Principals of Nutrition Article summary in APA format.... KIM WOODS?PHYLLIS YOUNG!!!!

May 2013; 64(3): 261–268

The relationship between emotions and food consumption

(macronutrient) in a foodservice college setting – a preliminary study

HEATHER J. HARTWELL, JOHN S. A. EDWARDS, & LORRAINE BROWN

Foodservice and Applied Nutrition Research Group, Bournemouth University, Poole, UK

Abstract

Many aspects of eating out have been studied, yet emotions remain an under-researched area, despite having been shown to play a significant role in food consumption. The aim of this research is to critically evaluate the relationship between emotions and food consumption (macronutrient) in a realistic eating environment, a college cafeteria. Subjects (n ¼ 408), diners using a cafeteria, completed an emotions questionnaire before and after freely choosing, paying for and consuming a hot main meal. The results demonstrated a greater feeling of contentment with a high fat, high energy meal, whereas with a low carbohydrate meal, participants felt unfulfilled. In addition, a high protein meal also leads to a feeling of contentment. These results are rather counterintuitive to public health nutrition policy but indicate the importance of inclusion of a protein or high carbohydrate item in any dish design in a foodservice setting.

Keywords: food intake, emotion, nutrition

Introduction

262 H. J. Hartwell et al.

Emotions and food consumption 263

Acquiring macronutrients from outside the home is becoming more prevalent and many aspects of eating out have been studied, yet emotions remain an underresearched area. The commonest way in which food can affect behaviour is through changes in mood and arousal that occur from before to after eating a meal (Gibson 2006), yet there is a paucity of research on how emotions influence or change when eating a selfselected meal out of the home and what effect these might have on macronutrient intake.

Past studies have concentrated on subjects predisposed to overweight and obesity where the tendency to eat more when experiencing negative emotions has been identified as contributing to these conditions, while inversely, the tendency to eat less when experiencing negative emotions has also been identified as contributing to being underweight (Nolan et al. 2010). This paper, part of a larger study on the role of emotions when eating out, addresses this scarcity of information and uses a real-life eating environment as context and in so doing, not only helps us to better understand the role of emotions, but also provides valuable data and insights into the body of knowledge on emotional response to an eating situation.

There is no standard definition, and emotion and mood are terms often used interchangeably. Plutchik (1980, in Ritchins 1997) reviewed 28 definitions of emotions and concluded that not only there is little consistency but also many were not sufficiently explicit to give a clear idea of what an emotion actually is. More recently Cabana (2002) reinforced this lack of consensus with King and Meiselman (2010) considering the distinction to be easier to delineate in theory than in practice.

Nevertheless, there is some agreement that an emotion is any short-term evaluative, affective, intentional, psychological state including happiness, sadness, disgust and other inner feelings, whereas mood is a temporary but relatively sustained and pervasive affective state, often contrasted in psychology and psychiatry with a more specific and short-term emotion (Gibson 2006; Dictionary of Psychology in Politics and Social Sciences 2010).

Emotions can be viewed from a number of perspectives and affect many aspects of an individual’s

life including, hunger, thirst and eating (Rolls 2007). In a review of the theories relating emotion and eating behaviours, Canetti et al. (2002) concluded that emotions influence eating behaviour in humans with negative emotions increasing food consumption, although the role of positive emotions is less conclusive. In addition, they also suggest that frequent emotions such as joy and anger have a greater impact; that the influence of emotions is stronger in obese rather than normal weight individuals and that obese individuals engage in emotional eating more than nonobese individuals, as do dieters. Emotions have been demonstrated to play a significant role in consumption. In Macht’s (1999) study, for example, higher levels of hunger were observed during periods of anger and joy than of fear and sadness; impulsive and sensory eating received higher scores during anger than other emotions and hedonic ratings were higher during joy than other emotions.

It is increasingly recognized that the eating context can influence an individual’s food and nutrient intake. Numerous studies have been undertaken illustrating an increase in food intakes when eating together compared with eating alone (Feunekes et al. 1995; Edwards and Hartwell 2004). This phenomenon of social facilitation is in evidence during all meals of the day, all days of the week and in many different locations.

The studyof food and emotions can be classified into two ways: those which focus on the effects of emotions on eating behaviour; and those which focus on the effects of eating behaviour on emotions (Desmet and Schifferstein 2008). This research covered the latter and specifically the change in emotion with macronutrient intake. Hence, the aim of this research is to critically evaluate the relationship between emotions and food consumption (macronutrient) of meals served in a realistic eating environment.

The research addresses the following hypotheses:

. H1: The emotional status of an individual influences what hot main meal (macronutrient content) is chosen.

. H2: The emotional status of an individual, when dining either alone or in the company of selfselected colleagues, is altered after eating a hot main meal (by macronutrient).

Methods

The research was undertaken using a dining facility (cafeteria) at Bournemouth University, England, run as a ‘food court’ which is open to students, members of staff and visitors. The counter used for the study offered a choice of two pizzas (one vegetarian), a pasta dish (the sauce changing daily), jacket potatoes with a choice of four toppings, three ‘traditional’ hot meals, including at least one vegetarian choice and salad items. These were defined as a hot meal for this study.

During term time, service was from 1200 to 1430h with the total number of customers on a ‘typical’ day ranging from 300 to 500.

Measuring emotional status

Just as there is no firm consensus on the definition of emotions, there is a lack of agreement as to exactly how emotions might be measured; hence the quest for suitable and appropriate instruments has a long and perhaps chequered history. A number of authors have investigated and proposed their own models which have involved questionnaires, verbal and non-verbal applications and ‘additional heart rate’ as an indicator of emotional arousal.

Most food-related studies to date have concentrated on developing instruments to assess emotions associated with a single food, primarily in a controlled laboratory environment, often using dieting and/or obese subjects or in Universities using psychology students. None has sought to measure emotions in a ‘live’ setting where complete meals are consumed. Of the numerous instruments which have been developed, each purporting to capture the emotions of individuals, a review of the literature failed to identify one that might be ‘entirely’ appropriate. Therefore, a comprehensive literature review provided the start point for this study. The work of Richins (1997) and King and Meiselman (2010) was considered as most suitable for the research context; however, in discussion with the co-researchers and potential subjects, it became apparent that, with the exception of one word (sexy), Richins used terminology ‘most’ appropriate to capture the data, although one word (bored) from King and Meiselman stood out as being a term that students could relate to and therefore was included. The concept of emotion is problematic to define; therefore, explicit well-validated items through which emotions have been investigated in previous research were used as the basis for data collection. The research instrument is presented in Table I.

There is no evidence of systematic order effects when administering this type of questionnaire (Richins 1997). However, in the studies reported, the questionnaires were only administered once as opposed to twice proposed for this study. It was decided, therefore, to reverse order the questions in the postquestionnaire, to help alleviate any ‘memory recall effects’. The resulting questionnaire presented in Table I was pre-tested on two groups (n ¼ 32) of first year University undergraduate students, providing a Cronbach’s avalue of 0.922.

Subjects and data collection

Subjects (n ¼ 408) were potential diners, recruited by asking them if they had time to take part and planned to eat a hot meal. Once confirmed, they were briefed as to the procedures and requirements of the study but

Table I. Pre-meal Emotional Status Questionnaire.

Notes: The above words describe different kinds of moods and feelings. Using the terms listed, please describe how you FEEL RIGHT NOW. Please rate each feeling by circling the appropriate number.

not the true purpose. Any questions or queries were addressed and, following agreement, subjects were invited to complete the Emotional Status Questionnaire for the first time. Once completed (approximately 5min), the time was noted on the questionnaire and a pseudonym added so that the subject’s completion of the subsequent questionnaire could be matched. Subjects then went to the service counter where they freely selected their hot meal before proceeding to the cash point to pay. They then found their own table in the cafeteria where they consumed their meal.

Once they had finished their meal, or all they wished to consume, they returned to one of the data collectors with their plate and completed the Emotional Status Questionnaire for the second time. Where any food remained uneaten, this was weighed out of sight and the amounts were recorded.

Data collection commenced at approximately 1150h and finished at 1400h daily. Prior to the start of each day’s meal service, details of the day’s menu were recorded along with average portion sizes. The latter were obtained by identifying and weighing a standard portion size and where necessary by purchasing a meal during service.

Data analysis

Data were coded, entered into the statistical package SPSS (version 18), checked and verified for accuracy. Descriptive techniques including ‘Frequencies’ were used to interrogate the data; thereafter Wilcoxon ranksum tests and Mann–Whitney U tests were used to compare ‘before’ and ‘after’ effects.

Food intake data were entered into a computer program Microdiet (version 2),1 which includes in its database standard food tables [McCance and Widdowson’s, The Composition of Foods, 6th Summary Edition], and the macronutrient content of meals was calculated. These data, discrete values (g) of protein, fat and carbohydrate, and percentages of energy from those nutrients, were also entered into the statistical package SPSS (version 18).

Spearman’s r was used to measure the magnitude and direction of the correlation between two variables, emotion and macronutrient. The analysis was extended by computing a new variable, ‘emotion difference’, pre- and post-meal, which was subjected to the same test. The data were then interrogated further by evaluating the effect of a macronutrient coded as a binary variable (e.g. low and high fat) on emotional difference using Mann–Whitney U test. A functional analysis approach was adopted.

Table II. The demographic characteristics of respondents (n¼ 408).

Results

The demographic profile of respondents is given in Table II.

For this part of the research respondents were considered as a homogenous group of diners, where when data are presented schematically, it can be seen that respondents generally demonstrate more positive emotions than negative (Figure 1). It can also be seen that emotions are flattened after a meal except for the emotion contentment which increases. Furthermore, after a meal there is a significant decrease in negative emotions, i.e. people are happier after a meal.

Spearman’s r correlations between all 47 emotion expression variables (pre- and post-consumption) and each macronutrient, fat, carbohydrate, protein and energy, identified weak correlations, these are presented in Table III.

Interestingly, testing emotion expression difference (pre- and post-consumption) variables and each macronutrient, fat, carbohydrate, protein and energy, again some weak correlations were found (Table IV).

It is not unsurprising that 63% of the respondent sample consumed dishes that contributed a high percent of energy from fat, higher than the USA and UK Dietary Reference Value (Department of Health 1991; US Dietary Guidelines 2010) of no more than 35% of energy recommended from this nutrient. Interrogating the data further revealed that the percentage of energy from fat was significantly higher for those respondents who showed a pre- and postdifference in the emotions, sentimental (p # 0.05) and discontented (p # 0.05). The intensity of feeling was greater for sentimental and less for discontented.

As fat and carbohydrate contribute nutrients in tandem, it is accepted that 86% of the participant sample consumed a low carbohydrate meal, lower than the UK Dietary Reference Value (Department of Health 1991) of no less than 50% of energy recommended from this nutrient. Interrogating the data further, it was revealed that the percentage of energy from carbohydrate was significantly lower for those respondents who showed a pre- and postdifference in the emotions such as unfulfilled (p # 0.05), humiliated (p # 0.02) and encouraged (p # 0.02). The lower the carbohydrate intake the more unfulfilled a participant felt, and paradoxically the higher the carbohydrate intake the greater the feelings of humiliated and encouraged were experienced.

Figure 1. Schematic presentation of pre- and post-meal subordinate level emotions.

Table III. Correlation between subordinate level of emotion expression and macronutrient.

*Carbohydrate; †Significantly different.

Within the population, protein intake is high and was demonstrated in this sample whereby 76% had intakes above the UK Dietary Reference Value of no more than 15% of energy recommended from this nutrient. There was one difference close to significance identified and this was for those respondents who showed a pre- and post-difference in the emotion discontented (p # 0.06). The higher the protein intake the greater the change in emotion pre- and postconsumption, that is the participants became less discontented.

Of the positive emotions, both males and females reported changes in their pre- to post-meal emotions for fulfilled, which increased, eager and proud, which decreased; and of the negative emotions, frustrated, irritated, worried and tense all decreased. Looking at the other expressions, hopeful, excited and loving for the positive affects; unfulfilled, bored and homesick for the negative affects, all decreased for females pre- to postmeal. Interestingly, there were no pre- to post-meal changes for the males.

In addition to the changes in pre- and post-meal, there were also a number of changes between the genders. Of the positive aspects, calm, peaceful, thrilled and proud, both pre- and post-meal; encouraged and eager, pre-meal; pleased, joyful, enthusiastic, relieved and passionate, post-meal, were all significantly different. The neutral aspect, surprised, was also significantly different in pre- and post-meal as were amazed, and astonished, for post-meal. In respect of the negative aspects, ashamed and humiliated in both pre- and postmeal; envious and homesick pre-meal; unfulfilled, bored, jealous and afraid, post-meal were also significantly different.

Notwithstanding the limitations of the small sample size of respondents eating alone, results show a large number of significant changes in positive and neutral emotions when eating alone or with others. Similarly, 19 of the negative emotions are reduced when eating with others compared with four when eating alone. However, when taking into account effect size, this number is reduced noticeably. Of the positive results, fulfilled increased positively pre- to post-meal when eating both alone and with others, whereas eager decreased again for both. Pleased increased and loving decreased when eating alone. Excited and proud both decreased when eating with others.

Of the negative results, worried and tense decreased both eating alone and with others, pre- to post-meal; Discontented and sad decreased when eating alone and frustrated, irritated, unfulfilled and bored decreased when eating with others. In addition to the changes in pre- and post-meal, there were also a number of changes depending on whether or not they were eating alone. Of the positive aspects, sentimental both pre- and post-meal; fulfilled, pre-meal; and hopeful, excited and enthusiastic post-meal were significantly different.

In respect of the negative aspects, nervous and lonely were significantly different both in pre- and post-meal; and bored, worried, tense and afraid were significantly different in pre-meal but not in post-meal.

In summary, results suggest the following:

. Rank order of emotions demonstrate that there is a greater propensity of positive over negative emotions.

. Emotions were altered by eating a hot main meal. In the main they tended to be flattened.

. After meal consumption, there is a significant decrease in negative emotions.

. Emotions, as measured by 47 words (expressions), had no subconscious influence on macronutrient selection.

. A high fat or high protein meal leads to a feeling of contentment.

. Participants became more unfulfilled after a low carbohydrate meal.

. The variables, gender and eating alone or with others, were considered to be important, pre- to post-meal. Males appear to be more positively disposed than females; the negative emotions, less clear by gender. Eating alone heightens positive and negative emotions.

During the data collection, plates were ‘inspected’ and in a very small number of cases, the amount of food remaining was weighed. However, the number of diners leaving food was not sufficient to provide for any meaningful analysis.

Discussion

The relationship between emotions and food consumption has been the subject of research for a number of years, although much of this interest has focused on those with ‘eating disorders’. Indeed, theories on the relationship between emotions and eating behaviours have their origin in the literature on obesity (Canetti et al. 2002). Notwithstanding, the role of emotions is important in the study of eating, eating out and the foodservice industry. Both positive and negative emotions, along with perceived service quality, are the three most important dimensions in customer satisfaction (Ladhari et al. 2008). Positive emotions are thought to exert a stronger impact than negative emotions although this contradicts the work of Liljander and Strandvik (1997). The authors suggest that this apparent contradiction might be due in part to thefactthatdifferentelementsoftheservicesectorwere studied, although the actual reasons remain unclear.

When considering emotion expression differences, it has been suggested that these variables can be co-varied from macronutrient intake (Stroebele and de Castro 2006). Many studies, particularly in the obesity field, have demonstrated that this is an area where short-term emotional response tends to overpower longer term more rational thinking (Cabinet Office 2010). Emotionally driven food behaviour takes over and can ‘trigger’ spontaneous consumption. There has much been written in the literature about ‘comfort’ food (Dube´ et al. 2005) and that consumption can be triggered by negative emotion, which in turn produces ‘guilt’. Anecdotal comments from participants in the study support this state particularly as the menu provided items such as pizza and pasta in their minds could evidently be categorized as ‘junk’ food.

Nutrients in food can activate certain amino acids that, in turn, produce or decrease the production of neurotransmitters such as dopamine, epinephrine and serotonin that regulate emotions (Thayer 2003). Certainly this research has demonstrated a greater feeling of contentment with a high fat, high energy meal which no doubt in part can be attributed to a feeling of satiety; whereas with a low carbohydrate meal, participants felt unfulfilled. These results are not unsurprising and reflect Benton and Brock’s (2010) assertion that a high carbohydrate meal leads to feelings of being happy and relaxed. In addition, a high protein meal also leads to a feeling of contentment which confirms research by Gibson (2006), linking an increase in positive emotions with a protein-rich lunch. Notwithstanding, although eating a meal will alter emotional predisposition, the flow of direction will depend on expectations and needs, which are personal to each individual.

From this research, two emotions (lonely and sentimental) are very weakly correlated with protein and carbohydrate/energy intake, respectively. In addition, it could be concluded that participants became less discontented after a high fat, high protein meal and more unfulfilled after a low carbohydrate meal. Unfortunately, this is counter to any nutritional guidance towards a healthy eating regime.

Gender variations were similar to the overall pattern of changes in emotions. Males appear to be more positively disposed than females and at a subordinate level Peacefulness (calm, peaceful), Joy (pleased, joyful), Surprise (surprised, amazed, astonished) and Excitement (thrilled) differed from male to female, post-meal consumption. The negative emotions are less clear by gender.

Dube´ et al. (2005) reported that for comfort foods at least men reported more intense positive affects prior to consumption, whereas women reported more negative affects. This is certainly the case from the current study where all significant values are higher for men pre-meal and also post-meal. It would appear, therefore, that overall gender is an important factor for some of the emotional (positive) expressions.

Eating tends to be a social activity although in many instances, such as in hospital, a cafeteria and soldiers in the field, the diner may be surrounded by others, but actually they may eat alone. A number of studies have shown the effects of eating with others. For example, energy intakes are higher, while satiety levels are lower (de Castro and de Castro 1989); the speed of consumption is increased (Rosenthal and McSweeney 1979); and the larger the group size the larger the meal (de Castro and Brewer 1992). This is seen across a range of meals (de Castro 1990) and in different settings, including hospitals (Edwards and Hartwell 2004). Mixed gender groups consume more than single sex groups (Klesges et al. 1984).

A number of suggestions have been offered to account for this phenomenon, although the most plausible seems to suggest that much is due to the time diners actually spend at the table, which increases when people eat together (Bell and Pliner 2003).

Results of the research show that when eating alone or with others, the patterns follow the overall rankings. Most of the changes, both positive and negative, occurred among those who dined with others, but did not meet the criteria of r $ 0.2. What is noticeable is that when eating alone, positive scores tend to be higher; and all significant scores are higher. The meal seems to be more important in raising emotional stateus when eating alone. Those who dined alone, of the negative emotions, nervous, worried and tense were much higher pre-meal. They seem to have been more afraid and concerned about being alone. The meal, as might be expected, reduced their feelings of being alone but they were still higher post-meal than those eating with others.

The consumption of food by people who eat in a group has never been explained by a change in a person’s emotion about the food (Barthomeuf et al. 2009). These authors investigated what effects photographs of facial expression (pleasure, neutral and disgust) might have on the desire to eat. The results show that for foods which are liked, the desire to eat was higher when foods were presented with no photograph, rather than with a photograph of a person expressing neutral emotions. When the photograph of person expressed pleasure, the desire to eat did not significantly increase. However, when the photograph of a person expressed disgust, the desire to eat decreased significantly. Many people will have positive and negative emotions, which can in part be reduced by consuming a meal. A greater change is brought about when consuming the meal with others. Although it could be argued that some of the relationships are rather tenuous, the overall thrust from the research is that dining with others has more of an influence than when dining alone. The second hypothesis:

H2: The emotional status of an individual, when dining either alone or in the company of selfselected colleagues, is altered after eating a hot main meal, is therefore supported.

It has been suggested that mood and emotion could influence food choice via physiological affects that change appetite. On the other hand, alteration of mood may be an outcome of food choice (Gibson

2006). Emotion, with the exception of one expression, sentimental, had no influence on what was chosen. The first hypothesis:

H1: The emotional status of an individual influences what hot main meal (macronutrient content) is chosen remains inconclusive.

Analysis of post-meal emotions identified five expressions: three positive, eager, optimistic and loving; one negative, angry; and one neutral amazed that were affected after the meal had been consumed. It is felt, however, that five emotions out of 47 can only indicate a weak association between the choice and consumption of a particular meal and changes in an emotion.

Summary and conclusions

In a foodservice eating environment, attention is generally paid to aspects, such as the food, ambiance and de´cor, all of which contribute to the overall experience and customer satisfaction. Little attention has been paid to emotions and the effects they have on the overall foodservice experience. Nevertheless, despite the lack of a consensus on what is understood by emotions and an agreed methodology to measure them, this research has, for the first time, tried to address some of the issues.

Using a previously validated instrument, changes in the emotional status of diners (n ¼ 408) consuming a self-selected hot main meal in a cafeteria (refectory) were investigated. Results demonstrate a greater feeling of contentment with a high fat, high energy meal; whereas with a low carbohydrate meal, participants felt unfulfilled. In addition, a high protein meal also led to a feeling of contentment. Practical implication of these results from a foodservice aspect is rather counter-intuitive to any public health nutritional advice, however, indicates the requirement for a range of dishes to be available for consumers to accommodate their emotional status, and in particular the inclusion of a protein or high carbohydrate item. It also indicates that from a physical setting encouraging people to eat together could lead to greater emotional stability.

The uniqueness of this research is that distinct emotions have been measured in a real-life situation rather than as a general valence response, adding valuable data and insights into the body of knowledge on emotional response to an eating situation.

Further study is clearly required in an area that is of emerging interest, for example, participants could be randomly allocated to one of the conditions with positive, negative or neutral emotions induced. A meal is then chosen and impact on meal choice could be measured. Strategies for encouraging healthy eating habits could be incorporated at this point. The moment-to-moment experience of the meal could be measured by videoing participant’s facial expressions and analysing them offline (perhaps using Ekman’s scheme). This could be allied to physiological emotion measures, some of which can be measured quite unobtrusively, e.g. heart rate.

Limitations

This research used undergraduate students as subjects who were probably ‘refuelling’ rather than sitting down to ‘appreciate’ their meal. Furthermore, in some cases, choice of meal may well have been predicated by financial considerations. In addition, there are individual differences in susceptibility to nutritional effects on emotion, which this study in design did not address. Notwithstanding, it is considered that these results are sufficiently robust to be able to draw conclusions, although the results may not necessarily be generalizable, indicating the need for further research.

Declaration of interest: This work was supported by funding from the US Army under contract number W911NF-10-1-0331. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Ethical considerations

Ethical approval was sought and granted through standard University procedures. Appropriate health and safety considerations, together with a risk assessment protocol, were carried out prior to the commencement of the primary research. Respondents’ agreement was assumed by the completion of questionnaires and participation in the interviews. Confidentiality and anonymity were assured at the start of the interviews.

Notes

1

A computer program (Salford University).

References

Barthomeuf L, Rousset S, Droit-Volet S. 2009. Emotion and food. Do the emotions expressed on other people’s faces affect the desire to eat liked and disliked food products? Appetite 52(1): 27–33.

Bell R, Pliner PL. 2003. Time to eat: the relationship between the number of people eating and meal duration in three lunch settings. Appetite 41:215–218.

Benton D, Brock H. 2010. Mood and the macro-nutrient composition of breakfast and the mid-day meal. Appetite 55:

436–440.

Cabanac M. 2002. What is emotion? Behav Process 60:69–83.

Cabinet Office. 2010. Applying behavioural insight to health. Available from: http://www.cabinetoffice.gov.uk/resourcelibrary/applying-behavioural-insight-health. Last accessed 10 February 2011.

Canetti L, Bachar E, Berry EM. 2002. Food and emotion. Behav Process 60(2):157–164.

de Castro JM. 1990. Social facilitation of duration and size but not rate of spontaneous meal intake in humans. Physiol Behav 47(6):

1129–1135.

de Castro JM, Brewer EM. 1992. The amount eaten in meals by humans is a power function of the number or people present. Physiol Behav 51(1):121–125.

de Castro JM, de Castro ES. 1989. Spontaneous meal patterns of humans influence of the presence of other people. Am J Clin Nutr 50:237–247.

Department of Health. 1991. Dietary reference values of food energy and nutrients for the United Kingdom (Report on Health & Social Subjects). London: Department of Health.

Desmet PMA, Schifferstein HNJ. 2008. Sources of positive and negative emotions in food experience. Appetite 50(2–3): 290–301.

Dictionary of Psychology in Politics and Social Sciences. 2010., Available from: http://www.oxfordreference.com/views/ SEARCH_RESULTS.html?q¼emotion&ssid¼159650658& scope¼global&time¼0.551431181011136. Last accessed 31 January 2011.

Dube´ L, LeBel JL, Lu J. 2005. Affect asymmetry and comfort food consumption. Physiol Behav 86(4):559–567.

Edwards JSA, Hartwell HJ. 2004. A comparison of energy intake between eating positions in a NHS hospital – a pilot study. Appetite 43:323–325.

Feunekes GIJ, De Graaf C, Van Staveren WA. 1995. Social facilitation of food intake is mediated by meal duration. Physiol Behav 58:551–558.

Gibson EL. 2006. Emotional influences on food choice: sensory, physiological and psychological pathways. Physiol Behav 89(1): 53–61.

King SC, Meiselman HL. 2010. Development of a method to measure consumer emotions associated with foods. Food Qual Prefer 21(2):168–177.

Klesges RC, Bartsch D, Norwood JD, Kautzman D, Haugrud S. 1984. The effects of selected social and environmental variables in the eating behavior of adults in the natural environment. Int J Eat Disord 3(4):35–41.

Ladhari R, Brun I, Morales M. 2008. Determinants of dining satisfaction and post-dining behavioral intentions. Int J Hosp Manag 27(4):563–573.

Liljander V, Strandvik T. 1997. Emotions in service satisfaction. Int J Serv Ind Manag 8(2):148–169.

Macht M. 1999. Characteristics of eating in anger, fear, sadness and joy. Appetite 33:129–139.

Nolan LJ, Halperin LB, Geliebter A. 2010. Emotional appetite questionnaire. Construct validity and relationship with BMI. Appetite 54(2):314–319.

Plutchik R. 1980. Emotion: a psychoevolutionary synthesis. New York: Harper & Row.

Richins ML. 1997. Measuring emotions in the consumption experience. J Consum Res 24(2):127–146. Rolls ET. 2007. Emotion explained. Oxford: Oxford University

Press.

Rosenthal B, McSweeney FK. 1979. Modeling influences on eating behavior. Addict Behav 4:205–214.

Stroebele N, de Castro JM. 2006. Influence of physiological and subjective arousal on food intake in humans. Nutrition 22: 996–1004.

Thayer R. 2003. Calm energy: how people regulate mood with food and exercise. Oxford: Oxford University Press.

US Dietary Guidelines. 2010., Available from: http://health.gov/ dietaryguidelines/dga2010/DietaryGuidelines2010.pdf

Copyright of International Journal of Food Sciences & Nutrition is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.