Traffic-light front-of-pack environmental labelling across food categories triggers more environmentally friendly food choices: a randomised controlled trial in virtual reality supermarket.

BACKGROUND: Food systems highly contribute to anthropogenic greenhouse gas emissions and shifting towards more environmentally friendly diets is urgently needed. Enabling consumers to compare the environmental impact of food products at point-of-purchase with front-of-pack labelling could be a promising strategy to trigger more environmentally friendly food choices. This strategy remained to be tested. METHODS: The effect of a new traffic-light front-of-pack environmental label on food choices was tested in a 2-arm randomised controlled trial in a virtual reality supermarket. Participants (n = 132) chose food products to compose two main meals for an everyday meal scenario and for an environmentally friendly meal scenario with or without the label. The environmental label (ranging from A: green/lowest impact, to E: red/highest impact) was based on the Environmental Footprint (EF) single score calculation across food categories. The effect of the label on the environmental impact of food choices in each scenario was tested using linear mixed models. RESULTS: In the everyday meal scenario, the environmental impact of meals was lower in the label condition than in the no label condition (-0.17 ± 0.07 mPt/kg, p = 0.012). This reduction was observed at no nutritional, financial nor hedonic cost. The effectiveness of the label can be attributed to a change in the food categories chosen: less meat-based and more vegetarian meals were chosen with the label. In the environmentally friendly meal scenario, we demonstrated that the label provided new information to the participants as they were able to further reduce the environmental impact of their food choices with the label (-0.19 ± 0.07 mPt/kg, p = 0.005). CONCLUSIONS: Implementing a front-of-pack environmental label on food products in real supermarkets could increase awareness of the environmental impact of food and contribute to drive more environmentally friendly food choices. TRIAL REGISTRATION: The study protocol was pre-registered prior to data collection at Clinicaltrials.gov (NCT04909372).

Using event-related potentials to study food-related cognition: An overview of methods and perspectives for future research.

Electroencephalography (EEG), and the measure of event-related potentials (ERPs) in particular, are useful methods to study the cognitive and cerebral mechanisms underlying the perception and processing of food cues. Further research on these aspects is necessary to better understand how cognitive functioning may influence food choices in different populations (e.g. obese individuals, individuals with eating disorders). To help researchers in designing future studies, this article provides an overview of the methods used in the current literature on ERPs and food-related cognition. Several methodological aspects are explored to outline interesting perspectives for future research, including discussions on the main experimental tasks used, the cognitive functions assessed (e.g. inhibitory control, attentional processing), the characteristics of the participants recruited (e.g. weight status, eating behaviors), and the stimuli selected (e.g. food pictures, odors). The issues generated by some of these methodological choices are discussed, and a few guidelines are provided.

Using CATA tests to capture consumers' mental representations elicited by images of pulse-based food products with different levels of processing.

Consumption of pulses plays a key role in achieving a healthier and more sustainable diet. Despite the increased availability of pulse-based foods on the French market, pulse consumption in France remains below recommended levels. The objective of this study was to capture French consumers' mental representations of various pulse-based foods with different levels of processing, using a Check-All-That-Apply (CATA) questionnaire with food images. We used CATA questionnaires, in two separate studies, to capture two aspects: consumers' beliefs about the products presented in the study, and the characteristics of those products, according to consumers. In Study 1, twelve images of pulses were presented (unprocessed, processed, and ready-to-eat) to investigate whether and how consumers differentiate between pulse-based products with different recipes and levels of processing. In Study 2, parallel CATA questionnaires were used: one with images of pulse-based products, and one with images of vegetables. We wanted to discover whether the factors underlying different representations for pulse-based products were specific to pulses, or were also observed with another family of food products. Results showed that products were differentiated more by their level of processing than by their main ingredient. Participants had more positive attitudes toward pulse products that were unprocessed, processed, or "new ready-to-eat", than toward other ready-to-eat pulse products (whether canned or frozen). Our results also highlighted that, in consumers' mental representations of pulse-based products, packaging format is a negligible factor in comparison with the nature of the product itself.

French consumers know the benefits of pulses, but do not choose them: An exploratory study combining indirect and direct approaches.

Pulses present nutritional advantages for human health, and also contribute to food security and environmental sustainability. Despite these beneficial properties, the consumption of pulses in France has decreased over the past century. This study explores the representation of pulses among French non-vegetarian consumers, using both an indirect approach, with scenarios evoking real-life situations, and a direct approach, with an online questionnaire. For the indirect approach, based on six different scenarios, the 120 participants composed dishes by selecting three food images out of the twenty provided. One week later, for the direct approach, the same participants completed a questionnaire, with ten questions about food products, using the same images. Results showed that spontaneous selection of pulses was low for all dish-composition situations, coherent with the low consumption of pulses declared in the questionnaire. Among possible barriers to pulse consumption were dislike of pulses, preparation difficulty, and perception of pulses as a food for vegetarians. Participants possessed theoretical knowledge about the benefits of pulses, but this knowledge seemed not to be applied to dish-composition situations. Results also highlighted the complementarity of these two methods, providing deeper understanding of consumer food habits in relation to pulses. Based on these findings, several levers are proposed to increase pulse consumption: improving familiarity with pulses, using the context of consumption, formulating trendy and innovative products, and promoting pulses as a "food for all".

Learned pleasure from eating: An opportunity to promote healthy eating in children?

Across the lifespan, eating is a common everyday act driven by the search for pleasure and reinforced by experienced pleasure. Pleasure is an innate indicator of the satisfaction of physiological needs, in addition to other attributes. Pleasure from eating is also learned and contributes to the development of children's eating habits, which remain mostly stable until adulthood. Based on classical models of determinants of food consumption behaviour, we identified three dimensions of pleasure from eating learned during childhood: 1/the sensory dimension, i.e., pleasure from sensory sensations during food consumption; 2/the interpersonal dimension, i.e., pleasure from the social context of food consumption; and 3/the psychosocial dimension, i.e., pleasure from cognitive representations of food. The objective of this narrative review is to explore whether these three dimensions may play a role in promotion of healthy eating behaviour among children. Up to now, it was assumed that providing nutritional information, pointing out which types of foods are "good" or "bad" for health, would drive healthier food choices in children. Today, we know that such strategies based on a cognitive approach toward eating have a limited impact on healthy choices and can even be counter-productive, leading children to avoid healthy foods. In the context of increasing rates of childhood obesity, new perspectives are needed to build efficient interventions that might help children adopt a healthy diet. This review suggests new directions for further research to test the efficacy of novel interventions that emphasize pleasure from eating.

When do healthiness and liking drive children's food choices? The influence of social context and weight status.

Children identify liking and healthiness of foods as factors influencing their food choices. However, the food decision making process is also influenced by both personal characteristics and food contexts. The present study explored the influence of liking and perceived healthiness of foods in normal- and overweight children's food choices intentions in a pleasure-oriented social eating context and a health-oriented social eating context. Children aged from 6 to 11 years old (n = 63; 34 children who were of normal weight and 29 who were overweight) were asked to select 5 foods among 10, based on food pictures, to make up a snack that would be suitable for their birthday party or a nutrition class. In addition, they rated their liking and healthiness perception of the foods. No significant difference in food choices was found between children who were of normal weight and children who were overweight. Both groups of children chose more healthy food items in a health-oriented social context (i.e., a fictive nutrition class) than in a pleasure-oriented social context (i.e., a fictive birthday party). Moreover, only liking significantly predicted food choices in the pleasure-oriented social context whereas both healthiness and liking significantly predicted food choices in the health-oriented social context. Overall these results advance our understanding on how children make food decisions and inform strategies that may help children to adopt a healthy diet. Because liking predicted children's food choices in both eating contexts, emphasizing the "good" taste of healthy foods and providing children with healthy foods they like could be efficient strategies to promote healthy eating habits in children.

Do hedonic- versus nutrition-based attitudes toward food predict food choices? a cross-sectional study of 6- to 11-year-olds.

BACKGROUND: Implicit and explicit attitudes are potential precursors of food choices and combine affective and cognitive components that can vary in their relative dominance. Yet, the affective and cognitive components of attitudes toward food can lead to distinct predisposition toward a food item and potentially to different food choices. In the food domain, the affective component pertains to the hedonic tone of consumption, while the cognitive component encompasses nutritional value or health consequences of food. The present study investigated whether hedonic- versus nutrition-based implicit and/or explicit attitudes toward food predicts children's healthy versus unhealthy food choices. METHODS: A total of 63 children (age range = 6.3-11.5) participated in a 90-min session at 5 pm (i.e., afterschool snack time in France). The children were asked to choose five food items from a buffet featuring five healthy and five unhealthy sweet foods pretested as being highly liked. Children ate what they had chosen. Moreover, their implicit attitudes were assessed with a pairing task in which children were presented with 10 food triplets and asked to choose two food items that "best go together". For each triplet, foods could be paired according to their hedonic or nutritional characteristics. Explicit attitudes were assessed with a task in which children placed each of 48 food items into one of the following categories: "yummy", "yucky" (i.e., hedonic categories), "makes you strong", or "makes you fat" (i.e., nutritional categories). RESULTS: Both implicit and explicit attitudes significantly influenced children's food choices. We observed that children with more hedonic-based implicit or explicit attitudes toward food were more likely to choose healthy food options from the buffet. Conversely, children with both implicit and explicit nutrition-based attitudes chose less healthy foods. CONCLUSIONS: Hedonic-based attitudes toward food seem to drive healthier food choices in children compared with nutrition-based attitudes in this particular eating context. These findings suggest that pleasure from eating might be an ally with regard to healthy eating among children. Additional research is needed to understand the etiology of children's attitudes toward food in order to provide insights on how to shape adequate children's attitudes to guide them toward healthy food choices.

Explicit and implicit tasks for assessing hedonic-versus nutrition-based attitudes towards food in French children.

Attitudes are important precursors of behaviours. This study aims to compare the food attitudes (i.e., hedonic- and nutrition-based) of children using both an implicit pairing task and an explicit forced-choice categorization task suitable for the cognitive abilities of 5- to 11-year-olds. A dominance of hedonically driven attitudes was expected for all ages in the pairing task, designed to elicit affective and spontaneous answers, whereas a progressive emergence of nutrition-based attitudes was expected in the categorization task, designed to involve deliberate analyses of the costs/benefits of foods. An additional exploratory goal was to evaluate differences in the attitudes of normal and overweight children in both tasks. Children from 3 school levels (n = 194; mean age = 8.03 years) were individually tested on computers in their schools. They performed a pairing task in which the tendencies to associate foods with nutritional vs. culinary contexts were assessed. Next, they were asked to categorize each food into one of the following four categories: "yummy", "yucky" (i.e., hedonic categories), "makes you strong", or"makes you fat" (i.e., nutritional categories). The hedonic/culinary pairs were very frequently selected (81% on average), and this frequency significantly increased through school levels. In contrast, in the categorization task, a significant increase in nutrition-driven categorizations with school level was observed. Additional analyses revealed no differences in the food attitudes between the normal and overweight children in the pairing task, and a tendency towards lower hedonic categorizations among the overweight children. Culinary associations can reflect cultural learning in the French context where food pleasure is dominant. In contrast, the progressive emergence of cognitively driven attitudes with age may reflect the cognitive development of children who are more reasonable and influenced by social norms.

Effects of psychological stress on the emission of volatile organic compounds from the skin.

Thirty-five women were included in a clinical study to characterize the volatile organic compounds (VOCs) emitted by the skin during exposure to psychological stress. An original silicon-based polymeric phase was used for VOC sampling on the forehead before and after stress induction. Cognitive stress was induced using specialized software that included a chronometer for semantic and arithmetic tasks. Assessment of stress was monitored using a State-trait anxiety inventory questionnaire, analysis of participants' verbal expressions and clinical measurements. Identification and relative quantification of VOCs were performed by gas chromatography-mass spectrometry. Stress induction was validated by a significant increase in state-anxiety as indicated by the questionnaire, modifications in electrodermal activity measurements and the expression of stress verbatims. In parallel, a sebum production increase and a skin pH decrease were observed. A total of 198 VOCs with different potential sources were identified. They were categorized in 5 groups: probable cosmetic composition, VOCs produced by the body or its microbiota, environmental origin, and dietary intake. In our qualitative statistical approach, three VOCs were found to be correlated with stress induction and 14 compounds showed significance in the paired Wilcoxon test. Fatty-acyls derived from lipids were predominantly identified as well as ethylbenzenes.

Chronic insomnia, high trait anxiety and their comorbidity as risk factors for incident type 2 diabetes mellitus.

The main objective of this study was to evaluate the association of the insomnia-anxiety comorbidity with incident type 2 diabetes (T2D) in a large prospective cohort. We selected adults without diabetes at baseline from the French NutriNet-Santé cohort who had completed the trait anxiety subscale of the Spielberger State-Trait Anxiety Inventory (STAI-T, 2013-2016) and a sleep questionnaire (2014); insomnia was defined according to established criteria. Using multivariable Cox models, we compared T2D risk across 4 groups: no insomnia or anxiety (reference), insomnia alone, anxiety alone (STAI-T ≥ 40), and comorbid anxiety and insomnia. Among 35,014 participants (mean baseline age: 52.4 ± 14.0 years; 76% women), 378 (1.1%) developed T2D over a mean follow-up of 5.9 ± 2.1 years. Overall, 28.5% of the sample had anxiety alone, 7.5%-insomnia alone, and 12.5%-both disorders. In the fully-adjusted model, a higher T2D risk was associated with anxiety-insomnia comorbidity (HR = 1.40; 95% CI 1.01, 1.94), but not with each disorder separately, compared to the group without insomnia or anxiety. The findings supported a positive association between anxiety-insomnia comorbidity and incident T2D among general-population adults. Future research using clinical diagnoses of mental disorders could confirm the findings and guide diabetes prevention programs.

A scoping review of emotion regulation and inhibition in emotional eating and binge-eating disorder: what about a continuum?

BACKGROUND: Emotional eating is defined as a nonpathological eating behavior, whereas binge-eating disorder (BED) is defined as a pathological eating behavior. While different, both share some striking similarities, such as deficits in emotion regulation and inhibition. Previous research has suggested the existence of an "eating continuum" that might reflect the increased severity of overeating behaviors, that is, from nonpathological overeating to BED. The main aims of this scoping review were to explore in the literature the idea of a continuum between emotional eating and BED and to observe whether deficits in emotion regulation and inhibition follow this continuum in terms of severity. The other aims were to hopefully clarify the ill-defined concept of overeating, to question the potential role of positive emotions and to identify potential knowledge gaps. METHOD: A systematic scoping review was conducted following the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines. Two databases (PubMed/Medline and PsycINFO) were examined in complete accordance with the beforehand sharply defined eligibility and exclusion criteria. The main criteria included adults (≥ 18) with emotional eating, BED or overeating and emotion regulation and inhibition as exposure criteria. RESULTS: Thirty-two studies were included in this scoping review. If the results showed a link between emotional eating and BED, with the presence of inhibition and emotion regulation deficits in both eating behaviors, no mention of a continuum between emotional eating and BED was found. CONCLUSION: In the absence of research directly comparing emotional eating and BED in the same studies and testing the potential increase in severity of emotion regulation and inhibition deficits along this continuum, there is currently no certainty that a continuum exists between emotional eating and BED. In the end, the idea of a continuum in terms of increased severity of overeating and in terms of emotion regulation and inhibition deficits between emotional eating and BED appears to be a gap in knowledge in the literature. This scoping review highlights the need for further research to identify knowledge gaps.

Emotional eating (EE) is defined as a nonpathological eating behavior, whereas binge-eating disorder (BED) is defined as a pathological eating behavior. While different, both share some striking similarities, such as deficits in emotion regulation (ER) and inhibition. Previous research has suggested the existence of an "eating continuum" that might reflect the increased severity of overeating behaviors, that is, from nonpathological overeating to BED. The main aims of this scoping review were to explore in the literature the idea of a continuum between EE and BED and to observe whether deficits in ER and inhibition follow this continuum in terms of severity. A systematic scoping review was conducted, and thirty-two studies were included in this review. If the results showed a link between EE and BED, with the presence of inhibition and ER deficits in both eating behaviors, no mention of a continuum between EE and BED, or in relation to a continuum, was found. Thus, in the absence of research directly comparing EE and BED in the same studies and testing the potential increase in severity of ER and inhibition deficits along this continuum, there is currently no certainty about the existence or absence of such a continuum.

Consumer Attitudes toward Pulses: Measuring the Implicit.

Research into consumer attitudes toward food products is important to help people adopt healthier, more sustainable diets. A positive attitude regarding an object is a prerequisite for its adoption. This study compares French consumers' implicit attitudes toward pulses and cereals. Many studies have measured attitudes by explicit methodologies (e.g., questionnaires). Such methods are often biased by social desirability, and consumers may not be consciously aware of their attitudes toward food. A Sorting Paired Feature Task measures the strength of automatic associations, pairing images of pulses or cereals and adjectives with positive or negative valence. Participants sorted 120 paired stimuli as fast as possible. Pairs composed of pulses and negative adjectives were sorted faster than pairs composed of cereals and negative adjectives. Cereals with positive adjectives were sorted more rapidly than pulses with positive adjectives. Mistaken associations were more frequent for pairs composed of cereals and negative adjectives than for pairs composed of pulses and negative adjectives. These results highlight more negative implicit attitudes toward pulses than cereals. This study provides the first potential evidence of negative implicit attitudes toward pulses, which could explain the low consumption of these products.

Inhibition and shifting across the weight status spectrum.

Executive functioning (EF) is of major interest in the study of cognitive factors involved in obesity. Among EF, shifting is related to behavioral flexibility, and inhibition to the ability to refrain from impulsive behavior. A deficit in those two EF could predict individual difficulties to maintain a healthy lifestyle. Weak evidence of deficits in shifting and inhibition in individuals of higher Body Mass Index (BMI) have been observed. The objective was to clarify the relationship between inhibition and shifting regarding weight status group differences in healthy adults. Two neuropsychological tests from the Test of Attentional Performance (TAP) battery were used to measure EF performance of three groups of men and women: normal-weight (NW, n = 38), overweight (OW, n = 40) and obesity (OB, n = 37). The results show that individuals with higher BMI have lower inhibition capacities and that classically used weight status categories might not capture cognitive variability. No differences in shifting were observed concerning weight status nor BMI. This paper provides new insights on cognitive factors in obesity by presenting data from healthy individuals with overweight and obesity. The results support that assessing inhibition capacities might be of interest in a clinical setting for patients with difficulties to lose weight.

Do Food Odors Differently Influence Cerebral Activity Depending on Weight Status? An Electroencephalography Study of Implicit Olfactory Priming Effects on the Processing of Food Pictures.

Attentional automatic processes and cerebral activity may differ between individuals with different weight statuses in the presence of food stimuli (e.g. odors, pictures). In the present study, we used an implicit olfactory priming paradigm to test the influence of non-attentively perceived food odors on the cerebral activity underlying the processing of food pictures, in normal-weight, overweight, and obese adults. A pear odor and a pound cake odor were used as primes, respectively priming sweet low-energy-density foods and high-energy-density foods. Event-related potentials were recorded while the participants passively watched pictures of sweet low and high-energy-density foods, under the two priming conditions plus an odorless control condition. The amplitude and latency of several peaks were measured (P100, N100, P200, N400). As a major result, we found that weight status influences the cerebral activity underlying the processing of food cues outside of consciousness, as early as the first detectable P100 peak.

Implicit food odour priming effects on reactivity and inhibitory control towards foods.

The food environment can interact with cognitive processing and influence eating behaviour. Our objective was to characterize the impact of implicit olfactory priming on inhibitory control towards food, in groups with different weight status. Ninety-one adults completed a modified Affective Shifting Task: they had to detect target stimuli and ignore distractor stimuli while being primed with non-attentively perceived odours. We measured reactivity and inhibitory control towards food pictures. Priming effects were observed on reactivity: participants with overweight and obesity were slower when primed with pear and pound cake odour respectively. Common inhibitory control patterns toward foods were observed between groups. We suggest that non-attentively perceived food cues influence bottom-up processing by activating distinguished mental representations according to weight status. Also, our data show that cognitive load influences inhibitory control toward foods. Those results contribute to understanding how the environment can influence eating behaviour in individuals with obesity.

Do French Consumers Have the Same Social Representations of Pulses as Food Industry Professionals?

Pulses present many advantages for human health, nutrition, sustainability, and the environment. Despite efforts in recent years by the pulse industry and national authorities to favor pulses, consumption in France remains relatively low, at 1.7 kg/per person in 2016, compared to 1920 when it was around 7.2 kg/per person. To understand social representations of pulses in France, 80 French nonvegetarian consumers and 35 professionals from the pulse industry were asked to say five words spontaneously evoked by the inductor "pulses". They then had to rank these five words in order of importance and rate their valence. The structural approach was used to analyze social representations for each group independently. Our results highlight differences in the structure and content of social representations for pulses. Consumer responses suggested only vague impressions of pulses, but taste evocations were nevertheless rated positively. By contrast, professionals tended to focus specifically on protein content and culinary preparation. These differences could explain some barriers to pulse consumption, and improved communication should be a key target. Efficient communication must consider the concepts most frequently used by consumers when referring to pulses, and those ranked as most important.

Weight Status and Attentional Biases Toward Foods: Impact of Implicit Olfactory Priming.

Objective: Numerous studies highlight the involvement of cognitive factors in the development and maintenance of obesity. We aimed to measure attentional biases (AB) toward foods (i.e., the individual tendency to automatically orient one's attention toward food stimuli) in normal-weight (NW) individuals and those with overweight (OW) and obesity (OB). We evaluated whether implicit or explicit exposure to olfactory food cues could modify AB. Methods: Eighty-five participants with different weight statuses took part in this experiment. We measured AB toward food pictures with an adapted visual probe task and the variations in AB, while participants were primed with olfactory food cues (within-subject design: no odor/low-energy dense food odor/high-energy dense food odor). Odors were non-attentively perceived during session 1 (implicit condition) and attentively perceived during session 2 (explicit condition). Results: Our results highlighted AB toward food pictures, especially when foods were energy dense, regardless of weight status (p < 0.001). The olfactory priming effect was only significant in the implicit condition. Participants with obesity had a stronger AB toward foods when they were primed with a non-attentively perceived high-energy dense food odor than with a non-attentively perceived low-energy dense food odor (p = 0.02). The trend was reversed for normal-weight participants, while no significant effect was found for participants with overweight. Conclusion: Our results support the hypothesis that an obesity-specific cognitive vulnerability may influence the processing of food-related stimuli and only while food cues are non-attentively perceived. Future research should seek to understand the mechanisms of this phenomenon.

A central place for meat, but what about pulses? Studying French consumers' representations of main dish structure, using an indirect approach.

Over time, meat has acquired a central place in French gastronomy. The role played by plant-based proteins, such as pulses, is less clear. In order to better understand it, this study seeks to identify how French non-vegetarian consumers structure their main dish, using an indirect approach. A total of 120 participants had to compose dishes, following six different scenarios, by selecting three images of food items out of the twenty proposed. Results provided information about: (1) the first food product chosen, and (2) how foods from different food-groups were associated. Our results indicated that (1) French non-vegetarian consumers generally constructed their main dishes using animal-based food products first. Some differences were nevertheless identified, in relation to the scenario presented, and the profile of the participants. Results also showed that (2) the food-groups most often associated within a dish were meat, starch, and vegetable. The pulse group, when used, was in general associated with the meat and vegetable food-groups. These findings provide new evidence of the structure of the French main dish. They also highlight possible levers that could be used to promote pulses to French consumers.

Impact of Olfactory Priming on Food Intake in an Alzheimer's Disease Unit.

Alzheimer's disease (AD) is often associated with feeding difficulties and changes in eating behavior with may lead to malnutrition. In French nursing homes, AD patients may live in special care units that better meet dementia residents' needs. However, meals are often delivered to AD patients by using meal trays coming from central kitchens. This led to the disappearance of cues that could help residents to foresee mealtime, such as the smell of food odors. The aim of the present study was to assess the impact of odorizing the dining room of AD Units with a meat odor before lunch on subsequent food intake and eating behavior. Thirty-two residents (>75 years old) from three AD Units were included in the study. They participated in two control lunches and two primed lunches, for which a meat odor was diffused in the dining room 15 minutes before the arrival of the meal tray (olfactory priming). Results of the first replication showed a significant effect of olfactory priming, with a 25% increase in meat and vegetable consumption compared to the control condition. Behavioral measurements also showed a significant increase of resident's interest toward the meal in the primed lunch. However, this effect was no longer observed when the priming session was replicated two weeks later with the same priming odor and the same menu. Although further research is needed to understand why this priming effect cannot be replicated, our experiment is one of the very first to investigate the effect of food odor priming on subsequent food intake in AD patients in a real-life setting.

Non-Conscious Effect of Food Odors on Children's Food Choices Varies by Weight Status.

OBJECTIVE: Food cues are omnipresent in the daily environment and may influence eating behavior even non-consciously. An increased reactivity to food cues, such as food odors, has been shown to be correlated with obesity in children. The objective of this study is to investigate whether the non-conscious influence of food odors on children's food choices varies by their weight status. METHODS: Seventy-four children, of whom 29 were obese, took part in this study. The children performed a food choice intention task presented as a computer game in which 30 pairs of food images (a fatty-sweet food picture vs. a fruit picture) successively appeared on the screen. The children had to choose the item "they most wanted to eat at the present moment" for each pair. While performing this task, the children wore a headset in which the microphone foam was odorized with a fruity odor, a fatty-sweet odor or no odor. They performed the intention task three times, one time for each olfactory condition. The odors were non-attentively perceived, i.e., none of the children were aware of the odorization of the microphone foams. The modeled probability is the probability to choose a fruit. RESULTS: In children with obesity, the fruity odor increased the likelihood of a fruit to be chosen compared to the no-odor condition [OR (95% CL) = 1.42 (1.13-1.78), P = 0.0028], while the fatty-sweet odor had no effect on food choice [OR (95% CL) = 1.07 (0.85-1.36), P = 0.55]. In children without obesity, both the fruity and the fatty-sweet odors decreased the likelihood to choose a fruit compared to the no-odor condition [OR (95% CL) = 0.76 (0.64-0.90), P = 0.0015, for the fruity odor and OR (95% CL) = 0.79 (0.66-0.93), P = 0.0062, for the fatty-sweet odor]. CONCLUSION: The different patterns of results obtained in both groups of children suggest differences in the mental representations activated by non-attentively perceived olfactory cues based on weight status.