Dr Herb Mieselman and the members of his team have been conducting research into the relationship between emotion and food. While there has been research done into a circular relationship between emotion and eating (see figure 1), the research Dr Mieselman and his research group have been involved with has focussed on consumer behaviour and the related emotions. This research has been largely devoted to determining what emotions can be elicited via eating, or even by being presented the name of a specific food.
In 2012 Cardello et al published the article ‘Measuring emotional responses to foods and food names using questionnaires’ in the journal Food Quality and Preference. This article investigated the relationship between a variety of emotions and foods, and also aimed to determine the reliability of the questionnaires that have been employed. While there are a number of validated questionnaires focussing on emotion, these are largely from the field of clinical psychology and have limited applicability to the consumer research field. This is generally because clinical psychology focusses on negative emotions, whereas consumer food research has a comparatively positive emotional range. As such, consumer research requires significantly more detail regarding positive emotion than is generally included in clinical questionnaires. Further, as this is an emerging field, there are limited consumer specific methodologies that have been adequately validated. As a result, part of this paper focusses on the reliability of the questionnaire used – the EsSense Profile, which was originally developed by King and Mieselman (2010).
Emotional responses to foods and food names
Participants were given 12 food names on separate pieces of paper and asked to rate each emotion on the EsSense Profile after looking at each name. There are 39 emotions addressed by this questionnaire, including joyful, interested, bored, aggressive, eager and good natured. A separate group of participants were then given four samples of different foods, each one with two varieties. For example, they were given potato chips as one of food variety, and were given both plain and barbeque flavours. Liking was measured in addition to emotion.
Significant differences in emotions were found in both study one and two after viewing food names and tasting different foods. However, there were no significant differences in emotional responses to the different brands/flavours of food that were tasted. Additionally, higher intensity emotions were recorded after participants were shown the names of foods rather than after tasting them. The authors postulate that this may be because an individual’s memory of a specific food is generally the best version of that food they have experienced, whereas consuming a food product in a laboratory may not be as positive an experience. A positive correlation between liking and emotional responses was also found when food products were tasted.
Reliability of the EsSense Profile
Four food names were given to participants, and their emotional responses were recorded, as in study one. These four names were chocolate, oatmeal, potato chips and carrots, and were chosen as a result of their variability in emotional responses according to the previous two studies. Participants completed the same emotion questionnaire as in the previous studies after reading the food names. Participants were retested one week later to determine reliability of responses
A separate group of participants were given four foods to tasted (chocolate, plain potato chips, coca cola and shortbread), with emotional responses measured after each. Participants were retested one week later.
Some emotions were highly correlated, while some weren’t for both study three and four. For example, words such as guilty, polite, tender, whole and satisfied were highly correlated for food names, while words such as joyful, eager, daring, aggressive and good natured were most correlated for tasted foods. However, the correlation range was between 0.309 to 0.767, with some emotions being less correlated for both names of foods and tasted foods.
Additionally, the 39 emotions that were investigated were correlated to investigate any similarities. This resulted in six groupings of similar emotions;
- Glad, good, pleased, happy, satisfied, enthusiastic, joyful, interested, pleasant, eager, energetic, good natured, calm, friendly, merry
- Quiet, understanding, tame, steady, tender, polite, warm, secure, whole, peaceful, mild
- Adventurous, wild, daring, aggressive, active
- Loving, affectionate, merry
- Disgusted, guilty, worried
From this study we can determine that there is indeed a relationship between emotions and food at a consumer level, and that most of the resulting emotions are positive. However, as this field is currently emerging and has limited established methodologies, there are potential validity issues with the questionnaires that are used. Particularly, there is no indication that the EsSense questionnaire is measuring the specific emotions it is designed to. As such, as this field develops further, it will be important to conduct further examinations of the specific tools being used. Additionally, when considering the consumer applications of research such as this, it may be relevant to create more sensitive measures, as the current questionnaires cannot determine any difference between brands or flavours (e.g. plain vs barbeque chips, or varieties of chocolate), which is particularly important for companies looking to succeed in the competitive food marketplace. Further, it may be beneficial in further investigations to include pre-test emotion questionnaires, to get a fuller understanding of the change in emotion following food presentation.
Cardello, A., Meiselman, H., Schutz, H., Craig, C., Given, Z., Lesher, L., Eicher, S. (2012). Measuring emotional response to foods and food names using questionnaires. Food Quality and Preference, 24, 243-250.
King, S., & Meiselman, H. L. (2010). Development of a method to measure consumer emotions associated with foods. Food Quality and Preference, 21, 168–177.