}
\multicolumn{1}{c}{\textbf{Digital Food Content}} & \multicolumn{2}{c}{\textbf{How Being Associated with Disordered Eating}} \\ \hline\hline
\multirow{8}{*}{Food Media} & \multirow{5}{*}{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED-Exacerbating Impact}} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED1} Stimulate Visually and Auditorily \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED2} Cannot Provide Vicarious Satisfaction \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED3} Present Abnormally Large Amount of Food Consumption \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED4} Trigger Overwhelming Thoughts about Food \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED5} Normalize Eating Disorders from Watching Other Patients \\ \cline{2-3}
& \multirow{3}{*}{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{Recovery-Supporting Impact}} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{REC1} Curb Urge to Binge Eating by Following Recipes \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{REC2} Provide Relief through Shared Experiences \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{REC3} Promote Empowerment for Change \\ \hline
\multirow{2}{*}{Food Delivery Apps} & \multirow{2}{*}{ \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED-Exacerbating Impact}} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED1} Stimulate Visually \\ \cline{3-3}
& & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED6} Lower Barrier of Getting Food \\\Xhline{2\arrayrulewidth}
\end{tabular}
}
\caption{\hr{Recovery-supporting and disorder-exacerbating impacts of digital food content, along with associated contributing factors, illustrating their ambivalence.}
}
\label{table:dfc_lead_mitigate}
\end{table*}
% 멋진표범, 귀여운캥거루, 산책하는여우, 기쁜플라밍고, 활기찬개구리, 자유로운고래, 당당한사자, 산책하는달팽이, 기쁜나무늘보, 느긋한달팽이, 아름다운고래
%Through the interviews, we gained insights into how digital food content can either trigger disordered eating, mostly binge eating, or support participants in overcoming their disordered eating behaviors.
\hr{The diverse motivations shaped how participants experienced digital food content, leading to both recovery‑supporting and \\ disorder‑exacerbating impacts. These contrasting effects often arose in relation to the same type of content, underscoring its ambivalent nature.} Table~\ref{table:dfc_lead_mitigate} outlines \hr{these impacts and the contributing factors that shaped them.} %the factors of different impacts of digital food content.
We identified five themes about how \textit{food media} contributes to eating disorder symptoms. Many participants discussed \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED1} visual and auditory stimuli (N=11)} incite their urge to binge eat. P10 said, ``\textit{I can't refrain myself from binge eating when I watch an eating broadcast with visual or auditory stimulation. The more my cravings are aroused [by watching it], the more and faster I eat.}'' % P10: 시각적 자극이나 청각적 자극을 먹방을 보면서 보면은 더 참을 수 없었기 때문에 오히려 더 먹게 되거나 먹는 자극을 더 크게 느껴서 욕구를 더 욕구가 더 크게 자극돼서 더 빨리 급하게 더 희마하게 먹게 되었습니다.
P2 commented, ``\textit{the enticing display of food makes [the food media] more stimulating}.'' %P2: 음식 자체를 맛있게 보이게 하니까 그래서 더 그런
Some participants mentioned that the eating sound makes digital food media %eating broadcasts
more invigorating. P17 said, ``\textit{it stimulates my appetite more when YouTubers eat food with vivid sounds.}'' % P17: 먹는 소리도 좀 더 생생하게 담아서 찍어 가지고 먹던 유튜버들 분들이 이제 좀 더 크게 내 식욕을 자극하는 거 같아요 % P13: 청각적인 것거랑 시각적인 게 제일 큰 것 같아요. 시각적으로는 좀 아무래도 많은 양을 한꺼번에 자극적으로 좀 먹다 보니까 그런 것도 있고 귀로 이제 생생하게 요즘은 마이크가 좋으니까 생생하게 그 소리가 들리니까 맛있겠다라는 생각을 좀 부추기는 것 같아요.
% 산책하는여우, 느긋한달팽이, 즐거운고슴도치, 수줍은펭귄, 당당한호랑이, 자유로운고래
Our participants also noted that, despite their expectations for vicarious satisfaction, watching food media \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED2} cannot provide them with vicarious satisfaction (N=6)} but instead increases the urge for food. P19 said, ``\textit{Looking back, I was falsely assuming that, oh, since the person in the video took my place and ate the food for me, I would be satisfied, which wasn't the case, and my urges piled up, leading me to binge eat. Afterward, I regretted my actions and vomited.}'' % 맛있어 보이고 먹고 싶으니까 이제 그걸 보면서 쟤가 대신 먹었으니까 나도 먹은 것처럼 생각을 한다고 내 스스로 이렇게 그냥 나는 만족한다라고 생각하고 있던 거지. 그게 아니라 그게 쌓여서 오히려 그 음식에 대한 갈망을 더 하게 되고 결국 그래서 먹고 먹고 나면 후회하니까 가서 뵙고 토하고 막 이렇게 되는 거더라고요.
Although a few studies claim that the general population gets vicarious satisfaction through watching eating broadcast~\cite{kircaburun2021psychology, anjani2020people}, our findings indicate that it does not always hold for people with EDs, for whom such content may instead lead to disordered eating behaviors.
%
Notably, every participant mentioned that the \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED3} abnormally large amount of food (N=22)} in food media has a negative impact, such as making them normalize binge eating. P6 said, ``\textit{People who eat a lot of food make me think, `It is okay to eat this much because there are other people who eat that much.}'~'' % P6: 많이 먹는 사람들이 음식을 먹는 먹방 그런 거 보면은 저 사람은 저만큼 먹으니까 나도 이만큼 먹어도 되지 않나 이런 생각이 좀 들어요.
This finding extends previous research by showing that, while large food portions in eating broadcasts can lead the general population to normalize overconsumption~\cite{kircaburun2021problematic}, they can also make individuals with EDs normalize their disordered eating behaviors, such as binge eating.
% 당당한사자, 산책하는달팽이, 수줍은펭귄, 당당한호랑이, 자유로운고래
Among all the participants, there was a unanimous assessment of eating broadcasts, indicating at least one adverse impact. However, participants had varying opinions on the effects of other food media types. For example, views on eating segments within non-food videos, such as vlogs, diverged. Some participants believed that eating segments were different from eating broadcasts and were not related to their ED symptoms. In contrast, others considered eating segments in non-food videos and eating broadcasts similar, attributing negative impacts to both.
%Among all the participants, there was a unanimous assessment of eating broadcasts, indicating that they were perceived to have either a negative or no impact.
% Cooking posts and videos
Conflicting assessments also emerged regarding cooking posts and videos. Two participants who identified seeking relief as their motivation for watching digital food media (Section~\ref{res:motivation}), expressed a positive view of cooking posts and videos. They found these resources helpful in \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{REC1} curbing the urge to binge eat by following the provided recipes (N=2)}.
On the other hand, some participants experienced the opposite effect, as exposure to such media led them to ``\textit{continuously ruminate on the recipes and food}'' (P4), \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED4} triggering overwhelming thoughts about food (N=5)}.
%However, for some other participants, exposure to such media had the opposite effect making them ``\textit{think about the recipe on the video, and thus food, endlessly}'' (P4): \textbf{trigger overwhelming thoughts about food (N=5)}. % P4: (요리영상을 봄으로 인해서) 증상이 좀 심해지는 경향이 더 있는 게. 이런. 만약에 같은. 그러니까 한 레시피를 보면 그 레시피에 대해서 이제 계속 생각을 하게 되고 내일 해 먹어야지 내일. 이런 생각이 계속 하루 종일 그냥 음식 생각만 하게 되는 거를 좀 그걸 심화시켜주는 것 같아요. 오히려 그런 영상을 찾아보는 게
% Binge eating broadcasts
For binge eating broadcasts, even the same individual experienced opposing impacts depending on the situation or circumstances.
They primarily fulfilled our participants' motivations, % (Section~\ref{res:interaction:motivation}),
such as seeking \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{REC2} relief through shared experiences (N=4)} and seeking \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightBlue]{REC3} empowerment for change (N=2)}.
Moreover, they have made promises not to binge eat after watching binge eating broadcasts. P1 said, ``\textit{In the video, there are many instances where people regret and blame themselves for binge eating, so when I see that, I think, `I should eat in moderation, not like that.'~}''
% 나중에 폭식하고 후회하고 자책하는 그런 것까지 같이 올려주시는 경우가 많아서 그거를 보고 있을 때는 나는 저렇게 꼭 하지 말고 적당히 먹어야지 하고 생각 들게 하기도 하고
In the meantime, P6 stated that she sometimes felt that she does ``\textit{not have to treat eating disorder, as many people do.}'' This \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED5} normalization of EDs from watching other patients (N=2)} could impede the recovery process for individuals with EDs.
%On the other hand, some watched the videos to pull themselves together. To quote P3, ``\textit{(Looking at people with eating disorders eat food) made me realize that such behavior looks very bad to other people. This made me reflect on myself.}'' % P3: 폭식 먹방 같은 거 보면은 되게 사실 되게 별로 보기 좋은 모습은 아니라서 이렇게 남들 눈에는 이렇게 보이는구나라고 해서 약간 자기 반성하게 되기도 하고 그랬어요.
%Furthermore, some search for and watch cooking videos to relieve their urge to binge; by following the videos and preparing their own food, they could alleviate their food cravings.
%While some mentioned the impact of watching binge eating broadcasts as positive, %providing relief and promoting self-reflection,
%some participants consider it negative, allowing them to justify their disorders as normal. P9 said, ``\textit{Since I know that (binge eating) is abnormal, out of curiosity, I searched for how much food people with bulimia eat. When I watched the related videos, I found myself justifying my behavior. Although my symptoms are abnormal, they are within the normal boundaries of most bulimics.}'' % 이게 비정상적인 걸 알기 때문에 다른 사람들도 이제 폭식증이 걸렸을 때 얼마큼의 음식을 먹는지 그게 너무 궁금해 가지고 찾아봤었어요. 처음에 그 사람들도 그렇게 먹는 거 보고 내 폭식증 증상이 이게 이게 비정상적이지만 그래도 그 폭식증 정상 그 범위에서는 정상이구나 다른 사람들도 이 정도의 음식을 먹는구나라고 생각을 좀 많이 해서 좀 합리화를 많이 시키는 것 같아요.
%Envy// P15: 어느 정도는 악화시켰다고 얘기할 수 있는 게저랑 비슷해 보이는 나이대의 사람들이 좀 많이 먹는 것 같다. 싶으면은 저 사람은 어떻게 저렇게 많이 먹고 말랐지 이렇게 생각이 들면서 자괴감에 빠져서폭식이 유도되는 경우가 있었고 반대로 너무 적게 먹는 사람의 그런 일상을 봤을 때 나는 왜 저 정도로 소식을 하지 못할까 하면서 자책하고 그런데 자신의 모습에 실망스러워서 다시 또 악순환이 반복되기도 했어요.
% P5: 다들 뭔가 이렇게 함께 즐기고 있고 막 그런 걸 보면서 이게 괜찮은 건가 그런 생각도 들고 그리고 그분이 뭔가 되게 이렇게 먹으면서 이렇게 중간중간 약간 살짝 살짝 웃음소리가 들렸거든요. 되게 먹으면서 기쁘게 먹는 것 같은 근데 이제 그 영상을 보고 나서 제가 이제 하루하고 한 반나절인가 지나서 이제 아이스크림을 똑같이 시켰어요. 시키고 이제 그걸 혼자 다 먹고 근데 또 결국에는 토를 했죠. 왜냐하면은 뭔가 건강하지 못한 걸 먹었다는 생각 때문에 약간 그걸 따라서 먹게 되는 그런 것도 있는 것 같아요.
% P5: 그게 쉬워졌다보니까 그런 것도 있고 그다음에 이제 두 번째로는 약간 뭐라고 해야 되지 그냥 정말 그냥 기초적으로 그냥 정말 폭식을 하고 싶어지는 그 마음도 자극이 되고 %It is like a copycat behavior. It makes me want to order what the YouTuber is eating on the video when I watch eating broadcasts. This is because, these days, food delivery apps are everywhere, so if you look for some food, you can get them anywhere.
% Food Delivery Apps
% 멋진호랑이, 멋진표범, 온화한플라밍고, 아름다운플라밍고, 즐거운고슴도치, 수줍은펭귄, 당당한호랑이, 아름다운고래, 자유로운표범, 활기찬고양이, 자유로운고래
Regarding how \textit{food delivery app} use contributes to eating disorders, we identified two themes about how it assists disordered eating.
% P9, P10
One common impact between food media and food delivery apps was \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED1} stimulating visually (N=2)}. P10 noted, ``\textit{With habitual use of food delivery apps, the enticing display of food stimulates my cravings.}''
% P10: 심심해서 배달의 민족 틀어보고 그림 보면서 자극받아서 시각적 자극을 받아서 이거 먹고 싶다. 먹고 싶다 하는 경우도 있고
%
In addition, many participants mentioned the \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightRed]{ED6} lower barrier of getting food (N=11)} on delivery apps compared with buying them in-person as a significant factor leading to disordered eating. P11 said, ``\textit{Food delivery apps enabled us to conveniently obtain food. People bring food to my door only with a few clicks. So I don't need to be self-conscious of how I look to other people. I can finish the (binge eating) process as if nothing happened. This is really bad.}'' %P11: 폭식이라는 행동이 이제 그 과정 까지 필요한 요소들이 있는데 일단 충동을 자극하는 자극하기에 충분한 그런 맛있는 음식 내 머릿속에서 생각하는 예전 보통 다이어트할 때 먹으면 안 되는 금지 음식들 그런 음식들을 접근을 아주 편리하게 할 수 있는 게 배달 어플인 것 같아요. 내가 손만 까딱하면은 내 문 앞에다가 그거를 돈만 주면 갖다 줘 그러니까 나는 어디 밖에 나가서 사람들 사람들의 시선 신경도 쓰지 않고 돈만 주면은 내가 이 과정을 아주 편하게 다 끝마치고 아무렇지도 않게 없었던 일처럼 끝낼 수가 있어 그게 아주 안 좋은 것 같아요. % P19: 네 야밤에 이제 음식을 준비하는 데는 시간이 오래 걸리는데 내가 폭식이 터졌을 때 욕구가 터졌을 때 배달에 판하면은 사실 편하게 배달시켜서 먹을 수 있잖아요. 그 유혹 그래서 그 왜냐하면 그 시간에 누워 있다가 내가 일어나서 뭔가를 차린다는 거는 귀찮아서라도 사실 안 하게 되는데 재료가 없거나 그러면 안 먹게 되잖아요.그럼 되는데 배달앱이 있으니까 굳이 한 번 들어가서 메뉴를 훑어 보고 주문 하나 그 주문만 넣으면 바로 오니까 결국 또 먹게 되고 영향이 확실히 있는 것 같아요. 너무 편해서
%The habitual use also often made them crave food and triggered their eating disorder symptoms. P1 stated, ``\textit{Upon browsing it, I am tempted to eat something.}'' % P1: 심심할 때 배달앱 들어가서 배고플 때나 심심할 때나 들어가서 비교 같은 거. 살펴보다 보면 금방금방 뭘 먹고 싶어지기도 하고
%The discount ads on delivery apps enticed them to binge eat even when they were not hungry (\textbf{enticed by discount ads to binge eat}). P4 described, ``\textit{Often, there are discount coupons. Regardless of whether I want to eat, I consider ordering food because of the discount.}'' % P4: 할인 쿠폰 같은 게 많이 올라오잖아요. 네 만약에 그렇게 딱히 먹고 싶은 욕구가 없어도 그렇게 좀 할인율이 높은 거를 보면 포장해서 먹을까라는 이렇게 딱 주문 직전까지 결제 직전까지 갔다가 취소하는 경우도 종종 있었어요.
%Some mentioned that they order needlessly large amounts of food to save delivery fees. To quote P17, ``\textit{I can get free delivery if I order over a certain amount. I can't even eat that much, but I order a lot to save the delivery fee which I think is a waste. Although I'm full and don't think I can finish the remaining food, I still finish all the food.}'' % P17: 얼마 이상이면은 배달비가 무료 이런 게 있잖아요. 그 금액을 채우기 위해서 오히려 좀 제가 딱 생각해도 못 먹을 양인데 그냥 금액을 채우는 경우가 있거든요. 저는 그냥 배달비가 좀 아까워서 근데 또 저는 제 성격상 눈앞에 있으면 다 먹어야 돼서 못 먹을 것 같은데 하면서도 배부르다 하면서도 그냥 다 먹어요.
In conclusion, \hr{digital food content had both recovery-supporting and disorder-exacerbating impacts on participants with ED, a coexistence we frame as \emph{impact ambivalence}. This ambivalence manifested differently depending on the individuals, circumstances, and subcategories.}
%while our participants with EDs often anticipate positive effects in overcoming EDs when they consume food media, in reality, these media frequently worsen the disorder instead. This discrepancy between expectations and outcomes stems from \emph{impact ambivalence}, the coexistence of conflicting effects where the same content
\subsection{Efforts to Avoid Digital Food Content Leading to Disordered Eating and Relapse}
\label{res:efforts}
\begin{table*}[t]
\resizebox{.65\textwidth}{!}{
\begin{tabular}{ll}
\Xhline{2\arrayrulewidth}
\textbf{Digital Food Content} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{\textbf{Avoidance Practices}} \\ \hline\hline
Food Media & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A1}Influence the Algorithm to Reduce Suggestions \\ \hline
\multirow{3}{*}{Food Delivery Apps} & \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A2}Uninstall Food Delivery Apps \\ \cline{2-2}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A3}Delete Accounts \\ \cline{2-2}
& \tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A4}Use Shared Accounts \\ \Xhline{2\arrayrulewidth}
\end{tabular}
}
\caption{Practices for avoiding digital food content leading to disordered eating.}
\label{tab:avoidance_practice}
\end{table*}
\hr{Aware of the potential harms from digital food content, many participants attempted to avoid it. However, these efforts were often short‑lived, with relapses driven either by a loss of self‑control or the anticipation of recovery‑supporting effects they had previously experienced, such as vicarious satisfaction. We examine the avoidance strategies participants used, the challenges in sustaining them, and the recurring cycle of disengagement and return. Table~\ref{tab:avoidance_practice} summarizes these avoidance practices. }
% 추천 영상 안보려고 뭔가를 한 사람들 (앱 지우기, 그냥 자의로 안보기, 알고리즘 영향미치기 등): 귀여운캥거루, 기쁜나무늘보, 기쁜플라밍고, 당당한호랑이, 멋진표범, 산책하는달팽이, 아름다운고래, 온화한플라밍고, 자유로운고래, 자유로운표범, 활기찬개구리
\hr{For \textit{food media}, eleven participants reported deliberate efforts to reduce or remove exposure.} % 알고리즘 영향 미치기: 귀여운캥거루, 온화한플라밍고, 기쁜나무늘보, 활기찬개구리, 당당한호랑이, 아름다운고래, 자유로운고래, 기쁜플라밍고
Most mentioned that they tried to \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A1}influence the algorithm to reduce suggestions~(N=8)} of food videos by unsubscribing from eating broadcast channels and/or deliberately searching for non-food content.
They utilized the `Not interested' and `Don't recommend channel' buttons on YouTube and the `Not interested' button on Instagram to remove food media from their personalized algorithm. P11 said, ``\textit{There are different measures you can take to make the (YouTube) algorithm not recommend certain videos. So, I tried to use them to remove recommendations for food-related videos.}'' % P11: 이제 알고리즘을 벗어나기 위해서 다 이렇게 추천을 하지 않으면 않는 그런 설정하는 게 있잖아요. 그걸 다 해가지고 되도록이면 그런 먹방이나 아니면 음식을 먹게끔 맛있게 보이게끔 하는 그런 영상들을 추천 안 하게 그렇게 만드는 노력을 좀 했던 것 같아요.
In addition, P13 ``\textit{deliberately searched for random keywords to change recommendations.}''
% 유튜브 사용을 좀 줄일 때도 있었고 아니면 검색어 키워드를 다른 걸로 이제 관심사를 다른 걸로 돌리려고 노력을 했었어요. % 추천되는 거를 안 보려고
Every participant who tried to modify their suggestion algorithm noted that such efforts often failed, \hr{as they eventually returned to actively searching for food-related videos, sometimes to satisfy urges and seek vicarious satisfaction, such as watching eating broadcasts to curb food cravings. }
%because they return to actively searching for food-related videos whenever they lose self-control.
The algorithm quickly responds to such behaviors. P2 said, ``\textit{\hr{Even after putting effort into removing food videos from my recommended feed, I eventually gave in to my urges and searched for them again. They quickly showed up on my feed once more. For example, when I felt hungry, I intentionally returned to such videos to ease the hunger, but that became the starting point, and I ended up eating a lot even when I was already full.}}'' %quote 1. % 구독 취소뿐만이 아니라 아예 알고리즘에 노출 안 시키려고 일부러 관심 없음. 아니면 이 채널을 더 이상 보여주지 않기 이런 버튼 눌러서 하긴 했는데. 결국은 또 제가 스스로 또 찾다 보니까 그 영상을 네 제가 스스로 이제 먹방을 검색해서 그 영상을 보면 다시 알고리즘에 뜨더라고요.... 배가 고프니까 좀 달래려고 일부러 그 영상을 조금 봤는데 이제 그게 시발점이 되어가지고 결국 엄청 배가 부른데도 굳이 먹고 그렇습니다.
\hr{Some participants described returning to food media not only because of sudden urges but also to recapture the comfort or satisfaction they had previously felt, such as vicarious satisfaction from watching eating broadcasts without actually consuming food.}
% 배달앱이나 account 지우기: 느긋한달팽이, 멋진 표범, 수줍은펭귄, 아름다운고래, 아름다운플라밍고, 온화한플라밍고, 자유로운고래, 자유로운표범, 즐거운다람쥐, 활기찬고양이, 활기찬개구리
Eleven participants reported that they tried to regulate \emph{food delivery app} use.
%To regulate food delivery app use,
All participants mentioned \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A2}uninstalling food delivery apps (N=11)} and some stated \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A3}deleting their accounts (N=3)}.
However, everyone described such efforts often prove ineffective as they tend to reinstall the apps whenever they lose self-control, as reinstalling or creating new accounts is trivial.
To quote P9, ``\textit{I repeatedly uninstall food delivery apps trying not to binge eat, but reinstall it whenever I want to binge eat.}'' % P9 한 번 시켜먹고 이제 이러지 말아야겠다 싶어서 배달 어플을 수시로 지웠다가 또 먹고 싶으면 깔았다가 이러고 있어요. 계속 지금도 깔려 있지는 않는데 또 먹고 싶으면 바로 또 설치하고 이런 식이에요.
%P4 said that she created a shared account with her family in a food delivery app to inform them of her use of the app to prevent herself from ordering food for binge eating and removed other delivery apps that do not have such functionality. % P4: 집에 만약에 부모님이 두 분 다 출근을 하시고 집에 혼자 있으면 이거를 시켜 먹고 증거 인멸을 하면 이게 제가 뭐했는지 모르잖아요. 그래서 배민 같은 경우에는 가족 연동 앱 연동 계좌 뭐지계정 연동이 있더라고요 네 그래서 그거를 해놨고 요기요나 이런 거는 아예 어플을 그냥 삭제를 해봤어요.
Two participants mentioned \textbf{\tcbox[on line, frame empty, boxsep=0pt,left=2pt,right=2pt,top=2pt,bottom=2pt,colback=LightViolet]{A4} using shared accounts~(N=2)} as a strategy to have an extra layer of vigilance in managing their engagement with food delivery apps. They explained that, by using shared accounts linked with family members, they introduced an element of accountability even when they were alone. For instance, P4 shared, ``\textit{If my parents are at work and I am home alone, I might order and binge eat something, and destroy the evidence, and they wouldn't know what I did. So, there is an option for a family account. Yes, I have set that up.}'' %(미연에 방지하기 위해서) 집에 만약에 부모님이 두 분 다 출근을 하시고 집에 혼자 있으면 이거를 시켜 먹고 증거 인멸을 하면 이게 제가 뭐했는지 모르잖아요. 그래서 배민 같은 경우에는 가족 연동 앱 연동 계좌 뭐지계정 연동이 있더라고요 네 그래서 그거를 해놨고
This approach allowed them to create a safeguard against impulsive app usage, as their actions would be visible to others, potentially discouraging disordered eating.
Note, however, that not every food delivery app offers the option of shared accounts.
\hr{In summary, our participants engaged with digital food content for diverse reasons and across multiple subcategories. While this engagement often led to ED-exacerbating impacts, some participants also turned to certain types of food media, such as cooking videos or binge eating broadcasts, as a way to alleviate their symptoms or support recovery. Across both food media consumption and food delivery app use, we observed a recurring pattern in which participants attempted to self-regulate their usage to prevent adverse impacts, yet frequently relapsed. These relapses were sometimes driven by sudden urges, but also by the anticipation of previously experienced benefits, resulting in an ambivalence‑driven cycle where the same content’s recovery-supporting and harmful aspects alternately pulled participants toward and away from engagement.}
%In summary, our participants engaged with digital food content for various reasons. While this engagement often resulted in ED-exacerbating impacts, it is worth noting that some participants also utilized different subcategories of food media, such as cooking and binge eating broadcasts, as a way to alleviate their ED symptoms. In addition, we observed a common perpetual behavioral pattern in both food media consumption and food delivery app use, where participants tried to self-regulate their usage to prevent adverse impacts, but their efforts often went in vain.
\section{Discussion}
\label{sec:discussion}
% specific -> broader (?)
Our study highlighted the conflicting impacts of digital food content on individuals with EDs, perpetuating the cycle in which they regulate their consumption to avoid ED-exacerbating impacts, yet eventually return when losing self-control or with the expectation of supportive effects on their condition. In this section, we discuss the process underlying their content consumption behavior based on dual systems theory~\cite{norman1986attention}, introducing the concept of \emph{ambivalence} (Section~\ref{sec:discussion:ambivalence}).
\subsection{Impact Ambivalence: Cognitive Mechanism of Digital Food Content Consumption} % by Individuals with ED
\label{sec:discussion:ambivalence}
% Rethinking Digital Food Content for Individuals with ED
%-- Impact ambivalence: ED-exacerbating impact vs. Recovery-supporting impact --%
%-- Attitudinal ambivalence: impact평가가 system 2 의사결정에 들어오면서, "그럼 내가 이걸 봐야할까 말아야할까"하는 태도적 양가성이 생김.
\begin{figure*}
\begin{subfigure}{0.4\textwidth}
\includegraphics[width=\linewidth]{fig/expected_value_1.png}
\caption{Evaluation of consuming vs. not consuming digital food content, where different actions lead to distinct, non-conflicting outcomes.} \label{fig:3a}
\end{subfigure}%
\hspace*{1.3cm} %\fill maximize separation between the subfigures
\begin{subfigure}{0.44\textwidth}
\includegraphics[width=\linewidth]{fig/expected_value_2.png}
\caption{Ambivalence in assessing the expected value of consumption, where the same action (engaging in digital food content) may lead to conflicting outcomes, such as recovery-supporting and/or ED-exacerbating impacts.} \label{fig:3b}
\end{subfigure}%
\vspace{-0.3cm}
\caption{Conceptual illustrations of expected value assessments of engaging with digital food content.}
\Description{Figure 2 conceptually illustrates expected value assessments associated with engaging with digital food content. (a) depicts a scale weighing the expected value of content consumption against that of self-regulation. (b) illustrates a bicycle with two people pedaling in opposite directions, symbolizing the conflicting impacts of content consumption.}
\end{figure*}
A recent study framed the cognitive mechanisms of digital food content consumption behavior by individuals with EDs using the extended dual systems model of self-regulation~\cite{choi2024foodcensor}. Dual systems model~\cite{norman1986attention} posits two distinct cognitive decision-making systems: system~1, which is fast, intuitive, and automatic responses, and system~2, which is slow, deliberate, and analytical thinking. System~1 handles routine or immediate decisions, while system~2 engages in more effortful, reflective thinking for complex or high-stakes decisions, ensuring rationality and accuracy but requiring more cognitive resources.
These systems interact dynamically, with system~1 providing quick responses (e.g.,~feeling an impulse to watch a food video when encountering it), which system~2 evaluates (e.g.,~assessing whether watching the food video aligns with personal ED-recovery goals), shaping how we process information and make decisions. When framing the cognitive mechanisms of digital food content consumption, prior work has largely emphasized the adverse effects of digital food content, assuming that individuals use system 2 primarily to regulate consumption in order to minimize harm ~\cite{choi2024foodcensor}.
%prior work focused on its adverse impacts on ED conditions and thus assumed that individuals aim to regulate their consumption to minimize such harm~\cite{choi2024foodcensor}.
However, our study revealed that individuals with EDs sometimes deliberately engage with digital food content to support recovery, such as by drawing motivation from various types of food media. This observation challenges the assumption in prior work that system 2's role is solely harm-avoidant.
Instead, our findings suggest that system~2 also incorporates the potential positive impacts of engaging with such content. In other words, when assessing the expected value of their digital food content consumption behavior, individuals may navigate two layers of decision-making: first, comparing the expected outcomes of consuming versus not consuming digital food content as suggested in the prior work (Figure~\ref{fig:3a}); and second, encountering conflicting expected values, recovery-supporting benefits and the risk of adverse effects, within the act of consumption itself (Figure~\ref{fig:3b}).
% Experiencing ambivalence entails the simultaneous occurrence of positive and negative evaluations toward the same object or situation
\hr{We describe this conflict as \emph{impact ambivalence}, extending the notion of attitudinal ambivalence, the coexistence of positive and negative evaluations toward the same behavior~\cite{thompson2014let, van2009agony, schneider2017mixed, van2015abc}. Impact ambivalence extends this idea to the level of behavioral consequences, where a single act of engagement can generate both supportive and harmful consequences. In our context, digital food content engagement is not just judged ambivalently but experienced ambivalently, producing both recovery-supporting and disorder-exacerbating outcomes. This makes engagement inherently tense, as individuals feel simultaneously drawn to and resistant to consumption, complicating System 2 evaluations with effects that are difficult to reconcile. Because these consequences are often unpredictable, individuals cannot reliably anticipate whether engagement will support recovery, exacerbate symptoms, or do both. This underscores the need for interventions that operate during digital food content engagement,} mitigating negative impacts while allowing potential positive effects to remain. In the following, we present design implications to foster healthier engagement with digital food content while addressing its ambivalent impacts.
%We contextualize this second layer of decision-making with the concept of \emph{attitudinal ambivalence}, defined as the coexistence of positive and negative evaluations toward the same behavior or phenomenon~\cite{thompson2014let, van2009agony, schneider2017mixed, van2015abc}. In our context, \emph{impact ambivalence} captures how digital food content simultaneously presents recovery-supporting benefits and risks of exacerbating disordered eating behaviors, thereby extending this attitudinal ambivalence to the level of behavioral consequences.
%Specifically, our findings highlight how such conflicting impacts render engagement itself ambivalent, with these impact-level tensions ultimately reinforcing attitudinal ambivalence---leaving individuals both desiring and resisting engagement. This inherent conflict complicates system~2 evaluations, as individuals face opposing impacts that are not easily reconciled. Prior work has similarly underscored ambivalence as a core aspect of long-term user experience with technology~\cite{park2023ambivalences}, and our findings extend this insight by showing how impact ambivalence manifests in digital food content consumption.
%In addition, it is important to note that this impact ambivalence embodies the unpredictability of consequences, meaning that individuals cannot reliably predict whether engaging in digital food content will result in recovery-supporting, ED-exacerbating, or both. Therefore, building on previous works that have suggested interventions once a decision has been made and behavior initiated (e.g.,~when a user has been searching for and viewing pro-ED content on websites)~\cite{lewis2012searching, xu2022typeout}, we suggest the need for interventions delivered during digital food content engagement, to moderate the emotional and behavioral consequences of digital food content consumption. Such interventions could be designed to mitigate negative impacts while allowing potential positive effects to remain.
% In-the-moment intervention
\subsection{Promoting Informed and Healthier Content Engagement}
\label{sec:discussion:informed}
\begin{figure*}[t]
\centering
\includegraphics[width=\textwidth]{fig/feedback_media.png}
\vspace{-0.6cm}
\caption{Example wireframes portraying design implications for reflective and collective feedback to surface media's various impacts. \circled{1}~Foster users to reflect on and share their feelings and thoughts while watching. \circled{2}~Provide visibility into others' in-the-moment reflections, surfacing how content may be experienced differently across viewers.}
\Description{Figure 3 is an example wireframe mock-up of a video player interface showing synchronized reflective and collective feedback features. The left side displays a paused food video at timestamp 0:42. Below the video timeline, a synchronized marker links the current time with a reflection entry box, prompting the viewer to add a comment. On the right side, a live comment panel shows time-stamped reflections from other viewers at nearby moments, including emotional responses, personal memories, and comparison concerns. Numbered callouts highlight two design implications: (1) enabling users to write reflections while watching and (2) allowing visibility into others’ in-the-moment reflections.}
\label{fig:feedback_media}
\end{figure*}
Research on decision-making has emphasized the importance of \emph{informed} decision-making~\cite{stacey2017decision, lindrup2021one}. Access to accurate, complete, and contextually relevant information helps individuals evaluate options and align decisions with their goals, values, or needs. Specifically, in dual systems theory, information facilitates system~2's deliberate and analytical thinking to match their decisions with their goals.
We found individuals with EDs have engaged with digital food content either by expecting positive impacts or through unconscious, habitual consumption. Within the framework of dual systems theory, this indicates that engagement decisions are often driven by system~2's expected value assessments, focusing mainly on positive outcomes, or by system~1's automatic processes. Meanwhile, our study stresses that engaging with digital food content sometimes carries ED-exacerbating impacts misaligned with individuals' expected recovery-supporting benefits, prompting our participants to spontaneously self-regulate their consumption behaviors. In this regard, we suggest that it is crucial to provide individuals with information that helps them recognize and evaluate the potential negative impacts of digital food content alongside the positive ones in their engagement decision-making process. By incorporating this information into system~2's expected value assessment, individuals could make balanced choices.
\subsubsection{Reflective and Collective Feedback to Uncover Media’s Ambivalent Impacts} % on Media Platforms
Our findings reveal that individuals with EDs engage with and respond to digital food content differently from the general population. While media platforms often implement content warnings to protect vulnerable users (e.g.,~content warnings on violent content for children)~\cite{charles2022typology, x2024your, netflix2021photosensitivity}, these warnings are typically designed for broad audiences and struggle to capture the nuanced needs of diverse populations~\cite{haimson2020trans}. As a result, less visible harms, such as the exacerbating impacts of digital food content on individuals with ED, are often overlooked.
To address these limitations, future research should investigate alternative deliberate approaches that could surface both overlooked harms and benefits. One promising direction is enabling viewers to articulate and share their emotional and cognitive responses while watching videos~(Figure~\ref{fig:feedback_media}~\circled{1}). This aligns with evidence that expressive writing can facilitate reflection and meaning-making, and that articulating reactions in situ can heighten self-awareness~\cite{smyth1998written}. Building on this, future research could also investigate interventions that make such reflections visible across viewers (Figure~\ref{fig:feedback_media}~\circled{2}), offering not only individual insight but also transparency into how the same media can be experienced differently by others.
When aggregated across multiple viewers, these responses could provide a form of \emph{social translucency}, surfacing how different segments and features of content are experienced. The aggregated view may highlight collective patterns---where reactions tend to converge---as well as points of divergence across subgroups, revealing how the same content may be interpreted differently depending on viewers' characteristics or vulnerabilities. This collective layer could help media platforms and creators identify where ambivalent impacts occur, for example, in our context, moments that simultaneously elicit recovery-supporting benefits and disorder-exacerbating risks. Further research is needed to explore how such insights might be leveraged to detect critical ambivalent impacts and explore ways of mediating them toward more positive outcomes. Rather than relying solely on one-size-fits-all warnings, viewer-informed signals may open pathways toward more personalized support for informed engagement.
We note, however, that such interventions could introduce unintentional risks, and future work should therefore investigate how reflective and collective feedback can be responsibly incorporated to support informed media engagement without creating new challenges.
\subsubsection{Prompts to Support Mindful Usage of Food Delivery Apps}
% Supportive Food Delivery App Environment
We identified a recurring pattern of uninstalling and reinstalling food delivery apps among individuals with EDs to avoid unhealthy usage of these apps. To foster healthier engagement, we suggest interventions to promote mindful food delivery app use.
For instance, a recent study demonstrated the effectiveness of reflective prompts in encouraging mindful engagement with food-related videos on YouTube for individuals with EDs~\cite{choi2024foodcensor}. In that study, participants were shown short prompts, such as ``Why do you want to watch this video now?'', which encouraged them to pause, evaluate their motivations, and sometimes disengage from potentially triggering content. These prompts did not block access but created a small reflective moment that supported more deliberate media choices. In the context of food delivery apps, a similar in-the-moment intervention could be triggered upon app entry, for example, asking, ``\textit{Will this order support your recovery goals today?}''. Such prompts could support mindful app usage~\cite{xu2022typeout}. A potential advantage is that these lightweight interventions respect autonomy and allow users to continue if they choose, while still nudging reflection. However, drawbacks include the risk of alert fatigue if prompts appear too frequently, or the possibility that prompts may feel judgmental and increase guilt, which could undermine recovery.
%We also identified opportunities to enhance social features within food delivery apps to encourage healthy eating. Two participants used the shared account feature with family members to manage their app usage, making their food ordering through the app visible to others to avoid disordered eating. Taking this further, within the context of family informatics, it is also important to recognize `\emph{what is not done}' by family members (e.g., not using the app, or using it without engaging in disordered eating)~\cite{lukoff2018tablechat}. Further research could explore interventions that enable family members to support each other's dietary health, including managing eating disorders, by making restraint or healthy usage more visible. For example, an intervention could provide a shared family dashboard that tracks and highlights days without using food delivery apps or excessive food orders. Additionally, the dashboard could be prompted as a reminder of the visibility of their app usage upon entering a food delivery app. Such an intervention could reinforce more mindful app usage while fostering accountability within families.
% Opportune Moments to Prevent Disordered Eating Post-
\subsubsection{Opportune Moments to Prevent Disordered Eating Triggered by Digital Food Content}
\label{sec:discussion:opportune}
% After Digital Food Content Consumption ~ Before Disordered Eating: Prevent ED-Promoting
The impact ambivalence of digital food content makes it difficult for our participants to assess the outcomes of engaging with such content. This unpredictability poses significant challenges for intervention design, particularly in determining when and how to intervene. For instance, interventions that aim to regulate engagement with digital food content to prevent its negative impacts~\cite{choi2024foodcensor} might inadvertently block opportunities for users to gain recovery-supporting benefits. In this regard, further intervention designs for healthy engagement with digital food content should account for the complex mechanisms underlying its conflicting impacts.
To guide the design of effective interventions, we suggest that researchers explore how users interact with different types of digital food content. These interactions are influenced by various motivations and impacts, and understanding them can help create strategies to prevent disordered eating behaviors.
% (1) food delivery app 보고 있는 중
% From our study, we recognized
Recognizing the uncertainty of impacts, our study identified a few moments that could offer a second chance to steer users away from the impacts that exacerbate EDs subsequent to engaging with digital food content. Some participants indicated prolonged browsing or frequent engagement with food delivery apps often precedes binge eating episodes, implying opportune moments for situated intervention.
In addition, our participants reported that food media sometimes triggered their use of food delivery apps, driven by the urge to eat the foods featured due to the media's visual and auditory stimuli. This sequence of engaging with food media and food delivery apps could imply their ED-exacerbating impacts, such as visual and auditory stimuli of food media and the low barrier of getting food from food delivery apps. These usage patterns detail the practices associated with ED-exacerbating impacts of digital food content, suggesting critical moments for timely interventions to mitigate its adverse effects. %These usage patterns detail the practices associated with overeating in food delivery apps~\cite{portingale2023tonight}. %We propose researchers investigate users' varying interaction patterns with digital food content based on different motivations and impacts and design interventions to deter disordered eating behaviors.
% (2) food delivery app으로 음식 이미 시켜버렸을 때
Another opportune moment we propose to leverage is the waiting period between ordering and food delivery on food delivery apps. Previous research has explored digital interventions such as showing videos that teach mindful eating or promote healthy body images just before meals to encourage healthy eating behaviors~\cite{allirot2018effects, zepeda2008think}. However, these interventions often require proactive user engagement, such as manually opening and watching the videos on their devices. By detecting associated app features or screens (e.g.,~food waiting screens), such interventions could be passively triggered and seamlessly integrated into the user experience.
For instance, an intervention could prompt users with short, guided questions, such as reflecting on their feelings or hunger levels to promote mindful eating~\cite{luo2019co}, while waiting for delivery. Such strategies could help foster self-reflection on food ordering, reduce impulsivity, and disrupt planned binge eating.
\subsection{Mindful Curation of Digital Food Content for Dietary Well-being}
\label{sec:discussion:curation}
In the field of human-food technology interaction, there has been a growing emergence of research focused on promoting healthy eating~\cite{khot2022understanding, covaci2023no}. Simultaneously, a substantial body of research has been dedicated to improving the eating experience through enhancing the realism or stimuli of virtual food~\cite{velasco2016multisensory, anjani2020people}. Recent studies have also investigated how tech-mediated auditory stimuli, such as different music playing or manipulating eating sounds (e.g.,~amplifying and delaying)~\cite{peng2021sounds, kleinberger2023auditory}, affect users' eating practices. Another research demonstrated that subtly adjusting the playback speed of food videos during eating could promote slower eating and more controlled intake~\cite{chen2025vifeed}. Understanding the impact of various stimuli related to food, as delivered through various technologies, on eating practices is essential. Coupled with our study, we believe this understanding could create a more health-conscious and enjoyable experience in food multimedia design and curation, ultimately enhancing users' experiences.
Our findings stress the importance of carefully curating food multimedia content that is increasingly captivating, engaging, and often exacerbating disordered eating~\cite{kinkel2022food}. Mindful digital food content curation is particularly important when promoting healthy eating among populations highly susceptible to the influential nature of digital content, such as individuals with EDs~\cite{choi2024foodcensor}. %It could offer a more supportive and balanced approach in response to the challenges posed by provocative food content.
Contrary to the prevailing content recommendation strategy, which reinforces users' exposure to similar content by relying on their recent viewing history to attract users~\cite{chaney2018algorithmic, youtubeSuggestionAlgorithm}, a more mindful curation approach is needed. %by focusing on understanding users' sensitivities and actively counteracting potential negative impacts. % youTubeReccommendation, instagramFeed, tickTokRecommendation, twitterRecommendation
%In light of our findings, HCI designers should adopt a more nuanced and empathetic approach to content delivery, recognizing the potential adverse effects of stimulus maximization on this vulnerable user group.
% (1) Counterbalancing stimulating content with healthier recommendations
For instance, multimedia platform designers could consider adopting a counterbalancing approach to prevent users from becoming trapped in the `filter bubble,' where they are repeatedly exposed to the same type of content~\cite{nguyen2014exploring}. For example, when users have heavily engaged with food media, the platform could strategically introduce content on different topics. This intentional diversification of content could help mitigate unconscious engagement with triggering food media for individuals with EDs by reducing their passive exposure to food-related media. Recent research also demonstrated that reducing exposure to food-related videos on YouTube can alleviate their negative impacts on individuals with EDs, leading to fewer obsessive thoughts about foods and overall improvement in quality of life~\cite{choi2024foodcensor}.
However, implementing such a counterbalancing strategy requires carefully considering the overall user experience. If not thoughtfully designed, this approach could disrupt the flow of content the users find engaging, leading to frustration or reduced platform satisfaction~\cite{wei2023measuring}. Therefore, future work should explore balancing content curation that breaks the filter bubble, supporting users' well-being while maintaining the recommended content attractive and engaging. Additionally, future research could explore strategies for providing a counterbalance to users and assessing its impact on both user engagement and health outcomes.
\subsection{Limitations}
Our study predominantly included participants residing in South Korea, with only two individuals from North America and Europe. While food media and food delivery apps have achieved global popularity~\cite{anjani2020people, onlineFoodDelivery}, the limited demographic diversity of our participants means that our findings may not be fully generalizable across different cultural and geographic contexts. %Additionally, the study's gender composition was predominantly female, reflecting the demographics associated with BED and BN in South Korea, where a significant majority (87.4\%~of identified BED patients) are female~\cite{kessler2013prevalence, galmiche2019prevalence, KoreaBEDProportion}.
\section{Conclusion}
We examined how individuals with eating disorders perceive and interact with digital food content. Our study offers insights into the complex dynamics between individuals with eating disorders and digital food content, uncovering its ambivalent roles as a recovery-supporting tool and a trigger for disordered eating behaviors. We demonstrated how this \emph{impact ambivalence} creates significant tension, perpetuating behavioral patterns characterized by cycles of self-regulated content engagement and relapse. Our design implications and proposed strategies for promoting mindful engagement with digital food content, and \hr{the potential opportune moments we identified to mediate its negative} aftereffects, \hr{together provide} a clear foundation for HCI designers to develop more personalized and supportive digital environments.
By addressing individuals with eating disorders' unique needs, HCI researchers and practitioners could contribute to shaping healthy and empowering digital spaces. Our research contributes to the HCI discourse by examining the potential for more responsible and health-conscious design practices in digital food content.
%directions for mindful digital food content engagement and moderating post-engagement impacts} provide a strategic foundation for HCI designers to create more tailored and supportive digital environments.
%This impact ambivalence often results in the disparity between anticipated positive effects and actual adverse outcomes of consuming such content.
%The disparity between anticipated positive effects and actual outcomes of consuming such content underscores the need for health-conscious food content design to prevent unintended negative impacts on actual eating. Our design implications and proposed directions for mindful content curation and enhanced user control over digital food content provide a strategic foundation for HCI designers to create more tailored and supportive digital food content platforms. By addressing individuals with eating disorders' unique needs, HCI researchers and practitioners could contribute to developing healthier and empowering digital environments. Our research contributes to the HCI discourse by examining the potential for more responsible and sensitive design practices in the realm of digital food content.
%%% -*-BibTeX-*-
%%% Do NOT edit. File created by BibTeX with style
%%% ACM-Reference-Format-Journals [18-Jan-2012].
\begin{thebibliography}{100}
%%% ====================================================================
%%% NOTE TO THE USER: you can override these defaults by providing
%%% customized versions of any of these macros before the \bibliography
%%% command. Each of them MUST provide its own final punctuation,
%%% except for \shownote{}, \showDOI{}, and \showURL{}. The latter two
%%% do not use final punctuation, in order to avoid confusing it with
%%% the Web address.
%%%
%%% To suppress output of a particular field, define its macro to expand
%%% to an empty string, or better, \unskip, like this:
%%%
%%% \newcommand{\showDOI}[1]{\unskip} % LaTeX syntax
%%%
%%% \def \showDOI #1{\unskip} % plain TeX syntax
%%%
%%% ====================================================================
\ifx \showCODEN \undefined \def \showCODEN #1{\unskip} \fi
\ifx \showDOI \undefined \def \showDOI #1{#1}\fi
\ifx \showISBNx \undefined \def \showISBNx #1{\unskip} \fi
\ifx \showISBNxiii \undefined \def \showISBNxiii #1{\unskip} \fi
\ifx \showISSN \undefined \def \showISSN #1{\unskip} \fi
\ifx \showLCCN \undefined \def \showLCCN #1{\unskip} \fi
\ifx \shownote \undefined \def \shownote #1{#1} \fi
\ifx \showarticletitle \undefined \def \showarticletitle #1{#1} \fi
\ifx \showURL \undefined \def \showURL {\relax} \fi
% The following commands are used for tagged output and should be
% invisible to TeX
\providecommand\bibfield[2]{#2}
\providecommand\bibinfo[2]{#2}
\providecommand\natexlab[1]{#1}
\providecommand\showeprint[2][]{arXiv:#2}
\bibitem[Aardoom et~al\mbox{.}(2012)]%
{aardoom2012norms}
\bibfield{author}{\bibinfo{person}{Jiska~J Aardoom}, \bibinfo{person}{Alexandra~E Dingemans}, \bibinfo{person}{Margarita CT~Slof Op't~Landt}, {and} \bibinfo{person}{Eric~F Van~Furth}.} \bibinfo{year}{2012}\natexlab{}.
\newblock \showarticletitle{Norms and discriminative validity of the Eating Disorder Examination Questionnaire (EDE-Q)}.
\newblock \bibinfo{journal}{\emph{Eating behaviors}} \bibinfo{volume}{13}, \bibinfo{number}{4} (\bibinfo{year}{2012}), \bibinfo{pages}{305--309}.
\newblock
\bibitem[Adema(2000)]%
{adema2000vicarious}
\bibfield{author}{\bibinfo{person}{Pauline Adema}.} \bibinfo{year}{2000}\natexlab{}.
\newblock \showarticletitle{Vicarious comsumption: Food, television and the ambiguity of modernity}.
\newblock \bibinfo{journal}{\emph{The journal of American culture}} \bibinfo{volume}{23}, \bibinfo{number}{3} (\bibinfo{year}{2000}), \bibinfo{pages}{113}.
\newblock
\bibitem[Allirot et~al\mbox{.}(2018)]%
{allirot2018effects}
\bibfield{author}{\bibinfo{person}{Xavier Allirot}, \bibinfo{person}{Marta Miragall}, \bibinfo{person}{I{\~n}igo Perdices}, \bibinfo{person}{Rosa~Maria Ba{\~n}os}, \bibinfo{person}{Elena Urdaneta}, {and} \bibinfo{person}{Ausias Cebolla}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{Effects of a brief mindful eating induction on food choices and energy intake: external eating and mindfulness state as moderators}.
\newblock \bibinfo{journal}{\emph{Mindfulness}} \bibinfo{volume}{9} (\bibinfo{year}{2018}), \bibinfo{pages}{750--760}.
\newblock
\bibitem[Anari and Eghtesadi(2023)]%
{anari2023relationship}
\bibfield{author}{\bibinfo{person}{Fatemeh~Manafi Anari} {and} \bibinfo{person}{Shahryar Eghtesadi}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \showarticletitle{The Relationship Between Watching Mukbang (Eating Show), Eating Behaviors, and Anthropometric Parameters in Iranian Female Students}.
\newblock \bibinfo{journal}{\emph{Journal of Research in Health Sciences}} \bibinfo{volume}{23}, \bibinfo{number}{1} (\bibinfo{year}{2023}).
\newblock
\bibitem[Anjani et~al\mbox{.}(2020)]%
{anjani2020people}
\bibfield{author}{\bibinfo{person}{Laurensia Anjani}, \bibinfo{person}{Terrance Mok}, \bibinfo{person}{Anthony Tang}, \bibinfo{person}{Lora Oehlberg}, {and} \bibinfo{person}{Wooi~Boon Goh}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{Why do people watch others eat food? An Empirical Study on the Motivations and Practices of Mukbang Viewers}. In \bibinfo{booktitle}{\emph{Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--13}.
\newblock
\bibitem[Aparicio-Martinez et~al\mbox{.}(2019)]%
{aparicio2019social}
\bibfield{author}{\bibinfo{person}{Pilar Aparicio-Martinez}, \bibinfo{person}{Alberto-Jesus Perea-Moreno}, \bibinfo{person}{Mar{\'\i}a~Pilar Martinez-Jimenez}, \bibinfo{person}{Mar{\'\i}a~Dolores Redel-Mac{\'\i}as}, \bibinfo{person}{Claudia Pagliari}, {and} \bibinfo{person}{Manuel Vaquero-Abellan}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{Social media, thin-ideal, body dissatisfaction and disordered eating attitudes: An exploratory analysis}.
\newblock \bibinfo{journal}{\emph{International journal of environmental research and public health}} \bibinfo{volume}{16}, \bibinfo{number}{21} (\bibinfo{year}{2019}), \bibinfo{pages}{4177}.
\newblock
\bibitem[Bachner-Melman et~al\mbox{.}(2018)]%
{bachner2018lives}
\bibfield{author}{\bibinfo{person}{Rachel Bachner-Melman}, \bibinfo{person}{Einat Zontag-Oren}, \bibinfo{person}{Ada~H Zohar}, {and} \bibinfo{person}{Helene Sher}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{Lives on the line: The online lives of girls and women with and without a lifetime eating disorder diagnosis}.
\newblock \bibinfo{journal}{\emph{Frontiers in psychology}} \bibinfo{volume}{9} (\bibinfo{year}{2018}), \bibinfo{pages}{2128}.
\newblock
\bibitem[Basso et~al\mbox{.}(2018)]%
{basso2018taste}
\bibfield{author}{\bibinfo{person}{Fr{\'e}d{\'e}ric Basso}, \bibinfo{person}{Olivia Petit}, \bibinfo{person}{Sophie Le~Bellu}, \bibinfo{person}{Saadi Lahlou}, \bibinfo{person}{A{\"\i}da Cancel}, {and} \bibinfo{person}{Jean-Luc Anton}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{Taste at first (person) sight: Visual perspective modulates brain activity implicitly associated with viewing unhealthy but not healthy foods}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{128} (\bibinfo{year}{2018}), \bibinfo{pages}{242--254}.
\newblock
\bibitem[Bodenlos and Wormuth(2013)]%
{bodenlos2013watching}
\bibfield{author}{\bibinfo{person}{Jamie~S Bodenlos} {and} \bibinfo{person}{Bernadette~M Wormuth}.} \bibinfo{year}{2013}\natexlab{}.
\newblock \showarticletitle{Watching a food-related television show and caloric intake. A laboratory study}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{61} (\bibinfo{year}{2013}), \bibinfo{pages}{8--12}.
\newblock
\bibitem[Braun and Clarke(2012)]%
{braun2012thematic}
\bibfield{author}{\bibinfo{person}{Virginia Braun} {and} \bibinfo{person}{Victoria Clarke}.} \bibinfo{year}{2012}\natexlab{}.
\newblock \bibinfo{booktitle}{\emph{Thematic analysis.}}
\newblock \bibinfo{publisher}{American Psychological Association}.
\newblock
\bibitem[Center(2021)]%
{netflix2021photosensitivity}
\bibfield{author}{\bibinfo{person}{Netflix Partner~Help Center}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \bibinfo{title}{Photosensitivity Prevention Best Practice Guildeline - Production}.
\newblock \bibinfo{howpublished}{\url{https://partnerhelp.netflixstudios.com/hc/en-us/articles/360058732874-Photosensitivity-Prevention-Best-Practice-Guideline-Production}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Center(2024)]%
{x2024your}
\bibfield{author}{\bibinfo{person}{X~Help Center}.} \bibinfo{year}{2024}\natexlab{}.
\newblock \bibinfo{title}{Your media settings}.
\newblock \bibinfo{howpublished}{\url{https://help.x.com/en/rules-and-policies/media-settings}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Chancellor et~al\mbox{.}(2017)]%
{chancellor2017multimodal}
\bibfield{author}{\bibinfo{person}{Stevie Chancellor}, \bibinfo{person}{Yannis Kalantidis}, \bibinfo{person}{Jessica~A Pater}, \bibinfo{person}{Munmun De~Choudhury}, {and} \bibinfo{person}{David~A Shamma}.} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{Multimodal classification of moderated online pro-eating disorder content}. In \bibinfo{booktitle}{\emph{Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{3213--3226}.
\newblock
\bibitem[Chancellor et~al\mbox{.}(2016a)]%
{chancellor2016post}
\bibfield{author}{\bibinfo{person}{Stevie Chancellor}, \bibinfo{person}{Zhiyuan Lin}, {and} \bibinfo{person}{Munmun De~Choudhury}.} \bibinfo{year}{2016}\natexlab{a}.
\newblock \showarticletitle{" This Post Will Just Get Taken Down" Characterizing Removed Pro-Eating Disorder Social Media Content}. In \bibinfo{booktitle}{\emph{Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1157--1162}.
\newblock
\bibitem[Chancellor et~al\mbox{.}(2016b)]%
{chancellor2016thyghgapp}
\bibfield{author}{\bibinfo{person}{Stevie Chancellor}, \bibinfo{person}{Jessica~Annette Pater}, \bibinfo{person}{Trustin Clear}, \bibinfo{person}{Eric Gilbert}, {and} \bibinfo{person}{Munmun De~Choudhury}.} \bibinfo{year}{2016}\natexlab{b}.
\newblock \showarticletitle{\# thyghgapp: Instagram content moderation and lexical variation in pro-eating disorder communities}. In \bibinfo{booktitle}{\emph{Proceedings of the 19th ACM conference on computer-supported cooperative work \& social computing}}. \bibinfo{pages}{1201--1213}.
\newblock
\bibitem[Chaney et~al\mbox{.}(2018)]%
{chaney2018algorithmic}
\bibfield{author}{\bibinfo{person}{Allison~JB Chaney}, \bibinfo{person}{Brandon~M Stewart}, {and} \bibinfo{person}{Barbara~E Engelhardt}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{How algorithmic confounding in recommendation systems increases homogeneity and decreases utility}. In \bibinfo{booktitle}{\emph{Proceedings of the 12th ACM conference on recommender systems}}. \bibinfo{pages}{224--232}.
\newblock
\bibitem[Charles et~al\mbox{.}(2022)]%
{charles2022typology}
\bibfield{author}{\bibinfo{person}{Ashleigh Charles}, \bibinfo{person}{Laurie Hare-Duke}, \bibinfo{person}{Hannah Nudds}, \bibinfo{person}{Donna Franklin}, \bibinfo{person}{Joy Llewellyn-Beardsley}, \bibinfo{person}{Stefan Rennick-Egglestone}, \bibinfo{person}{Onni Gust}, \bibinfo{person}{Fiona Ng}, \bibinfo{person}{Elizabeth Evans}, \bibinfo{person}{Emily Knox}, {et~al\mbox{.}}} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{Typology of content warnings and trigger warnings: Systematic review}.
\newblock \bibinfo{journal}{\emph{PloS one}} \bibinfo{volume}{17}, \bibinfo{number}{5} (\bibinfo{year}{2022}), \bibinfo{pages}{e0266722}.
\newblock
\bibitem[Chen et~al\mbox{.}(2025)]%
{chen2025vifeed}
\bibfield{author}{\bibinfo{person}{Yang Chen}, \bibinfo{person}{Felicia Fang-Yi Tan}, \bibinfo{person}{Zhuoyu Wang}, \bibinfo{person}{Xing Liu}, \bibinfo{person}{Jiayi Zhang}, \bibinfo{person}{Yun Huang}, \bibinfo{person}{Shengdong Zhao}, {and} \bibinfo{person}{Ching~Chiuan Yen}.} \bibinfo{year}{2025}\natexlab{}.
\newblock \showarticletitle{ViFeed: Promoting Slow Eating and Food Awareness through Strategic Video Manipulation during Screen-Based Dining}. In \bibinfo{booktitle}{\emph{Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--24}.
\newblock
\bibitem[Choi et~al\mbox{.}(2025)]%
{choi2025private}
\bibfield{author}{\bibinfo{person}{Ryuhaerang Choi}, \bibinfo{person}{Taehan Kim}, \bibinfo{person}{Subin Park}, \bibinfo{person}{Jennifer~G Kim}, {and} \bibinfo{person}{Sung-Ju Lee}.} \bibinfo{year}{2025}\natexlab{}.
\newblock \showarticletitle{Private Yet Social: How LLM Chatbots Support and Challenge Eating Disorder Recovery}. In \bibinfo{booktitle}{\emph{Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--19}.
\newblock
\bibitem[Choi et~al\mbox{.}(2024)]%
{choi2024foodcensor}
\bibfield{author}{\bibinfo{person}{Ryuhaerang Choi}, \bibinfo{person}{Subin Park}, \bibinfo{person}{Sujin Han}, {and} \bibinfo{person}{Sung-Ju Lee}.} \bibinfo{year}{2024}\natexlab{}.
\newblock \showarticletitle{FoodCensor: Promoting Mindful Digital Food Content Consumption for People with Eating Disorders}. In \bibinfo{booktitle}{\emph{Proceedings of the CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--18}.
\newblock
\bibitem[cop.(2018)]%
{KoreanEatingDisorderSocialSupportCommunity}
\bibfield{author}{\bibinfo{person}{Naver cop.}} \bibinfo{year}{2018}\natexlab{}.
\newblock \bibinfo{title}{Online Social Support Community for People with Eating Disorders}.
\newblock \bibinfo{howpublished}{\url{https://cafe.naver.com/jahayun}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Covaci et~al\mbox{.}(2023)]%
{covaci2023no}
\bibfield{author}{\bibinfo{person}{Alexandra Covaci}, \bibinfo{person}{Khawla Alhasan}, \bibinfo{person}{Mayank Loonker}, \bibinfo{person}{Bernardine Farrell}, \bibinfo{person}{Luma Tabbaa}, \bibinfo{person}{Sophia Ppali}, {and} \bibinfo{person}{Chee~Siang Ang}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \showarticletitle{No Pie in the (Digital) Sky: Co-Imagining the Food Metaverse}. In \bibinfo{booktitle}{\emph{Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--17}.
\newblock
\bibitem[De~Solier(2005)]%
{de2005tv}
\bibfield{author}{\bibinfo{person}{Isabelle De~Solier}.} \bibinfo{year}{2005}\natexlab{}.
\newblock \showarticletitle{TV dinners: Culinary television, education and distinction}.
\newblock \bibinfo{journal}{\emph{Continuum}} \bibinfo{volume}{19}, \bibinfo{number}{4} (\bibinfo{year}{2005}), \bibinfo{pages}{465--481}.
\newblock
\bibitem[Dingemans et~al\mbox{.}(2002)]%
{dingemans2002binge}
\bibfield{author}{\bibinfo{person}{AE Dingemans}, \bibinfo{person}{MJ Bruna}, {and} \bibinfo{person}{EF Van~Furth}.} \bibinfo{year}{2002}\natexlab{}.
\newblock \showarticletitle{Binge eating disorder: a review}.
\newblock \bibinfo{journal}{\emph{International journal of obesity}} \bibinfo{volume}{26}, \bibinfo{number}{3} (\bibinfo{year}{2002}), \bibinfo{pages}{299--307}.
\newblock
\bibitem[Donnar(2017)]%
{donnar2017food}
\bibfield{author}{\bibinfo{person}{Glen Donnar}.} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{‘Food porn’or intimate sociality: Committed celebrity and cultural performances of overeating in meokbang}.
\newblock \bibinfo{journal}{\emph{Celebrity Studies}} \bibinfo{volume}{8}, \bibinfo{number}{1} (\bibinfo{year}{2017}), \bibinfo{pages}{122--127}.
\newblock
\bibitem[Drivas et~al\mbox{.}(2024)]%
{drivas2024whatieatinaday}
\bibfield{author}{\bibinfo{person}{Magdalayna Drivas}, \bibinfo{person}{Olivia~Simone Reed}, {and} \bibinfo{person}{Maranda Berndt-Goke}.} \bibinfo{year}{2024}\natexlab{}.
\newblock \showarticletitle{\# WhatIEatInADay: The effects of viewing food diary TikTok videos on young adults’ body image and intent to diet}.
\newblock \bibinfo{journal}{\emph{Body Image}} \bibinfo{volume}{49} (\bibinfo{year}{2024}), \bibinfo{pages}{101712}.
\newblock
\bibitem[Eu and Sameeha(2021)]%
{eu2021consumers}
\bibfield{author}{\bibinfo{person}{Elisa Zhen~Rong Eu} {and} \bibinfo{person}{Mohd~Jamil Sameeha}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{Consumers' perceptions of healthy food availability in online food delivery applications (OFD apps) and its association with food choices among public university students in Malaysia}.
\newblock \bibinfo{journal}{\emph{Frontiers in nutrition}} (\bibinfo{year}{2021}), \bibinfo{pages}{522}.
\newblock
\bibitem[Fairburn and Beglin(2008)]%
{fairburn2008eating}
\bibfield{author}{\bibinfo{person}{Christopher~G Fairburn} {and} \bibinfo{person}{Sarah~J Beglin}.} \bibinfo{year}{2008}\natexlab{}.
\newblock \showarticletitle{Eating disorder examination questionnaire}.
\newblock \bibinfo{journal}{\emph{Cognitive behavior therapy and eating disorders}} \bibinfo{volume}{309} (\bibinfo{year}{2008}), \bibinfo{pages}{313}.
\newblock
\bibitem[Fairburn and Garner(1986)]%
{fairburn1986diagnosis}
\bibfield{author}{\bibinfo{person}{Christopher~G Fairburn} {and} \bibinfo{person}{David~M Garner}.} \bibinfo{year}{1986}\natexlab{}.
\newblock \showarticletitle{The diagnosis of bulimia nervosa}.
\newblock \bibinfo{journal}{\emph{International Journal of Eating Disorders}} \bibinfo{volume}{5}, \bibinfo{number}{3} (\bibinfo{year}{1986}), \bibinfo{pages}{403--419}.
\newblock
\bibitem[Feuston et~al\mbox{.}(2020)]%
{feuston2020conformity}
\bibfield{author}{\bibinfo{person}{Jessica~L Feuston}, \bibinfo{person}{Alex~S Taylor}, {and} \bibinfo{person}{Anne~Marie Piper}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{Conformity of eating disorders through content moderation}.
\newblock \bibinfo{journal}{\emph{Proceedings of the ACM on Human-Computer Interaction}} \bibinfo{volume}{4}, \bibinfo{number}{CSCW1} (\bibinfo{year}{2020}), \bibinfo{pages}{1--28}.
\newblock
\bibitem[Fortune(2019)]%
{youtubeSuggestionAlgorithm}
\bibfield{author}{\bibinfo{person}{Fortune}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \bibinfo{title}{On YouTube’s recommendation system}.
\newblock \bibinfo{howpublished}{\url{https://blog.youtube/inside-youtube/on-youtubes-recommendation-system/}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Gak et~al\mbox{.}(2022)]%
{gak2022distressing}
\bibfield{author}{\bibinfo{person}{Liza Gak}, \bibinfo{person}{Seyi Olojo}, {and} \bibinfo{person}{Niloufar Salehi}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{The distressing ads that persist: Uncovering the harms of targeted weight-loss ads among users with histories of disordered eating}.
\newblock \bibinfo{journal}{\emph{Proceedings of the ACM on Human-Computer Interaction}} \bibinfo{volume}{6}, \bibinfo{number}{CSCW2} (\bibinfo{year}{2022}), \bibinfo{pages}{1--23}.
\newblock
\bibitem[Galmiche et~al\mbox{.}(2019)]%
{galmiche2019prevalence}
\bibfield{author}{\bibinfo{person}{Marie Galmiche}, \bibinfo{person}{Pierre D{\'e}chelotte}, \bibinfo{person}{Gr{\'e}gory Lambert}, {and} \bibinfo{person}{Marie~Pierre Tavolacci}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{Prevalence of eating disorders over the 2000--2018 period: a systematic literature review}.
\newblock \bibinfo{journal}{\emph{The American journal of clinical nutrition}} \bibinfo{volume}{109}, \bibinfo{number}{5} (\bibinfo{year}{2019}), \bibinfo{pages}{1402--1413}.
\newblock
\bibitem[Haimson et~al\mbox{.}(2020)]%
{haimson2020trans}
\bibfield{author}{\bibinfo{person}{Oliver~L Haimson}, \bibinfo{person}{Justin Buss}, \bibinfo{person}{Zu Weinger}, \bibinfo{person}{Denny~L Starks}, \bibinfo{person}{Dykee Gorrell}, {and} \bibinfo{person}{Briar~Sweetbriar Baron}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{Trans time: Safety, privacy, and content warnings on a transgender-specific social media site}.
\newblock \bibinfo{journal}{\emph{Proceedings of the ACM on Human-Computer Interaction}} \bibinfo{volume}{4}, \bibinfo{number}{CSCW2} (\bibinfo{year}{2020}), \bibinfo{pages}{1--27}.
\newblock
\bibitem[Halford et~al\mbox{.}(2004)]%
{halford2004effect}
\bibfield{author}{\bibinfo{person}{Jason~CG Halford}, \bibinfo{person}{Jane Gillespie}, \bibinfo{person}{Victoria Brown}, \bibinfo{person}{Eleanor~E Pontin}, {and} \bibinfo{person}{Terence~M Dovey}.} \bibinfo{year}{2004}\natexlab{}.
\newblock \showarticletitle{Effect of television advertisements for foods on food consumption in children}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{42}, \bibinfo{number}{2} (\bibinfo{year}{2004}), \bibinfo{pages}{221--225}.
\newblock
\bibitem[Hudson et~al\mbox{.}(2007)]%
{hudson2007prevalence}
\bibfield{author}{\bibinfo{person}{James~I Hudson}, \bibinfo{person}{Eva Hiripi}, \bibinfo{person}{Harrison~G Pope~Jr}, {and} \bibinfo{person}{Ronald~C Kessler}.} \bibinfo{year}{2007}\natexlab{}.
\newblock \showarticletitle{The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication}.
\newblock \bibinfo{journal}{\emph{Biological psychiatry}} \bibinfo{volume}{61}, \bibinfo{number}{3} (\bibinfo{year}{2007}), \bibinfo{pages}{348--358}.
\newblock
\bibitem[Insights(2023)]%
{onlineFoodDelivery}
\bibfield{author}{\bibinfo{person}{Statistica~Market Insights}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \bibinfo{title}{Online Food Delivery - Worldwide}.
\newblock \bibinfo{howpublished}{\url{https://www.statista.com/outlook/dmo/online-food-delivery/worldwide}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Instagram(2024)]%
{InstaFoodporn}
\bibfield{author}{\bibinfo{person}{Instagram}.} \bibinfo{year}{2024}\natexlab{}.
\newblock \bibinfo{title}{Instagrm Search by \#foodporn}.
\newblock \bibinfo{howpublished}{\url{https://www.instagram.com/explore/tags/foodporn/}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[International(2025)]%
{foodservice_delivery}
\bibfield{author}{\bibinfo{person}{Euromonitor International}.} \bibinfo{year}{2025}\natexlab{}.
\newblock \bibinfo{title}{1 in 5 consumers chose foodservice delivery over dine-in in 2024}.
\newblock \bibinfo{howpublished}{\url{https://www.euromonitor.com/newsroom/press-releases/july-2025/1-in-5-consumers-chose-foodservice-delivery-over-dine-in-in-2024}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Jin(2018)]%
{jin2018interactive}
\bibfield{author}{\bibinfo{person}{Seunga~Venus Jin}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{Interactive effects of Instagram Foodies' Hashtagged\# Foodporn and peer Users' eating disorder on eating intention, envy, Parasocial interaction, and online friendship}.
\newblock \bibinfo{journal}{\emph{Cyberpsychology, Behavior, and Social Networking}} \bibinfo{volume}{21}, \bibinfo{number}{3} (\bibinfo{year}{2018}), \bibinfo{pages}{157--167}.
\newblock
\bibitem[KakaoTalk(2022)]%
{KakaoSocialSupportChatroom}
\bibfield{author}{\bibinfo{person}{KakaoTalk}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \bibinfo{title}{KakaoTalk Online Social Support Chatroom for People with Eating Disorders}.
\newblock \bibinfo{howpublished}{\url{https://open.kakao.com/o/guHeu70b}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Kang et~al\mbox{.}(2020)]%
{kang2020popularity}
\bibfield{author}{\bibinfo{person}{EunKyo Kang}, \bibinfo{person}{Jihye Lee}, \bibinfo{person}{Kyae~Hyung Kim}, {and} \bibinfo{person}{Young~Ho Yun}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{The popularity of eating broadcast: Content analysis of “mukbang” YouTube videos, media coverage, and the health impact of “mukbang” on public}.
\newblock \bibinfo{journal}{\emph{Health informatics journal}} \bibinfo{volume}{26}, \bibinfo{number}{3} (\bibinfo{year}{2020}), \bibinfo{pages}{2237--2248}.
\newblock
\bibitem[Kapoor and Vij(2018)]%
{kapoor2018technology}
\bibfield{author}{\bibinfo{person}{Anuj~Pal Kapoor} {and} \bibinfo{person}{Madhu Vij}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{Technology at the dinner table: Ordering food online through mobile apps}.
\newblock \bibinfo{journal}{\emph{Journal of Retailing and Consumer Services}} \bibinfo{volume}{43} (\bibinfo{year}{2018}), \bibinfo{pages}{342--351}.
\newblock
\bibitem[Khot et~al\mbox{.}(2022)]%
{khot2022understanding}
\bibfield{author}{\bibinfo{person}{Rohit~Ashok Khot}, \bibinfo{person}{Deepti Aggarwal}, {and} \bibinfo{person}{Nandini Pasumarthy}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{Understanding Screen-based Dining Practices through the Lens of Mindful Eating}. In \bibinfo{booktitle}{\emph{Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--19}.
\newblock
\bibitem[Kim et~al\mbox{.}(2021)]%
{kim2021binge}
\bibfield{author}{\bibinfo{person}{Jiye Kim}, \bibinfo{person}{Saegyeol Choi}, \bibinfo{person}{Hyekyeong Kim}, {and} \bibinfo{person}{Soontae An}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{Binge drinking and obesity-related eating: The moderating roles of the eating broadcast viewing experience among Korean adults}.
\newblock \bibinfo{journal}{\emph{International Journal of Environmental Research and Public Health}} \bibinfo{volume}{18}, \bibinfo{number}{15} (\bibinfo{year}{2021}), \bibinfo{pages}{8066}.
\newblock
\bibitem[Kinard(2016)]%
{kinard2016insta}
\bibfield{author}{\bibinfo{person}{Brian~R Kinard}.} \bibinfo{year}{2016}\natexlab{}.
\newblock \showarticletitle{Insta-Grams: The effect of consumer weight on reactions to healthy food posts}.
\newblock \bibinfo{journal}{\emph{Cyberpsychology, Behavior, and Social Networking}} \bibinfo{volume}{19}, \bibinfo{number}{8} (\bibinfo{year}{2016}), \bibinfo{pages}{481--486}.
\newblock
\bibitem[Kinkel-Ram et~al\mbox{.}(2022)]%
{kinkel2022food}
\bibfield{author}{\bibinfo{person}{Shruti~S Kinkel-Ram}, \bibinfo{person}{Cody Staples}, \bibinfo{person}{Diana Rancourt}, {and} \bibinfo{person}{April~R Smith}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{Food for thought: Examining the relationship between low calorie density foods in Instagram feeds and disordered eating symptoms among undergraduate women}.
\newblock \bibinfo{journal}{\emph{Eating Behaviors}} \bibinfo{volume}{47} (\bibinfo{year}{2022}), \bibinfo{pages}{101679}.
\newblock
\bibitem[Kircaburun et~al\mbox{.}(2021a)]%
{kircaburun2021compensatory}
\bibfield{author}{\bibinfo{person}{Kagan Kircaburun}, \bibinfo{person}{Sabah Balta}, \bibinfo{person}{Emrah Emirtekin}, \bibinfo{person}{{\c{S}}ule~Bet{\"u}l Tosuntas}, \bibinfo{person}{Zsolt Demetrovics}, {and} \bibinfo{person}{Mark~D Griffiths}.} \bibinfo{year}{2021}\natexlab{a}.
\newblock \showarticletitle{Compensatory usage of the internet: The case of mukbang watching on YouTube}.
\newblock \bibinfo{journal}{\emph{Psychiatry Investigation}} \bibinfo{volume}{18}, \bibinfo{number}{4} (\bibinfo{year}{2021}), \bibinfo{pages}{269}.
\newblock
\bibitem[Kircaburun et~al\mbox{.}(2021b)]%
{kircaburun2021psychology}
\bibfield{author}{\bibinfo{person}{Kagan Kircaburun}, \bibinfo{person}{Andrew Harris}, \bibinfo{person}{Filipa Calado}, {and} \bibinfo{person}{Mark~D Griffiths}.} \bibinfo{year}{2021}\natexlab{b}.
\newblock \showarticletitle{The psychology of mukbang watching: A scoping review of the academic and non-academic literature}.
\newblock \bibinfo{journal}{\emph{International Journal of Mental Health and Addiction}} \bibinfo{volume}{19}, \bibinfo{number}{4} (\bibinfo{year}{2021}), \bibinfo{pages}{1190--1213}.
\newblock
\bibitem[Kircaburun et~al\mbox{.}(2021c)]%
{kircaburun2021problematic}
\bibfield{author}{\bibinfo{person}{Kagan Kircaburun}, \bibinfo{person}{Cemil Yurdag{\"u}l}, \bibinfo{person}{Daria Kuss}, \bibinfo{person}{Emrah Emirtekin}, {and} \bibinfo{person}{Mark~D Griffiths}.} \bibinfo{year}{2021}\natexlab{c}.
\newblock \showarticletitle{Problematic mukbang watching and its relationship to disordered eating and internet addiction: A pilot study among emerging adult mukbang watchers}.
\newblock \bibinfo{journal}{\emph{International Journal of Mental Health and Addiction}} \bibinfo{volume}{19} (\bibinfo{year}{2021}), \bibinfo{pages}{2160--2169}.
\newblock
\bibitem[Kitchens(2021)]%
{tiktokwrapped2021}
\bibfield{author}{\bibinfo{person}{Magnet Kitchens}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \bibinfo{title}{TikTok Wrapped: The 2021 Food Trend Report}.
\newblock \bibinfo{howpublished}{\url{https://www.magnet.co.uk/the-2021-tiktok-food-trend-report/}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Kleinberger et~al\mbox{.}(2023)]%
{kleinberger2023auditory}
\bibfield{author}{\bibinfo{person}{Rebecca Kleinberger}, \bibinfo{person}{Akito~Oshiro Van~Troyer}, {and} \bibinfo{person}{Qian~Janice Wang}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \showarticletitle{Auditory Seasoning Filters: Altering Food Perception via Augmented Sonic Feedback of Chewing Sounds}. In \bibinfo{booktitle}{\emph{Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--15}.
\newblock
\bibitem[Lee et~al\mbox{.}(2017)]%
{lee2017impaired}
\bibfield{author}{\bibinfo{person}{Jung~Eun Lee}, \bibinfo{person}{Kee Namkoong}, {and} \bibinfo{person}{Young-Chul Jung}.} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{Impaired prefrontal cognitive control over interference by food images in binge-eating disorder and bulimia nervosa}.
\newblock \bibinfo{journal}{\emph{Neuroscience letters}} \bibinfo{volume}{651} (\bibinfo{year}{2017}), \bibinfo{pages}{95--101}.
\newblock
\bibitem[Legenbauer et~al\mbox{.}(2004)]%
{legenbauer2004anticipatory}
\bibfield{author}{\bibinfo{person}{Tanja Legenbauer}, \bibinfo{person}{Claus V{\"o}gele}, {and} \bibinfo{person}{Heinz R{\"u}ddel}.} \bibinfo{year}{2004}\natexlab{}.
\newblock \showarticletitle{Anticipatory effects of food exposure in women diagnosed with bulimia nervosa}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{42}, \bibinfo{number}{1} (\bibinfo{year}{2004}), \bibinfo{pages}{33--40}.
\newblock
\bibitem[Lindrup et~al\mbox{.}(2021)]%
{lindrup2021one}
\bibfield{author}{\bibinfo{person}{Martin~VA Lindrup}, \bibinfo{person}{EunJeong Cheon}, \bibinfo{person}{Mikael~B Skov}, {and} \bibinfo{person}{Dimitrios Raptis}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{One byte at a time: Insights about meaningful data for sustainable food consumption practices}. In \bibinfo{booktitle}{\emph{Proceedings of the 2021 ACM Designing Interactive Systems Conference}}. \bibinfo{pages}{683--696}.
\newblock
\bibitem[Liu et~al\mbox{.}(2013)]%
{liu2013website}
\bibfield{author}{\bibinfo{person}{Yong Liu}, \bibinfo{person}{Hongxiu Li}, {and} \bibinfo{person}{Feng Hu}.} \bibinfo{year}{2013}\natexlab{}.
\newblock \showarticletitle{Website attributes in urging online impulse purchase: An empirical investigation on consumer perceptions}.
\newblock \bibinfo{journal}{\emph{Decision support systems}} \bibinfo{volume}{55}, \bibinfo{number}{3} (\bibinfo{year}{2013}), \bibinfo{pages}{829--837}.
\newblock
\bibitem[Luo et~al\mbox{.}(2019)]%
{luo2019co}
\bibfield{author}{\bibinfo{person}{Yuhan Luo}, \bibinfo{person}{Peiyi Liu}, {and} \bibinfo{person}{Eun~Kyoung Choe}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{Co-Designing food trackers with dietitians: Identifying design opportunities for food tracker customization}. In \bibinfo{booktitle}{\emph{Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--13}.
\newblock
\bibitem[Maimaiti et~al\mbox{.}(2018)]%
{maimaiti2018we}
\bibfield{author}{\bibinfo{person}{Mayila Maimaiti}, \bibinfo{person}{Xueyin Zhao}, \bibinfo{person}{Menghan Jia}, \bibinfo{person}{Yuan Ru}, {and} \bibinfo{person}{Shankuan Zhu}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{How we eat determines what we become: opportunities and challenges brought by food delivery industry in a changing world in China}.
\newblock \bibinfo{journal}{\emph{European journal of clinical nutrition}} \bibinfo{volume}{72}, \bibinfo{number}{9} (\bibinfo{year}{2018}), \bibinfo{pages}{1282--1286}.
\newblock
\bibitem[Mejova et~al\mbox{.}(2016)]%
{mejova2016fetishizing}
\bibfield{author}{\bibinfo{person}{Yelena Mejova}, \bibinfo{person}{Sofiane Abbar}, {and} \bibinfo{person}{Hamed Haddadi}.} \bibinfo{year}{2016}\natexlab{}.
\newblock \showarticletitle{Fetishizing food in digital age:\# foodporn around the world}. In \bibinfo{booktitle}{\emph{Tenth International AAAI Conference on Web and Social Media}}.
\newblock
\bibitem[MurrayStuart et~al\mbox{.}(2018)]%
{murraystuart2018contribution}
\bibfield{author}{\bibinfo{person}{B MurrayStuart}, \bibinfo{person}{A McLeanSi{\^a}n}, {et~al\mbox{.}}} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{The contribution of social media to body dissatisfaction, eating disorder symptoms, and anabolic steroid use among sexual minority men}.
\newblock \bibinfo{journal}{\emph{Cyberpsychology, Behavior, and Social Networking}} (\bibinfo{year}{2018}).
\newblock
\bibitem[Nguyen et~al\mbox{.}(2014)]%
{nguyen2014exploring}
\bibfield{author}{\bibinfo{person}{Tien~T Nguyen}, \bibinfo{person}{Pik-Mai Hui}, \bibinfo{person}{F~Maxwell Harper}, \bibinfo{person}{Loren Terveen}, {and} \bibinfo{person}{Joseph~A Konstan}.} \bibinfo{year}{2014}\natexlab{}.
\newblock \showarticletitle{Exploring the filter bubble: the effect of using recommender systems on content diversity}. In \bibinfo{booktitle}{\emph{Proceedings of the 23rd international conference on World wide web}}. \bibinfo{pages}{677--686}.
\newblock
\bibitem[Norman(1986)]%
{norman1986attention}
\bibfield{author}{\bibinfo{person}{DA Norman}.} \bibinfo{year}{1986}\natexlab{}.
\newblock \showarticletitle{Attention to action: Willed and automatic control of behavior}.
\newblock \bibinfo{journal}{\emph{Consciousness and self-regulation: Advances in research and theory}} \bibinfo{volume}{4} (\bibinfo{year}{1986}).
\newblock
\bibitem[Ofli et~al\mbox{.}(2017)]%
{ofli2017saki}
\bibfield{author}{\bibinfo{person}{Ferda Ofli}, \bibinfo{person}{Yusuf Aytar}, \bibinfo{person}{Ingmar Weber}, \bibinfo{person}{Raggi Al~Hammouri}, {and} \bibinfo{person}{Antonio Torralba}.} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{Is saki\# delicious? the food perception gap on instagram and its relation to health}. In \bibinfo{booktitle}{\emph{Proceedings of the 26th International Conference on World Wide Web}}. \bibinfo{pages}{509--518}.
\newblock
\bibitem[Pater et~al\mbox{.}(2021)]%
{pater2021charting}
\bibfield{author}{\bibinfo{person}{Jessica Pater}, \bibinfo{person}{Fayika~Farhat Nova}, \bibinfo{person}{Amanda Coupe}, \bibinfo{person}{Lauren~E Reining}, \bibinfo{person}{Connie Kerrigan}, \bibinfo{person}{Tammy Toscos}, {and} \bibinfo{person}{Elizabeth~D Mynatt}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{Charting the Unknown: Challenges in the Clinical Assessment of Patients’ Technology Use Related to Eating Disorders}. In \bibinfo{booktitle}{\emph{Proceedings of the 2021 CHI conference on human factors in computing systems}}. \bibinfo{pages}{1--14}.
\newblock
\bibitem[Pater et~al\mbox{.}(2019)]%
{pater2019notjustgirls}
\bibfield{author}{\bibinfo{person}{Jessica~A Pater}, \bibinfo{person}{Lauren~E Reining}, \bibinfo{person}{Andrew~D Miller}, \bibinfo{person}{Tammy Toscos}, {and} \bibinfo{person}{Elizabeth~D Mynatt}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{" Notjustgirls" Exploring Male-related Eating Disordered Content across Social Media Platforms}. In \bibinfo{booktitle}{\emph{Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--13}.
\newblock
\bibitem[Pedalino and Camerini(2022)]%
{pedalino2022instagram}
\bibfield{author}{\bibinfo{person}{Federica Pedalino} {and} \bibinfo{person}{Anne-Linda Camerini}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{Instagram use and body dissatisfaction: The mediating role of upward social comparison with peers and influencers among young females}.
\newblock \bibinfo{journal}{\emph{International journal of environmental research and public health}} \bibinfo{volume}{19}, \bibinfo{number}{3} (\bibinfo{year}{2022}), \bibinfo{pages}{1543}.
\newblock
\bibitem[Peng and JEMMOTT~III(2018)]%
{peng2018feast}
\bibfield{author}{\bibinfo{person}{Yilang Peng} {and} \bibinfo{person}{JOHN~B JEMMOTT~III}.} \bibinfo{year}{2018}\natexlab{}.
\newblock \showarticletitle{Feast for the Eyes: Effects of Food Perceptions and Computer Vision Features on Food Photo Popularity.}
\newblock \bibinfo{journal}{\emph{International Journal of Communication (19328036)}} \bibinfo{volume}{12} (\bibinfo{year}{2018}).
\newblock
\bibitem[Peng-Li et~al\mbox{.}(2021)]%
{peng2021sounds}
\bibfield{author}{\bibinfo{person}{Danni Peng-Li}, \bibinfo{person}{Signe~L Mathiesen}, \bibinfo{person}{Raymond~CK Chan}, \bibinfo{person}{Derek~V Byrne}, {and} \bibinfo{person}{Qian~Janice Wang}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{Sounds Healthy: Modelling sound-evoked consumer food choice through visual attention}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{164} (\bibinfo{year}{2021}), \bibinfo{pages}{105264}.
\newblock
\bibitem[Petit et~al\mbox{.}(2016)]%
{petit2016can}
\bibfield{author}{\bibinfo{person}{Olivia Petit}, \bibinfo{person}{Adrian~D Cheok}, {and} \bibinfo{person}{Olivier Oullier}.} \bibinfo{year}{2016}\natexlab{}.
\newblock \showarticletitle{Can food porn make us slim? How brains of consumers react to food in digital environments}.
\newblock \bibinfo{journal}{\emph{Integrative Food, Nutrition and Metabolism}} \bibinfo{volume}{3}, \bibinfo{number}{1} (\bibinfo{year}{2016}), \bibinfo{pages}{251--255}.
\newblock
\bibitem[Pope et~al\mbox{.}(2015)]%
{pope2015viewers}
\bibfield{author}{\bibinfo{person}{Lizzy Pope}, \bibinfo{person}{Lara Latimer}, {and} \bibinfo{person}{Brian Wansink}.} \bibinfo{year}{2015}\natexlab{}.
\newblock \showarticletitle{Viewers vs. Doers. The relationship between watching food television and BMI}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{90} (\bibinfo{year}{2015}), \bibinfo{pages}{131--135}.
\newblock
\bibitem[Portingale et~al\mbox{.}(2023)]%
{portingale2023tonight}
\bibfield{author}{\bibinfo{person}{Jade Portingale}, \bibinfo{person}{Sarah Eddy}, \bibinfo{person}{Matthew Fuller-Tyszkiewicz}, \bibinfo{person}{Shanshan Liu}, \bibinfo{person}{Sarah Giles}, {and} \bibinfo{person}{Isabel Krug}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \showarticletitle{Tonight, I'm disordered eating: The effects of food delivery app use, loneliness, and mood on daily body dissatisfaction and disordered eating urges}.
\newblock \bibinfo{journal}{\emph{Appetite}} \bibinfo{volume}{180} (\bibinfo{year}{2023}), \bibinfo{pages}{106310}.
\newblock
\bibitem[Puhl and Suh(2015)]%
{puhl2015stigma}
\bibfield{author}{\bibinfo{person}{Rebecca Puhl} {and} \bibinfo{person}{Young Suh}.} \bibinfo{year}{2015}\natexlab{}.
\newblock \showarticletitle{Stigma and eating and weight disorders}.
\newblock \bibinfo{journal}{\emph{Current psychiatry reports}} \bibinfo{volume}{17}, \bibinfo{number}{3} (\bibinfo{year}{2015}), \bibinfo{pages}{1--10}.
\newblock
\bibitem[Qutteina et~al\mbox{.}(2019)]%
{qutteina2019media}
\bibfield{author}{\bibinfo{person}{Yara Qutteina}, \bibinfo{person}{Charlotte De~Backer}, {and} \bibinfo{person}{Tim Smits}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{Media food marketing and eating outcomes among pre-adolescents and adolescents: A systematic review and meta-analysis}.
\newblock \bibinfo{journal}{\emph{Obesity Reviews}} \bibinfo{volume}{20}, \bibinfo{number}{12} (\bibinfo{year}{2019}), \bibinfo{pages}{1708--1719}.
\newblock
\bibitem[Schienle et~al\mbox{.}(2009)]%
{schienle2009binge}
\bibfield{author}{\bibinfo{person}{Anne Schienle}, \bibinfo{person}{Axel Sch{\"a}fer}, \bibinfo{person}{Andrea Hermann}, {and} \bibinfo{person}{Dieter Vaitl}.} \bibinfo{year}{2009}\natexlab{}.
\newblock \showarticletitle{Binge-eating disorder: reward sensitivity and brain activation to images of food}.
\newblock \bibinfo{journal}{\emph{Biological psychiatry}} \bibinfo{volume}{65}, \bibinfo{number}{8} (\bibinfo{year}{2009}), \bibinfo{pages}{654--661}.
\newblock
\bibitem[Schneider and Schwarz(2017)]%
{schneider2017mixed}
\bibfield{author}{\bibinfo{person}{Iris~K Schneider} {and} \bibinfo{person}{Norbert Schwarz}.} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{Mixed feelings: The case of ambivalence}.
\newblock \bibinfo{journal}{\emph{Current Opinion in Behavioral Sciences}} \bibinfo{volume}{15} (\bibinfo{year}{2017}), \bibinfo{pages}{39--45}.
\newblock
\bibitem[Seal et~al\mbox{.}(2022)]%
{seal2022can}
\bibfield{author}{\bibinfo{person}{Ananya Seal}, \bibinfo{person}{SubbaRao~M Gavaravarapu}, {and} \bibinfo{person}{Archana Konapur}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{Can foodporn prime healthy eating? Thinking beyond digital gazing and satiety}.
\newblock \bibinfo{journal}{\emph{European Journal of Clinical Nutrition}} (\bibinfo{year}{2022}), \bibinfo{pages}{1--4}.
\newblock
\bibitem[Service(2023)]%
{KoreaBEDProportion}
\bibfield{author}{\bibinfo{person}{Health Insurance Review \&~Assessment Service}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \bibinfo{title}{100 Disease Statistics in South Korea}.
\newblock \bibinfo{howpublished}{\url{https://www.hira.or.kr/ebooksc/ebook_472/ebook_472_201803281057049800.pdf}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Smyth(1998)]%
{smyth1998written}
\bibfield{author}{\bibinfo{person}{Joshua~M Smyth}.} \bibinfo{year}{1998}\natexlab{}.
\newblock \showarticletitle{Written emotional expression: effect sizes, outcome types, and moderating variables.}
\newblock \bibinfo{journal}{\emph{Journal of consulting and clinical psychology}} \bibinfo{volume}{66}, \bibinfo{number}{1} (\bibinfo{year}{1998}), \bibinfo{pages}{174}.
\newblock
\bibitem[Spence et~al\mbox{.}(2019)]%
{spence2019digital}
\bibfield{author}{\bibinfo{person}{Charles Spence}, \bibinfo{person}{Maurizio Mancini}, {and} \bibinfo{person}{Gijs Huisman}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{Digital commensality: Eating and drinking in the company of technology}.
\newblock \bibinfo{journal}{\emph{Frontiers in psychology}} \bibinfo{volume}{10} (\bibinfo{year}{2019}), \bibinfo{pages}{2252}.
\newblock
\bibitem[Spence et~al\mbox{.}(2016)]%
{spence2016eating}
\bibfield{author}{\bibinfo{person}{Charles Spence}, \bibinfo{person}{Katsunori Okajima}, \bibinfo{person}{Adrian~David Cheok}, \bibinfo{person}{Olivia Petit}, {and} \bibinfo{person}{Charles Michel}.} \bibinfo{year}{2016}\natexlab{}.
\newblock \showarticletitle{Eating with our eyes: From visual hunger to digital satiation}.
\newblock \bibinfo{journal}{\emph{Brain and cognition}} \bibinfo{volume}{110} (\bibinfo{year}{2016}), \bibinfo{pages}{53--63}.
\newblock
\bibitem[Stacey et~al\mbox{.}(2017)]%
{stacey2017decision}
\bibfield{author}{\bibinfo{person}{Dawn Stacey}, \bibinfo{person}{France L{\'e}gar{\'e}}, \bibinfo{person}{Krystina Lewis}, \bibinfo{person}{Michael~J Barry}, \bibinfo{person}{Carol~L Bennett}, \bibinfo{person}{Karen~B Eden}, \bibinfo{person}{Margaret Holmes-Rovner}, \bibinfo{person}{Hilary Llewellyn-Thomas}, \bibinfo{person}{Anne Lyddiatt}, \bibinfo{person}{Richard Thomson}, {et~al\mbox{.}}} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{Decision aids for people facing health treatment or screening decisions}.
\newblock \bibinfo{journal}{\emph{Cochrane database of systematic reviews}} \bibinfo{number}{4} (\bibinfo{year}{2017}).
\newblock
\bibitem[Stephens et~al\mbox{.}(2020)]%
{stephens2020food}
\bibfield{author}{\bibinfo{person}{Janna Stephens}, \bibinfo{person}{Hailey Miller}, {and} \bibinfo{person}{Lisa Militello}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{Food delivery apps and the negative health impacts for Americans}.
\newblock \bibinfo{journal}{\emph{Frontiers in Nutrition}} \bibinfo{volume}{7} (\bibinfo{year}{2020}), \bibinfo{pages}{14}.
\newblock
\bibitem[Strand and Gustafsson(2020)]%
{strand2020mukbang}
\bibfield{author}{\bibinfo{person}{Mattias Strand} {and} \bibinfo{person}{Sanna~Aila Gustafsson}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{Mukbang and disordered eating: a netnographic analysis of online eating broadcasts}.
\newblock \bibinfo{journal}{\emph{Culture, Medicine, and Psychiatry}} \bibinfo{volume}{44}, \bibinfo{number}{4} (\bibinfo{year}{2020}), \bibinfo{pages}{586--609}.
\newblock
\bibitem[Thompson et~al\mbox{.}(2014)]%
{thompson2014let}
\bibfield{author}{\bibinfo{person}{Megan~M Thompson}, \bibinfo{person}{Mark~P Zanna}, {and} \bibinfo{person}{Dale~W Griffin}.} \bibinfo{year}{2014}\natexlab{}.
\newblock \showarticletitle{Let's not be indifferent about (attitudinal) ambivalence}.
\newblock In \bibinfo{booktitle}{\emph{Attitude strength}}. \bibinfo{publisher}{Psychology Press}, \bibinfo{pages}{361--386}.
\newblock
\bibitem[TikTok(2025)]%
{tiktokfood}
\bibfield{author}{\bibinfo{person}{TikTok}.} \bibinfo{year}{2025}\natexlab{}.
\newblock \bibinfo{title}{\#tiktokfood}.
\newblock \bibinfo{howpublished}{\url{https://ads.tiktok.com/business/creativecenter/hashtag/tiktokfood/pc/en?countryCode=SA&period=7}}.
\newblock
\newblock
\shownote{Accessed: \today}.
\bibitem[Van~Harreveld et~al\mbox{.}(2015)]%
{van2015abc}
\bibfield{author}{\bibinfo{person}{Frenk Van~Harreveld}, \bibinfo{person}{Hannah~U Nohlen}, {and} \bibinfo{person}{Iris~K Schneider}.} \bibinfo{year}{2015}\natexlab{}.
\newblock \showarticletitle{The ABC of ambivalence: Affective, behavioral, and cognitive consequences of attitudinal conflict}.
\newblock In \bibinfo{booktitle}{\emph{Advances in experimental social psychology}}. Vol.~\bibinfo{volume}{52}. \bibinfo{publisher}{Elsevier}, \bibinfo{pages}{285--324}.
\newblock
\bibitem[Van~Harreveld et~al\mbox{.}(2009)]%
{van2009agony}
\bibfield{author}{\bibinfo{person}{Frenk Van~Harreveld}, \bibinfo{person}{Joop Van~der Pligt}, {and} \bibinfo{person}{Yael~N De~Liver}.} \bibinfo{year}{2009}\natexlab{}.
\newblock \showarticletitle{The agony of ambivalence and ways to resolve it: Introducing the MAID model}.
\newblock \bibinfo{journal}{\emph{Personality and Social Psychology Review}} \bibinfo{volume}{13}, \bibinfo{number}{1} (\bibinfo{year}{2009}), \bibinfo{pages}{45--61}.
\newblock
\bibitem[Velasco et~al\mbox{.}(2016)]%
{velasco2016multisensory}
\bibfield{author}{\bibinfo{person}{Carlos Velasco}, \bibinfo{person}{Felipe~Reinoso Carvalho}, \bibinfo{person}{Olivia Petit}, {and} \bibinfo{person}{Anton Nijholt}.} \bibinfo{year}{2016}\natexlab{}.
\newblock \showarticletitle{A multisensory approach for the design of food and drink enhancing sonic systems}. In \bibinfo{booktitle}{\emph{Proceedings of the 1st workshop on multi-sensorial approaches to human-food interaction}}. \bibinfo{pages}{1--7}.
\newblock
\bibitem[Ventura et~al\mbox{.}(2021)]%
{ventura2021socialfood}
\bibfield{author}{\bibinfo{person}{Vera Ventura}, \bibinfo{person}{Alessia Cavaliere}, {and} \bibinfo{person}{Beatrice Ianno}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{\# Socialfood: Virtuous or vicious? A systematic review}.
\newblock \bibinfo{journal}{\emph{Trends in Food Science \& Technology}} \bibinfo{volume}{110} (\bibinfo{year}{2021}), \bibinfo{pages}{674--686}.
\newblock
\bibitem[Vermeir and Roose(2020)]%
{vermeir2020visual}
\bibfield{author}{\bibinfo{person}{Iris Vermeir} {and} \bibinfo{person}{Gudrun Roose}.} \bibinfo{year}{2020}\natexlab{}.
\newblock \showarticletitle{Visual design cues impacting food choice: A review and future research agenda}.
\newblock \bibinfo{journal}{\emph{Foods}} \bibinfo{volume}{9}, \bibinfo{number}{10} (\bibinfo{year}{2020}), \bibinfo{pages}{1495}.
\newblock
\bibitem[von Ash et~al\mbox{.}(2023)]%
{von2023associations}
\bibfield{author}{\bibinfo{person}{Tayla von Ash}, \bibinfo{person}{Rachel Huynh}, \bibinfo{person}{Carrie Deng}, {and} \bibinfo{person}{Marney~A White}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \showarticletitle{Associations between mukbang viewing and disordered eating behaviors}.
\newblock \bibinfo{journal}{\emph{International Journal of Eating Disorders}} (\bibinfo{year}{2023}).
\newblock
\bibitem[Wang et~al\mbox{.}(2022)]%
{wang2022tiktok}
\bibfield{author}{\bibinfo{person}{Chun-Han Wang}, \bibinfo{person}{Stephen Tsung-Han Sher}, \bibinfo{person}{Isabela Salman}, \bibinfo{person}{Kelly Janek}, {and} \bibinfo{person}{Chia-Fang Chung}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{“TikTok Made Me Do It”: Teenagers’ Perception and Use of Food Content on TikTok}. In \bibinfo{booktitle}{\emph{Interaction Design and Children}}. \bibinfo{pages}{458--463}.
\newblock
\bibitem[Wang et~al\mbox{.}(2024)]%
{wang2024viral}
\bibfield{author}{\bibinfo{person}{Chun-Han~Ariel Wang}, \bibinfo{person}{Stephen Tsung-Han Sher}, {and} \bibinfo{person}{Chia-Fang Chung}.} \bibinfo{year}{2024}\natexlab{}.
\newblock \showarticletitle{From Viral Content to Real-Life Cuisine and Beyond: Examining Teenagers' Interactions with TikTok Food Videos and the Influence on their Food Practices}.
\newblock \bibinfo{journal}{\emph{Proceedings of the ACM on Human-Computer Interaction}} \bibinfo{volume}{8}, \bibinfo{number}{CSCW2} (\bibinfo{year}{2024}), \bibinfo{pages}{1--30}.
\newblock
\bibitem[Weber et~al\mbox{.}(2021)]%
{weber2021s}
\bibfield{author}{\bibinfo{person}{Philip Weber}, \bibinfo{person}{Thomas Ludwig}, \bibinfo{person}{Sabrina Brodesser}, {and} \bibinfo{person}{Laura Gr{\"o}newald}.} \bibinfo{year}{2021}\natexlab{}.
\newblock \showarticletitle{“It's a Kind of Art!”: Understanding Food Influencers as Influential Content Creators}. In \bibinfo{booktitle}{\emph{Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--14}.
\newblock
\bibitem[Wei and Yan(2023)]%
{wei2023measuring}
\bibfield{author}{\bibinfo{person}{Shuyi Wei} {and} \bibinfo{person}{Pu Yan}.} \bibinfo{year}{2023}\natexlab{}.
\newblock \showarticletitle{Measuring users’ awareness of content recommendation algorithm: A survey on Douyin users in rural China}. In \bibinfo{booktitle}{\emph{International Conference on Information}}. Springer, \bibinfo{pages}{197--220}.
\newblock
\bibitem[Wu et~al\mbox{.}(2024)]%
{wu2024digging}
\bibfield{author}{\bibinfo{person}{Yu Wu}, \bibinfo{person}{Eva Kemps}, {and} \bibinfo{person}{Ivanka Prichard}.} \bibinfo{year}{2024}\natexlab{}.
\newblock \showarticletitle{Digging into digital buffets: A systematic review of eating-related social media content and its relationship with body image and eating behaviours}.
\newblock \bibinfo{journal}{\emph{Body Image}} \bibinfo{volume}{48} (\bibinfo{year}{2024}), \bibinfo{pages}{101650}.
\newblock
\bibitem[Xu et~al\mbox{.}(2022)]%
{xu2022typeout}
\bibfield{author}{\bibinfo{person}{Xuhai Xu}, \bibinfo{person}{Tianyuan Zou}, \bibinfo{person}{Han Xiao}, \bibinfo{person}{Yanzhang Li}, \bibinfo{person}{Ruolin Wang}, \bibinfo{person}{Tianyi Yuan}, \bibinfo{person}{Yuntao Wang}, \bibinfo{person}{Yuanchun Shi}, \bibinfo{person}{Jennifer Mankoff}, {and} \bibinfo{person}{Anind~K Dey}.} \bibinfo{year}{2022}\natexlab{}.
\newblock \showarticletitle{TypeOut: Leveraging Just-in-Time Self-Affirmation for Smartphone Overuse Reduction}. In \bibinfo{booktitle}{\emph{CHI Conference on Human Factors in Computing Systems}}. \bibinfo{pages}{1--17}.
\newblock
\bibitem[Yan and Corino(2019)]%
{yan2019digital}
\bibfield{author}{\bibinfo{person}{Jie Yan} {and} \bibinfo{person}{Gianni Corino}.} \bibinfo{year}{2019}\natexlab{}.
\newblock \showarticletitle{Digital foraging: how social media is changing our relationship with food and nutrition?}
\newblock \bibinfo{journal}{\emph{DigitCult-Scientific Journal on Digital Cultures}} \bibinfo{volume}{4}, \bibinfo{number}{3} (\bibinfo{year}{2019}), \bibinfo{pages}{39--46}.
\newblock
\bibitem[Yeo et~al\mbox{.}(2017)]%
{yeo2017consumer}
\bibfield{author}{\bibinfo{person}{Vincent Cheow~Sern Yeo}, \bibinfo{person}{See-Kwong Goh}, {and} \bibinfo{person}{Sajad Rezaei}.} \bibinfo{year}{2017}\natexlab{}.
\newblock \showarticletitle{Consumer experiences, attitude and behavioral intention toward online food delivery (OFD) services}.
\newblock \bibinfo{journal}{\emph{Journal of Retailing and Consumer services}} \bibinfo{volume}{35} (\bibinfo{year}{2017}), \bibinfo{pages}{150--162}.
\newblock
\bibitem[Zepeda and Deal(2008)]%
{zepeda2008think}
\bibfield{author}{\bibinfo{person}{Lydia Zepeda} {and} \bibinfo{person}{David Deal}.} \bibinfo{year}{2008}\natexlab{}.
\newblock \showarticletitle{Think before you eat: photographic food diaries as intervention tools to change dietary decision making and attitudes}.
\newblock \bibinfo{journal}{\emph{International Journal of Consumer Studies}} \bibinfo{volume}{32}, \bibinfo{number}{6} (\bibinfo{year}{2008}), \bibinfo{pages}{692--698}.
\newblock
\end{thebibliography}