Can eating pleasure be a lever for healthy eating? A systematic scoping review of eating pleasure and its links with dietary behaviors and health

The aims of this review were to map and summarize data currently available about 1) key dimensions of eating pleasure; 2) associations of eating pleasure, and its key dimensions, with dietary and health outcomes and 3) the most promising intervention strategies using eating pleasure to promote healthy eating. Using the scoping review methodology, a comprehensive search of the peer-reviewed literature (Medline, PsycInfo, Embase, ERIC, Web of Science, CINAHL, ABI/Inform global and Sociology Abstract) and of the grey literature (ProQuest Dissertations & Theses and Google) was carried out by two independent reviewers. We included 119 of the 28,908 studies found. In total, 89 sub-dimensions of eating pleasure were grouped into 22 key dimensions. The most frequently found related to sensory experiences (in 50.9% of the documents), social experiences (42.7%), food characteristics besides sensory attributes (27.3%), food preparation process (19.1%), novelty (16.4%), variety (14.5%), mindful eating (13.6%), visceral eating (12.7%), place where food is consumed (11.8%) and memories associated with eating (10.9%). Forty-five studies, mostly cross-sectional (62.2%), have documented links between eating pleasure and dietary and/or health outcomes. Most studies (57.1%) reported favorable associations between eating pleasure and dietary outcomes. For health outcomes, results were less consistent. The links between eating pleasure and both dietary and health outcomes varied according to the dimensions of eating pleasure studied. Finally, results from 11 independent interventions suggested that strategies focusing on sensory experiences, cooking and/or sharing activities, mindful eating, and positive memories related to healthy food may be most promising. Thus, eating pleasure may be an ally in the promotion of healthy eating. However, systematically developed, evidence-based interventions are needed to better understand how eating pleasure may be a lever for healthy eating.

Introduction Diet quality has been identified as a major determinant of health. However, in developed countries, many individuals still not consume enough fruits, vegetables and wholes grains, while consuming too much processed foods high in energy, sodium, free sugars and saturated fats [1,2]. Despite numerous public health campaigns aimed at improving dietary habits, recent data suggest that diet quality has remained suboptimal over years [3]. These observations are challenging for policymakers and health professionals, underlying the necessity to improve food policies and interventions, and to identify new approaches to promote healthy eating that may have a greater impact on diet quality over time. These efforts could lead to considerable health benefits, along with reducing disease-related costs [4].
Some authors have proposed that the strong focus on the healthiness of food that is typically used to encourage individuals to change their food habits could explain, at least partly, the limited success of strategies used to improve diet quality [5][6][7][8]. Indeed, it is likely that the uneven importance attributed to the health value of food compared with other food values (e.g., related to preferences and human experiences) can nurture the beliefs that the healthiness of food is incompatible with these other food attributes [9]. A good example of this is the well-documented unhealthy equal tasty association. Some studies have shown that food presented as being healthy is often considered as being less tasty than "unhealthy" food [10][11][12][13][14][15]. This perception seems to be influenced by the culture. Previous research observed that, for Americans, unhealthy food is implicitly associated to tastiness [15]. However, the opposite intuition has been observed in France, French people spontaneously associating unhealthy food with bad taste, while linking healthy food to tastiness [16]. Considering that taste is a predominant determinant of food choices [17,18], the unhealthy equal tasty intuition can clearly thwart the adoption of healthy food choices.
A growing number of authors have suggested that pleasure should be emphasized more in the promotion of healthy eating [6][7][8][19][20][21]. This is even reflected in the latest dietary guidelines in some countries, including Canada [22] and Brazil [23], but also in intervention strategies using the "intuitive eating" approach (i.e., an approach that implies eating in response to hunger and satiety cues and rediscovering the pleasures of eating) [24]. This new paradigm is supported by a growing body of research that has reported, using different study designs, favorable associations between eating pleasure and dietary and health outcomes. For example, various cross-sectional studies have linked higher eating pleasure to better nutritional status [25], healthier food choices [26][27][28], increased subjective diet-related quality of life [29] and decreased depressive symptoms [30]. Other studies of both short-term and long-term interventions using eating pleasure also observed promising results. For example, Petit et al. [7] found that focusing on the healthiness of food decreased activity in the gustatory and reward brain areas, whereas focusing on the tastiness of food increased brain activity in self-control, gustatory and reward areas, and most importantly increased the percentage of healthy food choices. Overall, the authors concluded that promoting the tastiness of healthy food may lead to more successful self-control and healthy dietary behaviors. In a randomized controlled trial conducted by our research group [31], a sensory-based intervention addressing the topics of relationship with food, reliance on hunger and satiety cues, food tasting and eating pleasure showed that women who were allocated to the intervention decreased their level of disinhibition in comparison to the control group. Sasson et al. [32] did an experiment with US students who explored food and nutrition from historical, cultural and culinary perspectives during a study abroad program in Italy. Six months after the end of the program, it was found that students had increased their use of fresh and locally grown foods, improved their cooking skills and were less worried about calories [32]. Authors suggested that focusing on eating pleasure can nurture lifelong healthy eating habits. However, other studies have observed a negative link between eating pleasure and diet quality [27,33,34] or body-mass index (BMI) [35,36]. In some cross-sectional studies, eating pleasure has been associated with the consumption of convenience food [33], late-night snaking [34], and has been identified as a factor motivating the consumption of sweets [13]. Moreover, results from Cornil & Chandon [37] showed that focusing on sensory pleasure made sated eaters and dieters choose larger portions of hedonic foods, while the opposite was observed for hungry eaters and nondieters. Eating pleasure has also been reported as one of the most important reason for weight gain over years in a sample of women and men living with obesity [35].
The use of different approaches to define eating pleasure may partly explain the inconsistency of these outcomes. Studies aimed at understanding overeating and self-regulation failures [21,[38][39][40][41] mostly describe proper weight management as incompatible with eating pleasure, which they define as visceral impulses triggered by external cues or emotional urges driven by biological preferences for high-sugar, high-fat and high-sodium food. This way of defining eating pleasure is associated with overeating, loss of control and poor diet quality [21,39]. However, other approaches to define eating pleasure have been proposed in the literature [21,[42][43][44]. For example, Cornil & Chandon [21] contrast short-term visceral pleasure with what they call epicurean eating pleasure, or "the enduring pleasure derived from the aesthetic appreciation of the sensory and symbolic value of food". This definition includes not only food characteristics (i.e., taste and appearance) but also other dimensions of the eating experience such as social eating and cooking. They showed that, unlike visceral pleasure, those scoring high on epicurean pleasure preferred smaller portions and displayed a higher level of well-being [21]. Interestingly, epicurean eating pleasure displays similarities with the dimensions of eating pleasure identified in some qualitative studies [42][43][44]. These studies underline the multidimensional nature of eating pleasure, suggesting it includes dimensions related to food characteristics, the individual, the environment and the social context. The wide variety of definitions of eating pleasure could thus lead to inconsistencies in assessments of the relationship between eating pleasure and healthy eating. Establishing a common understanding of eating pleasure therefore requires identifying its key dimensions and using these to analyze in more detail how eating pleasure is associated with dietary behavior and health outcomes.
Considering the increased interest within the scientific and clinical communities in the integration of eating pleasure in the promotion of healthy eating, a comprehensive mapping of current knowledge about this topic is essential to inform health professionals and policymakers whether and how the pleasure of eating can be a lever for healthy eating. This will also help to identify gaps that will guide next research efforts in this area. In this comprehensive scoping review we therefore aimed to (1) identify the key dimensions of eating pleasure; (2) map the data currently available about associations of eating pleasure, and its key dimensions, with dietary behaviors and health outcomes and (3) identify most promising intervention strategies using eating pleasure to promote healthy eating. Scoping review design represents a rigorous methodology that allows comprehensive assessment of emerging evidence, and a first step in research development [45]. This technique aims to map the key concepts, types of evidence available, and gaps in research related to a defined field by systematically searching, selecting, and synthesizing existing knowledge [46,47]. This is particularly relevant to disciplines with emerging evidence, since researchers can incorporate a range of study designs in both published and grey literature, and address questions beyond those related to intervention effectiveness [48]. Accordingly, a scoping review was deemed more appropriate than a systematic review or a meta-analysis in the present context because (1) the objectives address a broad, multidisciplinary topic, (2) this topic has been explored using a range of research designs and (3) this complex and emerging topic has not been reviewed comprehensively before, thereby preventing the identification of more specific objectives that can be the focus of a systematic review or a meta-analysis [49].

Methods
Based on the Arksey and O'Malley methodological framework for scoping reviews [47] and enhanced according to the recommendations of Levac et al. aimed at clarifying each stage of the framework [48], six steps were performed: (1) identifying the research question, (2) identifying relevant studies, (3) selecting eligible studies, (4) charting the data, (5) collating, summarizing, and reporting results and (6) consultation. The study protocol has been registered on Open Science Framework (https://osf.io/5p7b8/) [50].
Step 1: Identifying the research questions and eligibility criteria Research questions. The purpose of this review was to assess whether and how eating pleasure can be a lever for healthy dietary behaviors and health. Accordingly, three research questions were addressed: Q1: How is eating pleasure conceptualized (i.e., key dimensions) in scientific research and organization/government documents?
Q2: What is the current available scientific evidence about associations of eating pleasure, and its key dimensions, with dietary behaviors and health outcomes?
Q3: What are the most promising strategies using eating pleasure to promote healthy dietary behaviors identified through intervention studies?
In parallel, this review sought to identify gaps that research efforts should address to further explore the place of eating pleasure in the promotion of healthy eating, and to inform health professionals and policymakers about how to use eating pleasure in their practical strategies.
Our scoping review was conducted between February 2018 and August 2019. It was conceptualized (SL and AB) and reviewed by coauthors with expertise in scoping review methodology (VP and SD).
Population and context. The population targeted was individuals of at least five years old, with no specific condition that would interfere with eating pleasure or alter the perception of eating pleasure. We decided not to include individuals under the age of five because we were interested in eating pleasure as experienced or reported by the individuals themselves, and children under five may have some trouble reporting explicit, conscious judgements due to their limited cognitive maturity, and especially language development [51]. As a result, eating pleasure is usually measured in infants and young children using observer-based measures or parent-rated measures [51,52]. Intervention strategies may also be very different at these ages, focusing more on play-based approaches and often related to neophobia and food familiarity [53][54][55][56][57][58]. In addition, as we wanted to identify strategies that could be applied in interventions aiming to prevent nutrition-related chronic diseases (e.g., obesity, type 2 diabetes, cardiovascular diseases), we excluded documents targeting people with conditions requiring a strict diet or increased dietary intake. Finally, since issues related to healthy eating differ between underdeveloped and developed countries [59], only studies conducted in developed countries (i.e., countries with high and very high human development index according to the Human Development Report 2016 [60]) were considered for inclusion.
Concepts. No a priori definition for eating pleasure was used since we aimed to conceptualize eating pleasure by identifying its key dimensions in this review. Scientific research referred to peer-reviewed literature and theses/dissertations. Given that this is an emerging area in the scientific literature a broad range of articles and study designs were considered, including but not limited to qualitative, cross-sectional, retrospective, prospective, intervention and mixed methods studies. Organization/government documents referred to guidelines, health-promotion tools, reports, and program descriptions published by a health-related organization (i.e., an organization whose ultimate mission is to promote and/or foster the health of individuals/populations) or government. Dietary behavior was defined as any actions associated with the act of eating. Health was defined according to the definition of the World Health Organization ("a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity" [61]). A promising strategy was defined as an intervention method that led to a significant and beneficial impact on dietary behavior and/or health outcomes.
Inclusion and exclusion criteria. In addition to responding to our research questions, documents considered for inclusion were: 1) mainly focused on eating pleasure dimensions or reporting data about the links between eating pleasure and dietary behavior and/or health outcomes; 2) targeting individuals of at least five years old; 3) written in English or in French (languages spoken by the research team); 4) conducted in developed countries and 5) interested in eating pleasure as experienced or reported by the individuals themselves. Excluded sources were those that 1) reported results in animals, 2) targeted individuals with a specific condition that would require a strict diet or an increased dietary intake (e.g., type 1 diabetes, malnutrition, decreased appetite, neophobia), 3) targeted people with a condition that may interfere with eating pleasure or alter the perception of eating pleasure (e.g., depression, eating disorder pathologies, pain), 4) reported research objectives aiming to evaluate eating pleasure in relation to a specific food/food group (except for vegetables and fruits, as they are considered as the cornerstone of a healthy diet [62]) and 5) described eating pleasure only in terms of the biological reward mechanisms (e.g., with brain imaging). Conference abstracts, proceedings of symposiums or conferences, editorials, expert opinions, blogs, books, book chapters, book reviews and literature reviews were also excluded.

Step 2: Identifying relevant studies
With the collaboration of a librarian specializing in nutrition, we developed a comprehensive search strategy using both specific keywords and controlled vocabulary (i.e., standardized terms used by database indexers to describe and categorize articles based on content) related to eating and pleasure. Using May 6, 2018, as a cut-off date, we searched eight electronic databases of peer-reviewed literature: Medline, PsycInfo, Embase, ERIC, Web of Science, CINAHL, ABI/Inform global and Sociology Abstract. This range of databases allowed us to capture a comprehensive sample from the many disciplines related to eating pleasure, namely medical sciences, nutrition, psychology, business, public health, education and sociology. Search terms were adapted for each database and combined using Boolean operators to narrow the results. No date restrictions were implemented in the search. As an example, the Medline search strategy is presented in Table 1. Strategies for other databases are available in S1 Table. The references of included studies were also scanned to identify other relevant results. To conceptualize eating pleasure in a more comprehensive way and to retrieve unpublished studies, we conducted additional searches in the grey literature using ProQuest Dissertations & Theses Global (cut-off date: July 10, 2019) and the search engine "Google" (between July 15, 2019 and August 26, 2019). For Google, simplified search strategies in English and French (when relevant) were developed and validated by the specialized librarian for organization websites (site:. org) and for the government website of each country for which a relevant peer-reviewed article had been identified in peer-reviewed databases (total of 16 government websites; see S1 Table  for search strategies). For Google search strategies, the government websites of the United States, France and Canada as well as organization websites were initially searched. For each of these, the first 100 hits were analyzed. Since the Google engine displays results by relative importance using a link analysis algorithm [63] and all relevant websites were in the first Google hits for these first Google searches, for the remaining governments, we decided that if at least one website was identified as relevant in the first group of 20, then the next 20 websites were searched, for up to a maximum of 100 per government search strategy. Finally, suggestions from authors of this scoping review were gathered to identify additional potentially relevant documents from the literature.

Step 3: Selecting eligible studies
All retrieved records were imported into EndNote software (Version X9), and duplicate records were deleted. First, two reviewers (POL and LMG) and the research assistant (AB) tested the selection process with 250 articles to ensure a shared understanding of inclusion and exclusion criteria. Thereafter, the two reviewers independently determined the eligibility of all articles using a two-stage screening process consisting of a title and abstract scan followed by a full-text review. If we were unable to obtain the full text of an article, the research assistant contacted the corresponding author for the article in question. For the Google searches, two reviewers (POL and CTG) determined eligibility after reading each web page in detail. Any discrepancies between the two reviewers were resolved through weekly discussion with the research assistant and, if needed, the principal investigator (SL). '/ ' indicates that a term is a subject heading (i.e., controlled vocabulary). Specific keywords are indicated in lines 2 and 5. ' � ' represents truncation. For example, "pleasur � " finds terms that begin with the root term "pleasur", such as pleasure, pleasures, pleasurable, etc. '.tw.' is an alias for all of the fields in a database which contain text words and which are appropriate for a subject search. https://doi.org/10.1371/journal.pone.0244292.t001 Step 4: Charting the data A standardized data extraction spreadsheet (Microsoft Excel) was developed by the authors. The following information was extracted: document details, design and methodology, relevant results and authors' conclusions. Subdomains are outlined in Table 2. A similar but simplified template was used for web pages. The extraction of data was undertaken by two independent coders (POL and CTG).
Prior to data extraction, this form was pilot tested by the two coders with ten selected records that were sufficiently different in terms of design and objectives to make sure that the template would be appropriate for different types of publications. The form was revised as needed during the data extraction to extract relevant information as comprehensively as possible. Each of the two coders extracted data for half of the records. The other coder then revised the extracted data and any discrepancies between the two coders were resolved by discussion and consensus. If consensus was not reached, the research assistant was consulted to resolve the disagreement.
Step 5: Collating, summarizing and reporting results As recommended by Levac et al. [48], this step was broken into three distinct tasks: (1) analyzing data (including descriptive numerical summary analysis); (2) reporting results and outcomes relevant to the research questions and (3) considering the meaning and implications of the findings for future research, practice and policy. First, a summary of each included document is presented, including the following information: author(s), year of publication, country, publication type, study/document design, journal discipline, first author's discipline, target population, word(s) used to designate "pleasure", and objective(s) targeted by the document. For each research question, a quantitative analysis for each relevant characteristic is presented. Second, data outcomes were analyzed and reported according to research questions. For the first research question (Q1: i.e., key dimensions of eating pleasure), documents analyzed were mainly those focusing on the identification of the dimensions of eating pleasure as well as those reporting data about the links between eating pleasure and dietary behavior and/or health outcomes. Each document was uploaded into qualitative NVivo Software (version 10, QSR International, Burlington, MA) and a thematic analysis of the dimensions of eating pleasure was undertaken. A basic tree node structure was built independently by the two coders and the research assistant based on 10 key articles reporting a detailed approach of eating pleasure [8,[42][43][44][64][65][66][67][68][69]. These basic tree nodes were then pooled together, and each dimension and sub-dimension of eating pleasure was described based on existing definitions and the literature. The two coders then reviewed all documents independently and extracted all dimensions of eating pleasure. New dimensions and sub-dimensions were created as needed during the extraction process. For peer-reviewed articles and theses, only the methodology and results sections were coded in order not to duplicate findings. The two coders met regularly to compare coding and to discuss all new added nodes and those for which agreement was not perfect (kappa coefficient under 100%). Any discrepancies were discussed and resolved with the research assistant. The tree node structure was discussed with the principal investigator at the beginning, middle and end of the extraction process. For the second research question (Q2: i.e., links between eating pleasure and dietary behaviors and health outcomes), the reported outcome was categorized in line with study authors' judgement as favorable (i.e., eating pleasure was associated with favorable outcomes), unfavorable (i.e., eating pleasure was associated with unfavorable outcomes), mixed (i.e., eating pleasure was associated with both favorable and unfavorable outcomes) or neutral (i.e., eating pleasure was not significantly associated with any outcomes). Outcomes most frequently studied in association with eating pleasure (i.e., diet quality, food choices, portion size, restrained eating, weight/BMI and depressive symptoms) were analyzed separately. Finally, associations between eating pleasure and outcomes were analyzed according to the ways to conceptualize (i.e., dimensions) eating pleasure as identified by coders. For the third research question (Q3: i.e., most promising intervention strategies using eating pleasure to promote healthy eating), we included studies evaluating the effect of an intervention using eating pleasure and also studies that showed any increase in eating pleasure in response to their intervention. For these studies, we made sure to give enough information about the nature of intervention groups, data analysis and measurement of outcomes in the text and tables in order to fully understand the context in which these studies were carried out. Finally, a table that groups together the dimensions based on their likely favorable or unfavorable impact and ordering the dimensions based on level of evidence was used to summarize the findings. Melnyk & Fineout-Overholt's guidelines were used to assess the level of evidence for each dimension [70]. In this model, evidence is assigned a score from 1 to 7, with 1 being the highest level of evidence. The meaning and implications of the findings for research, practice, and policy, are addressed in the Discussion, along with gaps identified through this review.
Step 6: Consulting with an expert committee An expert committee was set up at the beginning of the study, including the authors of this study and two research assistants with expertise in scoping review methodology. This expert committee was consulted for their feedback and deeper understanding of the data at each critical step of our scoping review. In addition, to help to put our findings on the dimensions of eating pleasure into perspective and to interpret them in the context of our population, results were compared to those from focus groups addressing perceptions about eating pleasure in the French-Canadian population conducted earlier by our research team [42].

Results
From the search, 27,974 records were found from the databases and 934 records identified through other sources (hand-searching references, government and organization websites, and expert committee) for a total of 28,908 records imported into Endnote (Fig 1). After exclusion of duplicates, 18,002 records were screened, from which 16,606 were excluded on the basis of Titles/Abstracts. Therefore, 1,396 were reviewed in full and 119 met our eligibility criteria and were included in this review. S2 Table provides
e Program descriptions included documents detailing objectives and/or strategies of a health-based program in which eating pleasure was integrated. f For articles, theses and dissertations, words identified as designating pleasure were those found in the title/abstract. For web pages, words identified were those found in the entire document. g The journal name was used to find the associated discipline using Journal Citation Reports (InCites, 2018). When the journal was not indexed in this database, the scope of the journal was used to identify the discipline. See S3 Table for  i For articles, theses and dissertations, the target population was children when the mean age was between 5 and 12 years, adolescents when the mean age was between 12 and 18 years, young and middle-aged adults when the mean age was between 18 and 65 years and older adults when the mean age was over 65 years. When no mean age was provided, the age range was used to identify the target population most represented. For web pages, we used the target population identified in the document by the authors. The target population was considered as "all" when the document targeted all strata of the population or when no target population was identified by authors. https://doi.org/10.1371/journal.pone.0244292.t003

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review Samplers in the mindful raisineating task condition indicated higher levels of enjoyment of the sampled food than did samplers in either the nonmindful raisineating control condition or the no-task baseline condition.
--Mindful eating had no effect on the likelihood of sampling food. The mindfulness manipulation produced no differences in participants' mood. (Continued)

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review -Little change was seen from pre to post-course survey with regard to the consumption of foods that had neither been promoted nor discouraged in course videos (e.g., dairy). (Continued)

PLOS ONE
Eating pleasure and healthy eating: A scoping review

PLOS ONE
Eating pleasure and healthy eating: A scoping review reported more frequent monthly purchases of food from farmers' markets, 50% reported more frequent monthly purchases of olive oil, 75% reported increases in purchases of wine and wine consumption with meals, 81% reported increasing weekly wine consumption with meals, 50% reported an increase in the frequency of consuming "more satisfying" meals, 31% reported decreasing portion sizes whereas 6% reported an increase, 37% reported increasing "the amount of time I spend eating meals", more than 50% reported changes in eating behaviors. The number of meals prepared at home, the use of fresh ingredients, trying new recipes, interest in improving cooking skills, the variety of foods used to prepare food, and the use of olive oil in cooking were all increased by the majority of participants. In addition, 31% reported that they have decreased "the amount that I worry about calories" whereas 6% reported an increase in this behavior, and more than 80% reported an increase for "the amount that I make time for cooking, eating and enjoying food".  [148]. Nine studies [21,26,28,74,76,136,140,148,152] included multiple measures of eating pleasure while six intervention studies [7,31,32,37,159,161] that used eating pleasure in their strategies did not measure the impact of their interventions on eating pleasure. Dietary behavior and health outcomes. Dietary behavior and health outcomes for each study are presented in Table 5. Of the 45 studies identified, 31 (68.9%) included only dietary outcomes [7,13,[25][26][27][28][29]33,34,37,75,76,84,86,88,109,111,125,126,136,140,[145][146][147][148][150][151][152]159,160,161], 10 (22.2%) only health outcomes [30,35,36,74,89,112,137,[156][157][158] and 4 (8.9%) included both dietary and health outcomes [21,31,32,141].
Links between dimensions of eating pleasure and dietary/health outcomes. For each article, up to five dimensions of eating pleasure were identified by coders. In order of
In order to evaluate associations between dimensions of eating pleasure and health outcomes, we focused on dimensions identified in at least two different documents. When eating pleasure was conceptualized as sensory experiences, 3 out of 4 studies (75%) [21,30,89] showed that eating pleasure was associated with positive health outcomes. When eating pleasure was conceptualized as novelty, 2 out of 3 studies (66.7%) [89,137] showed that eating pleasure was associated with positive health outcomes. No dimension identified in at least two documents was predominantly associated with negative outcomes.

Q. 3 -Most promising intervention strategies using eating pleasure to promote healthy dietary behaviors
Description of included documents. Among the 13 intervention studies identified in this review (Table 5), one was excluded from this analysis since it did not permit identification of intervention strategies using eating pleasure to promote healthy dietary behaviors [125]. In addition, two studies reported results from the same long-term intervention [31,161]. Therefore, 12 intervention papers were included in this analysis [7,32,37,136,140,141,147,148,[159][160][161], reporting results from 11 independent interventions. Seven (63.6%) were short-term, single exposure interventions while four (36.4%) were longer-term, prolonged exposure interventions. Studies were all published after 2007. Interventions were conducted in the United States of America (n = 3; 27.3%), United Kingdom (n = 3; 27.3%), France (n = 1; 9.1%), Canada (n = 1; 9.1%), Germany (n = 1; 9.1%) or in multiple countries (n = 2; 18.2%). In sum, these intervention studies included 10,243 participants, of whom 7,835 (76.5%) were women, 1,569 (15.3%) were men and 839 (8.2%) for whom the sex was not specified. Sample size ranged from 16 to 7,385 participants. Nine were conducted in young and middle-aged adults only (81.8%), one in children only (9.1%) and one in both children and young and middle-aged adults (9.1%).
Intervention strategies using eating pleasure to promote healthy dietary behaviors. First, two studies investigated whether inducing a pleasure mindset influences portion sizes and food choices. Hege et al. [159] investigated whether different mindsets (healthiness mindset, pleasure mindset, fullness mindset) were associated with altered activity in certain brain areas during the pre-meal selection of portion sizes. Results showed that, compared with a control condition, participants reduced their portion sizes when considering eating for health or pleasure while increasing their portion sizes when considering eating to be full until the next meal. Petit et al. [7] assessed whether drawing attention to the pleasure of eating healthily increased healthy food choices in individuals with a high BMI. Participants evaluated pictures of 64 selected food items in three different "attention conditions" in which participants were either instructed to consider the benefits of eating healthy food (health condition), to consider the pleasure of eating healthily (pleasure condition), or to consider any features of the food that came to their mind (no diet condition). Results showed that participants were more likely to mention that they would eat the pictured healthy foods in the pleasure diet condition than in the no diet condition. Taken together, these results suggest that, compared with a control condition, a pleasure mindset may have favorable effects on portion sizes and food choices.
Four intervention studies investigated whether focusing on sensory sensations influences dietary intakes, eating behaviors and food choices. Gravel et al. [31,161] investigated whether a sensory-based 6-week intervention influenced eating related attitudes and behaviors among restrained women (dieters). Specific themes addressed were the relationship with food, hunger and satiety cues, sensory sensations, and the pleasures associated with eating. Compared with a waiting list control group, women in the sensory-based intervention decreased disinhibition (i.e., the tendency to overeat in response to cognitive or emotional cues), while increasing the intuitive eating subscale related to the unconditional permission to eat and the use of terms associated with the senses and with pleasure to describe food. No effects on restrained eating, susceptibility to hunger, and BMI were observed. Moreover, Cornil and Chandon [37] investigated how multisensory imagery can make people happier with smaller food portions. During the intervention, participants were asked to vividly imagine the multisensory pleasure (taste, smell, and texture) of three hedonic foods before choosing the size of another hedonic food. This short intervention led to mixed results, showing that, compared with several control conditions (e.g., non-food sensory imagery condition), focusing on sensory pleasure made hungry eaters and nondieters to choose and prefer smaller portions of hedonic foods while sated eaters and dieters chose larger portions. Two interventions using mindful eating strategies and focusing on the sensory properties of food were also performed. Arch et al. [140] showed that directing participants' attention as fully as possible towards the pleasure associated with the sensory experience of eating increased the enjoyment of a commonly pleasurable food (chocolate), and a food with generally more mixed associations (raisins), and led to lower calorie consumption of unhealthy food, an effect mediated by greater eating enjoyment as compared with a control condition. Finally, in a study by Hong et al. [141] participants were randomly assigned to a mindful raisin-eating task condition (i.e., subjects instructed to be nonjudgmental and fully aware, with all senses, of different aspects of the raisin), a non-mindful raisin-eating task (control) condition, or a no-task baseline condition. All participants were thereafter offered different types of food that they could sample. Subjects in the mindful raisin-eating task condition indicated higher levels of enjoyment of the sampled food than did subjects in the two other conditions. However, mindful eating had no effect on the likelihood of trying novel or typically avoided foods. In sum, results suggest that sensory-based interventions can have positive impact on eating behaviors and portion sizes. Mindful eating strategies focusing on the sensory properties of food also seem to increase eating enjoyment and to lower energy intake from unhealthy food while having no impact on the likelihood to sample a novel or typically avoided food.
Two interventions included aspects related to food preparation and/or sharing. Adam et al. [160] assessed the effectiveness of open, online nutrition and cooking instruction in improving eating behaviors of participants. Results showed that, in addition to increase the percentage of participants who found yesterday's dinner to be very or extremely enjoyable, the intervention resulted in significant changes in eating behaviors and meal composition, mirroring the messaging of the course. Mean overall perceived ease of making healthy food choices and preparing home cooked meals also increased significantly over time. Ensaff et al. [136] evaluated Jamie Oliver's Kitchen Garden Project (JOKGP). This project aims at developing food and cooking skills in children, increase their willingness to try new foods, and gain a better understanding of where food comes from. In the intervention school, children reported an increase in cooking enjoyment. They also reported an increase in helping with cooking at home and taste description score, and a decrease in food neophobia and fussiness. At follow-up, higher scores were reported for the children in the intervention school compared with those from the control school. Taken together, these results suggest that interventions including cooking lessons and sharing increase the enjoyment of eating and cooking, in addition to improving eating behaviors, diet quality, taste description, and involvement in cooking.
Two studies performed by Robinson et al. [147,148] included aspects related to memories. First, they examined the relationship between remembered enjoyment of vegetable eating, predictions about liking vegetables, and vegetable consumption [147]. Results showed that the recall of positive and enjoyable memories about past vegetable consumption resulted in higher liking for vegetables and the choice of a larger portion size of vegetables when compared with recall of a personal non-food memory, a non-vegetable food memory, or when visualizing someone else enjoying eating vegetables. They then examined if a simple instruction to rehearse what participants found enjoyable about a food immediately after eating it could be used to increase remembered enjoyment of a food (a Mediterranean vegetable quiche) and whether this would result in individuals choosing to eat more of this food at a later date [148]. Compared with controls (i.e., neutral meal rehearsal of the ingredients in the vegetable quiche and how long it took to eat it), participants in the intervention group showed a 90% increase in the amount of low-energy Mediterranean vegetable quiche chosen and eaten. In sum, these results suggest that recalling enjoyable eating memories related to healthy foods may increase the portion size of these foods later.
Finally, Sasson et al. [32] assessed the effects of a 3-week study program conducted in Italy that emphasizes the importance of enjoying the cooking and eating experience, taking time to eat, and valuing high-quality, unprocessed food. The program positively influenced behaviors related to food shopping, preparation, eating, and dieting six months later. More precisely, a majority of participants (i.e., 50% or more) reported increases in purchases of olive oil, locally grown food, seasonally and organically grown food, and food from farmers' markets. The consumption of more satisfying meals, the number of meals prepared at home, the use of fresh ingredients, the interest in improving cooking skills, the variety of foods used to prepare food, the use of olive oil in cooking and trying new recipes were also all increased by the majority of participants. Decreases in portion sizes and in worrying about calories were also observed. This program had no impact on self-reported weight. These results suggest that a more comprehensive intervention using eating pleasure as the core strategy may have many beneficial effects on dietary behaviors.
A summary of findings is presented in Table 7. This table groups together the dimensions based on their likely favorable or unfavorable impact and presents the dimensions in an order based on the level of evidence. Also, where applicable, intervention strategies that have shown promising effects for each dimension are also presented.

Discussion
Many authors now claim that eating pleasure is a promising approach to fostering healthy dietary behaviors [19,31,69,142,162] and some dietary guidelines, such as Canada's Food Guide [22], now include the enjoyment of food among their healthy eating recommendations. In spite of this, until now, no comprehensive review assessing the links between eating pleasure and dietary behaviors was available, making difficult to determine whether and how the pleasure of eating can be a lever for healthy eating and health. Our review contributes to filling some important gaps. First, it provides a more comprehensive overview of the key dimensions of eating pleasure, which may serve as a foundation for researchers interested in studying this concept. This literature review also mapped the evidence to date on the links relating eating pleasure with healthy eating and health, a key step for determining whether the pleasure of eating can be a potential ally in the promotion of healthy eating. This review also identified strategies using eating pleasure that seem promising to include in interventions aimed at promoting healthy eating. However, some major gaps remain in our understanding of how to use eating pleasure in the promotion of healthy eating habits, which we discuss below.
In total, 110 studies were examined for their way to conceptualize eating pleasure, including both scientific and grey literature. The review of the literature suggests that eating pleasure is a multidimensional concept. Indeed, in our qualitative assessment, two independent coders identified 22 different key dimensions used to define eating pleasure. The dimensions most frequently cited were sensory experiences, social experiences, food characteristics besides sensory attributes, food preparation process, novelty, variety, mindful eating, visceral eating, place where food is eaten, and memories associated with eating. This way to conceptualize eating pleasure is much broader than those found in previous reviews, which usually address only one or two dimensions of eating pleasure [163][164][165][166][167][168]. It is interesting to note that several similarities exist between this broader way to conceptualize eating pleasure and key dimensions we gathered in a study we recently conducted. In our focus groups [42], participants mostly defined eating pleasure in terms of taste, aesthetics, cooking, sharing a meal, relaxing, variety, nutritional aspects of food, and discovering new foods. In sum, our conceptual framework of eating pleasure provides evidence-based key dimensions and terminology that researchers may use to build consistency in the field and develop our understanding of the role of eating pleasure in the promotion of healthy eating habits. These results also highlight that eating pleasure may have different meanings, and that it is important for policymakers to clarify what is meant by this term when using it in policies and promotional materials. The purpose of this scoping review was to assess whether and how eating pleasure can be a lever for healthy dietary behaviors and health. Although results were not uniform across studies, our scoping review highlights that eating pleasure is mostly (57.1%) associated with positive dietary behavior outcomes. This was especially the case for food choices, as six out of eight studies showed favorable links. However, most studies investigating the links between eating pleasure and dietary behavior outcomes were cross-sectional, hindering the assumption of causality between eating pleasure and better dietary behaviors. Nevertheless, when we looked only at intervention studies, we observed even more consistent results. In addition to leading to better diet quality [140,160], healthier portion sizes [32,37,147,148,159] and better food choices [7,160], pleasure-related strategies used in these intervention studies also increased perception of tastiness and the liking of healthy food [147], time for and perceived ease of cooking and eating [32,136,160], proportion of meals prepared at home [32] and interest in improving cooking skills [32], while decreasing disinhibition (i.e., the tendency to overeat in response to cognitive or emotional cues) [31] and food neophobia and fussiness [136]. However, although these intervention studies had promising results concerning the potential impact of using eating pleasure in the promotion of healthy eating, these findings still need to be confirmed since to date, only 11 independent intervention studies have been conducted, and these used different methodologies, evaluated different outcomes and most of the time integrated only one or a few dimensions of eating pleasure. This may be explained by the fact that the integration of eating pleasure in the promotion of healthy eating is still at a nascent stage, all intervention studies being published after 2007.
The link between eating pleasure and health outcomes has been less studied and seems more inconsistent. Only two outcomes were studied in more than two studies: body weight/ BMI and depressive symptoms. For body weight and BMI, most studies reported that eating pleasure was associated with a higher weight/BMI. However, it is important to note that the two intervention studies that documented weight/BMI observed a neutral effect. Three crosssectional studies also assessed the link between eating pleasure and depressive symptoms, with 2 out of 3 studies showing no association between these two variables. This literature review therefore highlights a gap in research on links between eating pleasure and health-related variables, including body weight but also other health indicators less studied such as depression and anxiety symptoms, well-being, self-esteem and body weight dissatisfaction.
Tools used to measure eating pleasure in studies examining the links between eating pleasure and dietary and health outcomes used a great variety of approaches. In total, 37 different tools were used to measure eating pleasure, which greatly compromises comparisons across studies. As there seems to be no universal definition of eating pleasure in the literature, authors could define this concept according to their own perception. Inconsistent approaches and measurement of eating pleasure between studies may be at least partly responsible of divergence in results reported. The development of standardized and validated measurement tools is thus essential to have a better understanding of the links relating eating pleasure, and its key dimensions, with dietary and health outcomes.
To overcome the inconsistency in the measurement of eating pleasure, we decided to describe the associations between eating pleasure and dietary and health outcomes, taking into consideration the different dimensions that defined eating pleasure in each study. On the one hand, when eating pleasure was conceptualized as sensory experiences, mindful eating, memories, social experiences, food preparation process and food anticipation, favorable dietary outcomes were reported in the literature. In addition, findings from intervention studies suggested that sensory-based interventions (e.g., food-tasting activities; multisensory imagery), mindful eating strategies focusing on the sensory experience of eating, the rehearsal of enjoyable eating memories related to healthy foods, cooking lessons/experiences and sharing activities may result in beneficial effects on dietary behaviors. It is worth noting that these findings are consistent with those of other studies that did not specifically focus on eating pleasure, but that observed favorable outcomes using similar intervention strategies as those identified as promising in this review [169][170][171][172]. On the other hand, visceral eating pleasure, defined as eating pleasure associated with short-term visceral impulses triggered by hunger, external cues or internal emotional urges, was more strongly related to detrimental effects on dietary outcomes according to cross-sectional studies. For health outcomes, most cross-sectional studies found that sensory experiences and novelty were favorably associated with these outcomes. Taken together, these results suggest that the different dimensions of eating pleasure are characterized by a specific pattern of associations with dietary and health outcomes, some being more favorable than others. These promising avenues may then inform whether and how the pleasure of eating can be a lever for healthy eating. However, even if this review was an essential first step to propose impactful strategies in the promotion of healthy eating, it will be essential for health professionals and policymakers to tailor messages and intervention strategies to fit the local context, including social, cultural and socioeconomic factors, to ensure that these strategies are well aligned with community needs and interests. Additional steps, including a comprehensive needs assessment, are therefore needed before using these promising strategies in policies and dietary interventions within a specific population.
It is important to highlight some gaps identified by this review. First, considering the high number of dimensions of eating pleasure (89 sub-dimensions grouped into 22 key dimensions), some of these have been less studied (e.g., place and atmosphere where food is consumed, eating on special occasions, having the choice, taking time). The lack of studies, however, does not necessarily indicate that these dimensions of eating pleasure are not favorably or unfavorably related to dietary behaviors and health outcomes. Rather, these findings highlight the fact that research on the pleasure of eating in the context of promoting healthy eating is still in its infancy and further studies are needed to get a more comprehensive perspective on the place of eating pleasure, and its multiple dimensions, in the promotion of healthy eating. In addition, some limitations of the included intervention studies need to be highlighted. First, intervention studies were more likely to include only one dimension of eating pleasure, to be short-term single exposure interventions, and to be performed with women. However, some reasons may explain these methodological choices. Short-term exposure can be more easily used to test hypotheses at the outset, allowing to identify more efficiently dimensions of eating pleasure that seem to be more promising in the promotion of healthy eating. These promising dimensions can then be tested in larger, longer-term interventions. Also, the testing of one dimension at a time allows for a better assessment of the impact of this dimension on the variables of interest, which can be very useful when designing more complex interventions afterwards. The fact that more women have been included in previous intervention studies is congruent with what is observed in other systematic reviews in nutrition, highlighting most of the time a greater proportion of women than men in these studies [173][174][175][176]. These gender disparities may be explained by gender-specific differences in many areas of nutrition, including the fact that women have a greater interest in nutrition and seek nutrition counselling more frequently than men do [177]. Finally, no intervention study examined the effectiveness of intervention strategies by sex, age or BMI, and only two comprehensive intervention studies using different dimensions of eating pleasure were reported in the present review [31,32]. Therefore, these findings highlight some research gaps and the need for conducting well-developed and evidence-based intervention studies before reaching a firm conclusion about the place of eating pleasure, and its multiple dimensions, in interventions aiming to promote healthy eating.

Methodological strengths and limitations
This scoping review has many methodological strengths. First, the search strategy was developed with the help of a librarian specializing in nutrition-related research, limiting the risk of missing relevant documents. The search strategy included a wide range of databases, allowing us to capture a comprehensive sample from the many disciplines related to eating pleasure. In addition, the search strategy included both scientific and grey literature, allowing for a more comprehensive approach of eating pleasure. A rigorous methodology comprising two independent coders (one specialized in nutrition and one specialized in social sciences) who reviewed evidence at each step of the scoping review resulted in a substantial reduction in possible errors. In addition, the use of the scoping review per se is a strength, allowing us to address a broad, multidisciplinary topic using a wide range of research designs. Finally, we used the PRISMA extension for scoping reviews checklist to report these results, ensuring methodological and reporting quality [178]. Some limitations of the present review must also be acknowledged. First, we included only studies conducted in developed countries, with participants aged 5 years old and older, limiting the generalization of our results. Second, we reviewed only documents written in English or French, thus we cannot exclude the possibility that we missed some studies. Finally, several documents mentioned dimensions of eating pleasure only very briefly, thus making interpretation difficult for coders. In order not to bias the results of this review, the coders therefore analyzed them as they were described, without over-interpreting them. Accordingly, we do not exclude the possibility that some dimensions can be combined in further analyses.

Conclusion
This comprehensive review of the literature about eating pleasure showed that this concept involves numerous key dimensions. It showed favorable links between eating pleasure and dietary outcomes and identified promising strategies for using eating pleasure in intervention strategies. However, the role occupied by studies evaluating the effects of eating pleasure on healthy eating is still minimal, and additional well-developed and evidence-based intervention studies are needed before reaching a firm conclusion about the place that eating pleasure, and its multiple dimensions, should occupy in the promotion of healthy eating.
Supporting information S1