Recipient Organization
UNIVERSITY OF TENNESSEE
2621 MORGAN CIR
KNOXVILLE,TN 37996-4540
Performing Department
Food Science & Technology
Non Technical Summary
Food texture is one of the two most important food attributes for acceptability. However, we do not understand much about what makes a food texture desirable or aversive. This project will first define and quantitate the prevalence of texture aversions in the United States. Texture aversions are related to picky-eating, a behavior that has been shown to have negative health outcomes. We will also see to understand the role of expectation on texture acceptance. Previous research has shown that deviations from expected sensation can be key to understanding food/beverage acceptability. Lastly, food texture is not typically perceived in a vacuum. More specifically, food texture is often perceived with accompanying tastes, aromas, sights, and sounds. The last goal of this project is to see how associating tastes with a particular texture can modify that texture's acceptability.
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
0%
Goals / Objectives
1. Define the characteristics of texture aversion - Define the frequency that individuals reject food due to sensory attributes while ascertaining the types of foods commonly rejected for its texture. Address any relationships between subjective touch sensitivity and food texture aversions.2. Better understand the role of expectations in texture acceptance - Determine the physiological and psychological changes to adapting texture expectation and its role in the acceptance of a food.3. Determine the role of taste in conditioning texture preferences - To better understand the development of food aversion across different sensory modalities through conditioning an acceptance or rejection of a neutral texture with basic tastes in two age groups (school-aged and young-adult).
Project Methods
Goal 1. Part 1: A survey will be distributed to individuals (n = 500) who are native English speakers, born in and currently reside in the U.S.A, and report no dietary restrictions. The survey will be designed to determine the frequency that individuals reject food due to sensory attributes while ascertaining the types of foods commonly rejected for its texture. First, motivations behind food rejections will be explored. Three motivations, set by Rozin and Fallon (1987), will be provided (and defined) and then participants will be asked to rate those for their involvement in the rejection of food: 1) Unpleasantness of the sensory attributes (e.g. appearance, sound, texture/mouthfeel, taste/flavor and aroma), 2) Anticipation of consequences (or harm) after eating, either short term (e.g. upset stomach) or long term (e.g. cancer), and 3) Other factors such as consuming nontraditional (e.g. insects) or religiously-restricted (e.g. pork for Muslims) foods (Rozin & Fallon, 1987). Next, participants will be asked to list all foods that they reject to eat due to its sensory attributes. On the same page, and prior to the question, a list 5 sensory attributes and definitions will be given: 1) Aroma: the smell of the food before you eat it (e.g. smell of baked bread), 2) Texture: the way a food feels in your hands or mouth (e.g. rough, crispy), 3) Flavor/taste: the distinctive taste of the food (e.g. sweet, fruity), 4) Sound: the noise of the food (e.g. sound of sizzling meat), and 5) Appearance: the way the food looks (e.g. color, shape). On the following page all foods mentioned will be listed and individuals will be asked to check-all-that-apply for which sensory attribute(s) make them not eat this food. A listing of all sensory attributes defined earlier were available. Lastly, the next page will repopulate foods that were checked for 'texture' as a sensory attribute contributing to the rejection of the food. For each of these foods, participants will be asked to describe the texture(s) that were unpleasant in this food. After all food-related questions are answered, the participants will fill out the touch sensitivity subscale of the Sensory Perception Quotient (SPQ) which was developed for adults with and without autism (Tavassoli et al., 2014). This scale has 9 statements (e.g. I notice the weight and pressure of a hat on my head) which will be rated on a six-point scale (Strongly Disagree, 1; Strongly Agree, 6).Statistical Analysis A linear mixed effects model will be used to understand motivation differences for individuals with higher touch sensitivities. In this model, the motivations will be used as the within variables and composite touch sensitivity score as continuous variable for each participant. The interaction between these terms will also be included in the model. A mixed effects logistic model will be built. In this model, foods will be nested within subjects and each food with a binary rejection outcome for a particular sensory attribute (e.g. rejected for texture and flavor, but not sound, appearance, and aroma). Additionally, the touch sensitivity of the subject will be included as well as its cross-level interaction with the sensory attribute. However, for textures, anecdotally, a more common response of food aversion is "there is 'a' food I won't eat due to its texture." This tests a different notion than the previous question, it asks "How many people reject a food due to a sensory attribute?" To answer this question, foods will be collapsed within individuals and each sensory attribute will be made binary (mentioned or not mentioned as a reason to reject one or more of the foods listed). This resulted in different frequencies than the previous question. The texture-specific section of the survey will be analyzed to understand how people use aversive texture terms in relation to foods. Due to the diversity of what can constitute a texture attribute, texture terms are often indexed into classes which are related to different aspects of texture. For example, mechanical attributes are a texture class concerned with reactions to stress (e.g. hardness, brittleness) while the geometrical and surface attributes are related to food size/shape/arrangement (e.g. porous, gritty) or moisture level/fat content of the food (e.g. moist, greasy), respectively (Guinard & Mazzucchelli, 1996; Szczeniak, 1963). All texture quality terms were categorized into an established classification system for texture types (Guinard & Mazzucchelli, 1996; Szczeniak, 1963) (Table 1). Next, cluster analysis techniques will be used to classify individuals into groups of similar disliked textures. Foods were collapsed within individuals and the texture term types were summed. The texture terms will be analyzed using principle component analysis (PCA) without a weighted outcome of sensitivity. Texture term types will be reduced into seven principal components (PCs) using PCA. Non-parametric Kruskal-Wallis rank tests were done for each texture type from the classification (Table 1), with the clustered groups as the main factor and post-hoc comparisons among groups were done using the Wilcoxon method.Goal 2. Condition individuals (n = 70) to expect a "hard" or "soft" texture over three sessions using a series of triangle tests. At the beginning of each session, they will try one gummy of their condition ("hard" or "soft") and rate on several attributes similar to a regular sensory affective test. On the forth session, they will be given either a "hard" gummy and rate with the same affective test ballot as prior sessions. During all sessions, a camera will record and capture their jaw movements during mastication.T-test will be used to look at rating and jaw movement differences between the "hard" and "soft" groups during the fourth session. A MANOVA will be used to look at changes in jaw movements and preferences from sessions 1 to 3 for hard and soft conditions.Upon completion of this study we will have developed a better understanding of the role of texture expectations in the acceptance of a food.Goal 3. Increasing oral transit times (i.e., chewing) decreases consumption and this can be modified by texture. Additionally, the amount of saliva and its composition of enzymes modifies food texture and may be modified through exposure. We will test the hypothesis that you can condition a person to accept or reject a food texture through inherent basic tastes leading to healthy eating. After selecting neutral textures (varying in transit times), a classical conditioning paradigm will be performed pairing a conditioned stimulus (CS+) with an inherent taste (the unconditioned stimulus, US) to form a conditioned response (CR, either positive or negative). Sweet and bitter tastes will be used for the US as they respectively have a positive and negative unconditioned response (UR). More specifically, the positive conditioned response will be made on a longer transit time texture (e.g., chewy) and negative conditioned response on a shorter transit time texture (e.g. soft). Mastication parameters (via camera) and saliva samples will be collected on the CS+ and CS- pre and post-conditioning as well as affective and preference measurements.Depending on time and IRB stipulations with underage participants, adolescents may be used as a second study sample. Children which have not learned to like or dislike bitter and sweet respectively may show greater responses to conditioning than adults. Consequently, a longitudinal study with several follow-ups may be more appropriate to see preference effects into adulthood.