This item is only available as the following downloads:
FOOD FOR THOUGHT: THE EFFECTS OF MENU DESCRIPTIONS AND PLATING AESTHETICS ON CONSUMER RATINGS AND EXPECTATIONS BY NICOLE NOUJAIM A Thesis Submitted to the Division of Social Sciences New College of Florida in partial fulfillment of the req uirements for the degree Bachelor of Arts Under the Sponsorship of Gordon Bauer, PhD Sarasota, Florida April, 2013
ii Dedication For my sister Ashley It's us against the world.
iii Acknowledgements First and foremost, I want to thank my sponsor Professor Gordon Bauer who, even without any specific ties to my topic beyond vague connections to perception, took on an over zealous third year and allowed me to pursue a topic that was of great interest to me and relevant to my future career goals. I'd also like to thank Professor Michelle Barton for not only being a member of my committee, but also my stand in sponsor while Professor Bauer was on leave. Her dedication and rigor as a sponsor is one of the only reasons I was so productive during fall semester. To Professor Steven Graham, thank you for being a member of my committee and a recurring pillar of support throughout my years as a psychology student. Additional acknowledgements also go out to Professor Heidi Harley. Beyond her contributions as a professor, hearing her laugh echo through Bon House was, on numerous occasions, the only thing that kept me sane after hours of staring at the same four walls. Thank you to everybody that participated in my pilot and thesis studies, namely those who were last minut e replacements when people didn't show up for their appointments. If it weren't for you all, I wouldn't have a thesis. I also want to thank all those who have accompanied me on this grand adventure. To Mia Newell and Michelle Patteson, you've been there fo r me since day one and I love you both so much. Without you two I would've never made it past first year, much less through the next three. To Jennie Caskey, I may know more about dolphins and camp life than I ever wanted (or needed) to, but your friendshi p and support means more to me than you will ever know. You were there for the breakdowns, the sleep deprived giggle fits, the late nights in Bon, and everything in between. To Adam Flowerday and Kyle
iv Saunders, thank you for putting up with living with me, for all your help with my pilot study shenanigans, and for making my life more bearable in general. To Nick Daugherty, thank you for being the constant outlet for my frustrations this year. Also thank you to all my other friends for your continued support and entertainment I don't know what I would've done without you all. Finally, I want to thank Elizabeth Ashley Noujaim, my sister, my other half, my constant support system, my partner in crime. You are always there for me no matter what I need and I lo ve you so much. You are what family should be and regardless of the hands we've been dealt we can get through it together. It's you and me against the world sissy, and that will never change.
v Table of contents DEDICATION ii ACKNOWLEDGEMENTS iii TABLE OF CONTENTS v ABSTRACT vi INTRODUCTION 1 Factors That Affect Food Consumption and Liking 2 Sensory Expectations and their Influence on Perception 7 Menu Descriptions and Labeling 10 Aesthetics and Plating 12 THE CURRENT STUDY 15 METHOD 1 6 RESULTS 20 DISCUSSION 25 FIGURE 1 32 FIGURE 2 33 APPENDIX A 34 APPENDIX B 35 APPENDIX C 36 REFERENCES 40
vi FOOD FOR THOUGHT: THE EFFECTS OF MENU DESCRIPTIONS AND PLATING AESTHETICS ON CONSUMER RATINGS AND EXPECTATIONS Nicole Noujaim New Colle ge of Florida, 2013 ABSTRACT Many factors affect consumer food preferences and ratings, including how the food item is described and presented pre consumption. Previous research has also demonstrated that perceptions of liking are created before a foo d item is even tasted, and these expectations can change over the course of a meal. The current study examined how exposure to different menu descriptions (descriptive vs. plain) and different plating aesthetics (refined vs. crude) influenced consumer rati ngs and satisfaction. College students were given questionnaires to fill out after menu exposure, after plating exposure, and then post consumption. Results confirmed previous findings in that the descriptive menu was better liked, receiving higher ratings than the plain menu, though only significantly so for two items. No significant results were found concerning plating aesthetics or any interaction between the plating and menu conditions, however interesting trends in the data are mentioned. Implications and suggestions for future research are examined. __________________________ Gordon Bauer Division of Social Sciences
1 FOOD FOR THOUGHT: THE EFFECTS OF MENU DESCRIPTIONS AND PLATING AESTHETICS ON CONSUMER RATINGS AND EXPECTATIONS "One of the very nicest things about life is the way we must regularly stop whatever it is we are doing and devote our attention to eating." Luciano Pavarotti Food is a concept that spans every city, every nation, and every culture; it is something every person experiences. How each individual experiences food, however, is drastically different. Even when the scope is limited to western cultu re, there are still many distinctive aspects of food that come into play concerning an individual's eating habits and preferences. While taste does play a key role in an eating experience, it is not the only factor with influence. Many other aspects (e.g., packaging, presentation, color, texture, and descriptions) play a large role in the enjoyment of food related experiences, and the perceptions people create about specific food items (Lyman, 1989; Oliver, Wardle, & Gibson, 2000; Piqueras Fiszman, Alcaide Roura, & Spence, 2012; Stroebele & Castro, 2004). Preconceived expectations concerning these influential aspects also affect consumer satisfaction and overall hedonic evaluation (Lyman, 1989; Wansink & Park, 2002). Many improvements to meal experiences c an occur by gaining a better understanding of the roles that these factors play and then manipulating them accordingly. Specifically, professional eating establishments can alter their restaurants appropriately and potentially benefit from increased sales, repatronage, and customer satisfaction ( Gueguen & Jacob, 2010; Wansink, van Ittersum, & Painter, 2002, 2005) However, these alterations ought be monetarily practical to be successfully applied in a
2 real world setting. Though many factors have been identi fied as influential during an eating experience, few inquiries into how these factors interact with each other have surfaced. The following is a review of the current literature concerning different aspects of eating experiences that affect individual atti tudes and food preferences. Focus is drawn to the effects of food item descriptions and plating presentation as well as the role of sensory expectations and their influence on the perception of food. After this review, the current study investigates the re lationship between menu descriptions and presentation aesthetics, as well as how this relationship affects liking towards an eating experience. Factors That Affect Food Consumption and Liking Many assume that the basis for food preference falls to taste, smell, and texture. This, however, is not entirely true and Lyman (1989) dedicated his book A Psychology of Food: More than a Matter of Taste to dispelling this myth. Food preferences can vary from day to day. If a food is initially disliked, then it will continue to be seen as unfavorable, but preferences for liked foods are not this stable and tend to fluctuate (Lyman, 1989). Context is one of the most important factors when considering preference fluctuations. Context for any item in a meal is all other food related components present, from other food items on the plate to the service and atmosphere. Lyman (1989) explains that whenever food is taken out of its normal context, meaning may be changed and preference may be altered. Physical and social setti ngs can arouse preferences for particular foods based on associations from the past (i.e. the smell of grandma's cookies), which can cause specific preferences or even cravings for foods, even though these items
3 were not previously desired (Lyman, 1989). F or this reason, the generalizability of certain food lab testing situations may not always be entirely accurate in predicting real world consumer behaviors. There is much evidence that confirms the effect that these factors have on consumer preferences a nd how food is perceived. Stroebele and Castro (2004) generated a review of the different facets of ambience that play a role in choices concerning food. This review opens with the bold assertion that sometimes the ambience of a location can have a greater influence on the consumer's perception than the actual food item will. Ambience is very much all encompassing calling upon visual, auditory, olfactory, and textual cues in the immediate environment. While societal factors, such as media, cultural norms, and advertising, may also influence intake and food choice, the consumer is mainly exposed to these well before consumption. The researchers also explain how some cues hinge on the border between societal norms and ambiance. For example, social facilitati on is a phenomenon in which the number of people present for a meal is positively correlated with the quantity of food and calories consumed. Another such cue is social modeling; this theory posits that whom you are with affects your eating. For example, i ndividual eating habits may change in the presence of a parent or potential mate. Stroebele and Castro (2004) continue on in this review to mention more straightforward ambient factors like the presentation of food cues (there is a greater desire to eat a food after being visually exposed to it), the accessibility of food (greater availability leads to greater intake), and the presentation of portions (physiological signs of fullness might be ignored when larger portions offer the possibility to consume mo re
4 of the desired food). Other, subtler cues like color, smell and lighting can also play a role in consumer intake and preference. Color, whether ambient or tied to the food, can unconsciously influence mood, sensation, appetite, hunger, and attractivenes s. Light can either energize or relax a person's movements and it is suggested that exposure to dimmer light promotes disinhibited eating. This effect is one of the reasons self conscious eaters tend to consume more at night. Aroma is also a very interesti ng ambient factor, considering olfaction is directly connected to the perception of flavor. It has been shown that aroma can influence expectations of food quality, food choice, and amount of intake. Weber, King, and Meiselman (2004) have conducted more sp ecific work regarding ambience. Four tests concerning the effects of social interaction, physical environment and food choice freedom were run with 406 participants. Four central location tests of about 100 people each were carried out. All participants w ere given the same meal (personal sized pizza, side salad, and iced tea), while social atmosphere, environment, and choice options were manipulated. In the first test, participants were seated in a food laboratory, given the meal, told to face the wall and not to discuss anything with other participants. In the second test, participants were seated around tables in a mock caf with friends and co workers in groups of two to six people with whom discussion was allowed. Up until this point only paper and soft plastic utensils/dinnerware were used. The third test condition introduced real silverware and hard plastic plates into the mix. The final test introduced all of the aforementioned characteristics as well as food item choice: participants were allowed to choose their type of pizza, salad dressing, and iced tea. Questionnaire surveys were administered to all
5 participants during testing. Results suggest that consumption was always lowest for the meal alone condition (test 1) where no contextual enhancements were present. Consumption of pizza and tea were increased when the environment was more like a restaurant (test 3) with no further increase seen when a choice was given (test 4). However, given a possible ceiling effect (considering 50% of participants ate all of their food), these results may not be appropriately generalizable. Overall these results do suggest, however, that physical environment is critical when considering food consumption and preference. Meiselman, Johnson, Reeve, and Crouch (2000) have also demonstrated the influence of environment on food acceptance. These researchers predicted that food served in different settings would produce reliably different ratings for the same items. A meal was offered to participants in three different meal te sting environments: a training restaurant, a food science lab, and a cafeteria. After participants ate their meal they were asked to answer a questionnaire concerning hedonic ratings. Participants scored the food highest in the training restaurant, followe d by the food science lab, with the cafeteria setting receiving the lowest overall ratings. These results reiterate the concept that when the same food is served in different environments, acceptance of the food can vary according to setting. Individual co nsumer differences beyond ambience also affect food preferences. Stress, for instance, plays a role in the types of food individuals prefer. In a laboratory study, Oliver, Wardle, and Gibson (2000) demonstrated how acute stress alters food choices in healt hy adults. Participants were assigned to either a stressed condition or a control condition. Participants in the stressed condition were asked to prepare a speech
6 with the understanding that it would be filmed and assessed after a midday meal (though speec hes were not actually presented), while participants in the control condition instead listened to a passage of neutral text before the meal. Mood, hunger, heart rate and other measures were assessed at baseline and after the speech/passage task. Participan ts were then given an array of foods to choose from for a midday meal they were told was compensation for study participation. Options included sweet, salty, and bland foods, as well as both high and low fat choices. Researchers recorded food intake and re sults revealed a difference between conditions. Stressed, emotional eaters more often preferred sweeter and high fat foods as well as choosing a more energy dense meal compared to their unstressed, non emotional counterparts. The results of this study show how individual consumer variables external to the meal itself (like a stressed state) can also affect food preferences. There are many other individual consumer differences that can also affect the hedonic perception of food items. Genetics can leave indi viduals predisposed to different levels of sensitivity to bitter taste. Studies of age show that younger people prefer familiar, sweet, energy dense foods while elderly adults prefer strong salty and sweet tastes, most likely due a decrease in taste sensit ivity. Even a consumer's nicotine intake prior to eating has shown to reduce the intensity of fat taste (Drewnowski, 1997; Perkins, Epstein, Stiller, Fernstrom, Sexton, & Jacob, 1990). These individual factors, however, are out of the service industries' hands. The owner of a professional eating establishment cannot control these aspects of an eating experience. While ambiance can be manipulated, a total change of setting is not always the most easily attainable goal. This process can be long and expensive calling upon
7 many different factors to be kept in mind. Considering the goal of this research is aimed towards identifying practical and easily applied changes, the remainder of this literature review focuses on how sensory expectations can affect consum er perceptions as well as how menu descriptions and plating aesthetics (two easily manipulated aspects of a dining experience) play into this. Sensory Expectations and their Influence on Perception Perceived quality of food is the result of a perceptual pr ocess dependent on consumers' judgments (Ophius & Trijp, 1995). These judgments are formed on the bases of visible or invisible characteristics that may have actually been experienced by the consumer or are believed by the consumer to be associated with th e evaluated item. This perceived quality depends on the product, the person, and the environment, with the perceptual aspects of each holding sway. Two factors come into play here: quality cues and quality attributes. Quality cues can be established by the individual consumer's senses prior to consumption, whereas quality attributes are "benefit generating" aspects that cannot be observed prior to consumption (Ophius & Trijp, 1995). Taking into consideration the goals and interests of the current study, the main focus of discussion here will concern quality cues, such as labeling and presentation, and their role in consumer ratings and satisfaction. Though less empirical research has been done concerning the effects of quality expectations, appearance, and liking compared to menu labeling and plating which both affect food preferences and consumption, some literature does exist. One such study by Hurling and Sheperd (2003) is notable in these respects. This study focused on how consumers' visually influenced expectations of a product can enhance or degrade the
8 quality perception of that product before the item is ever tasted. After presenting female UK residents with raw and then cooked breaded fish steaks that were either visually appealing or unappealing (a ppealing conditions had a nice, consistent color and intact breading as opposed to the unappealing conditions which involved torn, flaky breading), participants answered questions concerning their expectations of the food. Results suggested that the appear ance of a raw product influenced expectations of cooked products in participants, in that unappealing raw items were expected to be lower in quality when cooked. This study showed the impact of expectations and how these effects follow an assimilation patt ern. Expectations from one stage in consumers' consumption process can influence expectations of subsequent stages of dining. Even perceived "freshness" can have an effect on sensory and quality expectations. Peneau, Hoehn, Roth, Escher, and Nuessli (2006) conducted a study to observe consumer expectations and perceptions concerning the quality of apples. At a public food exhibition, a computer testing station presented participants with a computerized questionnaire concerning liking and apple evaluation. R esults suggested that "freshness" was the primary criterion for apple choice regardless of age, gender, or apple consumption. Appearance and aroma, in that order, were the next two most highly ranked criterions. It is interesting that freshness, a rather v ague characteristic that cannot really be gauged before consumption, played such an important role in consumer acceptance and evaluation. These results showed how preconceived perceptions created by sensory attributes such as aroma, appearance, and perceiv ed "crispness" function together to affect consumer quality ratings even before the fruit was tasted. This
9 emphasizes the importance of pre consumption exposure to food items and how that exposure can have a vast impact on consumer ratings. In a study conc erning differences in sensory expectations (Wansink, Payne, & North, 2007), wine bottles with the same contents were labeled as either being from California (a region known for its wine) or North Dakota (a region not known for its wine). Participants were given questionnaires before and after consumption of the wine to gauge pre consumption sensory expectations and post consumption evaluations. Results showed that those in the California condition had more favorable taste expectations and higher post consum ption ratings not only for the wine, but also the companion food with which it was served compared to the North Dakota condition. This study provides even more evidence that taste expectations (established during pre consumption stages) can dramatically bi as sensory evaluations and also shows the effects that expectations can have on accompanying food items. Namkung and Jang (2007) investigated food quality perceptions and behavioral intentions as fundamental components concerning consumer satisfaction in a n attempt to identify factors that increase repatronage and overall dining satisfaction in a restaurant setting. Testing was conducted with the use of a questionnaire developed specifically for this study in five different mid to upper scale restaurants. These questionnaires gauged consumer intentions, quality expectations, satisfaction, and other similar concepts. Results show a positive link between food quality, satisfaction, perceptions and dining location intention. Regression analysis showed that p resentation (a quality cue, as mentioned previously) is a significant contributor towards customer satisfaction, receiving a higher statistical value than even taste. These results show just how critical
10 perceptual expectations (established in this study t hrough presentation) are when evaluating consumer attitudes towards food items. Cardello (1994), through his analysis of consumer expectations, investigated the concept that sensory attributes (e.g. appearance and/or presentation) and expectations (includ ing how food is described preconsumption) have a vast effect on an individual's central integration of an eating experience. This integration of each aspect that makes up an eating experience directly affects the acceptance, or liking, of that experience. Of the many different factors that potentially affect consumer acceptance of food items, the effects of and relationship between menu descriptions and plating aesthetics are key elements to understanding and manipulating sensory expectations and attributes that could potentially increase consumer ratings. Menu Descriptions and Labeling During World War II, a study was conducted on consumers' acceptance of meat products (as cited in Wansink, 2002). Considering that at this point in history choice meat was s carce for many, people were forced to eat organ meats due to the sheer lack of availability of other quality cuts. When the type of meat was specifically named, the food was found to be repulsive, but, when the type of meat was undisclosed, it was instead generally accepted. This study was the basis of research done by Wansink and Park (2002) concerning sensory suggestiveness and labeling. These researchers wanted to know to what extent labels suggestively influence liking for the taste of a product, and wi th that notion the researchers investigated soy. Soy is endorsed as a healthy food, but many American consumers report disliking the taste of soybeans. Wansink and Park wanted to know whether people honestly disliked soy or if this was another organ meat
11 s cenario. Participants in this study were given a nutrition bar that was labeled as either soy or just protein. They were then asked to answer a question concerning their intent to purchase the item, instructed to taste the product, then asked to complete t he rest of the survey on taste perceptions. Results showed that soy labeling negatively influenced product perceptions and responses concerning hedonics, but also lead to stronger beliefs concerning claims of healthiness. Overall, this study shows how infl uential labeling can be, especially concerning intention to purchase, and how consumers will often experience what they want or expect to taste based on judgments made pre consumption. When considering pre consumption factors, there have been multiple st udies concerning the effect of menu labels on consumer attitudes and sales. For example, Wansink and colleagues (Wansink et al., 2002; Wansink et al., 2005) have repeatedly investigated descriptive food names and their effects on consumer attitudes, sensor y perceptions, willingness to pay, and repatronage. Utilizing two 6 week field studies conducted in a university cafeteria, these researchers systematically rotated the displayed menu names of six different food items with their normal label and then with a descriptive label using geographic, nostalgic, or sensory related names every two weeks (e.g., "seafood filet" vs. "succulent seafood filet"). Any individual that chose one of the target items was asked by the cashier to complete a short questionnaire (s lightly different variations were used in the 2002 and 2005 experiments). The questionnaires asked about appeal, caloric estimate, liking, and intention to repurchase, as well as included a free response comment section. Collective results from both studie s revealed that when items were described with more illustrative names they were perceived as tastier and more appealing. Additionally, descriptive labels generated more positive comments from
12 consumers than their regularly named counterparts. These result s point to the conclusion that menu names can suggestively influence the perceived taste of food (Wansink et al., 2002; Wansink et al., 2005). Gueguen and Jacob (2010) have also investigated the effect of menu labels on consumer menu choice and sales. Duri ng a study in Brittany, France, researchers selected one item from three menu categories (salad, main dish, and dessert) that were the second most popular items in each category based on orders from the previous three months. Target food item labels were a ltered to produce descriptive names that evoked patriotic (Brittany Apple Pie), familial (Grandma's Apple Pie), and traditional (Traditional Apple Pie) ties. These alternatives were tested in daily rotation with the normal menu label, which served as a con trol, and tracked for number of orders. Results suggested that the descriptive labels were associated with increased sales compared to the usual menu names. Familial labels proved to be the most effective in influencing customer choice, followed by traditi onal and patriotic labels respectively. This may be due to positive emotions associated with family, as well as connotations of quality. These results confirmed the role that menu labels can play in the perceived attractiveness of a food item, such that pe ople chose items with descriptive labels more often than the normally labeled items. Aesthetics and Plating Appearance also plays a large role in food acceptance (Lyman, 1989). It can influence not only which foods people select but also how those food s are perceived to taste, which consequently affects hedonic ratings. Visual perception has many characteristics including organization, figure ground relationships, and other principles
13 that govern what individuals find appealing. These characteristics ha ve many implications for the realm of foodservice. Owners of eating establishments should consider these characteristics when attempting to build or improve their current business. Organization comes into play not only when considering the entire atmospher e of the restaurant, but also when planning how food will be displayed on a plate (the plating). Studying figure ground relationships can potentially help guide better choices for the development of food plating. Zellner and colleagues understood the impor tant role of plating and conducted multiple experiments testing how it affects liking for the taste of food (Zellner, Lankford, A,brose, & Locher, 2010; Zellner, Siemers, Teran, Conroy, Lankford, Agrafiotis, Ambrose & Locher, 2011). One such experiment gau ged the effect of balance and color on attractiveness and liking for food (Zellner et al., 2010). Water chestnuts and tahini paste were arranged on disposable plates and presented to participants. Participants were grouped in one of four conditions: unbala nced and monochromatic, unbalanced and multi colored, balanced and monochromatic, or balanced and multi colored. Balanced conditions received a symmetrical array of the food items while the unbalanced conditions received an asymmetrical array. Color was ma nipulated by adding different amounts of red and green food dye to the tahini. Ratings of attractiveness, willingness to try the food item, and liking were measured using 100 point Likert scales. Results revealed that color enhanced the attractiveness of t he balanced presentation, but not the unbalanced condition. Researchers suggest this is due to color adding complexity, which is more favorable and can be predicted when taking studies of artwork into account. Asymmetry and color were too complex, thus eli citing lower scores, while balance and
14 color complexity complimented each other. Participants also rated their willingness to try the food item as higher in the balanced condition. In another study, two more experiments were carried out, one testing balanc e and symmetry, the other testing clean presentation versus messy presentation, in an attempt to measure these effects (Zellner et al., 2011). The first experiment tested balance on a plate: participants were offered vegetables, hummus, and pita chips on a plate in either a symmetrical design or the same ingredients haphazardly tossed to one side of the plate. One hundred point Likert scales were used to gauge willingness to try the hummus and liking of the taste of the humus. Results suggest that balance d oes have an effect on liking for taste, producing higher ratings than the unbalanced plate. However, the researchers were worried that it wasn't asymmetry the participants saw, but instead a messy presentation, so a second experiment was carried out to acc ount for this. In the second experiment, portions of chicken salad were either neatly placed on the center of a lettuce leaf on a plate, or instead messily strewn across the lettuce leaf. The same measures of liking were used in the second experiment as we re the first. Results suggested that messiness did not affect willingness to taste the food, however it did have an effect on participants' liking for the taste of the chicken salad revealing higher scores. Together, the results of these experiments sugges t that neater, more balanced presentation of food items increases liking above that of its messy counterpart and show noticeable importance concerning the role of plating aesthetics in consumer liking. Not only can the appearance of food on a plate affect consumer ratings, but the plate itself can also have notable effects. In a study by Piqueras Fiszman, Alcaide, Roura, and Spence (2012) the influence of plate color on taste ratings was measured. Participants
15 were presented with the same strawberry mousse on both a black and a white plate (half of the participants received the black plate first, half received the white plate first) and then asked to complete scales measuring sweetness, quality, intensity, and overall liking for each. Results revealed that white plates received significantly higher marks for sweetness, intensity, and liking, although quality differences between plates did not reach significance. This suggests that plating color does indeed have an effect on consumer ratings, despite the fact that quality perceptions were not affected. Current Study Research involving the alteration of menu descriptions and presentation to manipulate attitudes seems to be most promising as far as cost and time efficiency is concerned. A menu overhaul is a lot less costly to restaurant owners than, for example, changing the entire interior atmosphere of the establishment. Given this notion, the purpose of the current study will focus on the relationship between two variables, presentation (fancy vs. crude) and m enu descriptions (descriptive vs. plain), concerning consumers' sensory perceptions and expectations surrounding a food item. The current literature lacks any research that tests both of these factors within the same study (which is more comparable to the actual nature of a restaurant experience). The research methodology aims to address this problem, as well as attempts to establish if one of these factors (naming or presentation) has a stronger effect on hedonic perception of taste than the other, since n o other research could be found that addresses these factors in such a way. Overall, the goal of this research was to identify factors that are easily altered in a restaurant setting, which could potentially increase not only sales but also the reception o f food items in professional establishments.
16 It was hypothesized that the more descriptive menu (Gueguen & Jacob, 2010; Wansink & Park, 2002; Wanksink et al., 2005, 2007) and more involved (fancier) presentation (Hurling & Shepard, 2003; Zellner et al., 2 010, 2011) would result in higher ratings for the food item based on previous findings in the literature Concerning sensory perception and expectations, it was hypothesized that overall ratings would be affected by condition pairings in the following desc ending order (such that the first pairing will receive the highest ratings and the last will receive the lowest ratings): descriptive menu/refined presentation (DR), plain menu/refined presentation (PR), plain menu/crude presentation (PC), and finally desc riptive menu/crude presentation (DC). This hypothesis was based on the hypothesis that sensory p erceptions are created and can be altered over the course of exposure (Cardello, 1994; Sorensen, Moller, Flint, Martens, & Raben, 2003). Exposure to the menu cr eates one set of expectations and whether or not these perceptions are congruent with what is presented to the participant during the presentation and tasting stages of testing should affect overall ratings as discussed above. Method Participants Sixty (40 females, 20 males) non vegan undergraduate students (age: M = 19.9 years, SD = 1.34) from a small liberal arts college in southwest Florida were recruited in person by the researcher or online via the student forum. Only non vegan students without food allergies were included in this sample because the nature of this study involves attitudes towards, as well as the tasting of, non vegan food items.
17 Materials Food item. Each participant was presented with the same food item regardless of condition. Th is item was a chocolate cupcake topped with strawberry whipped cream and a chocolate drizzle/shell. The researcher prepared all food items, paying specific attention to visible details in attempts to ensure the highest level of consistency. Any items that had visible defects or size differences were discarded. Menu stimuli. Two different menu names and descriptions, one per condition, were generated for this study. The menu for the descriptive condition incorporated a fanciful name Hi Hat Royale and a larg e number of adjectives in its item description: A scrumptious chocolate cupcake topped with fresh Florida strawberry whipped cream drenched in a decadent chocolate shell The menu used in the plain condition had a basic name Chocolate Strawberry Cupcake and a more simplistic description: Chocolate cupcake topped with strawberry whipped cream and chocolate Both menus were printed in black ink centered on white paper using a size 12 Bookman Old Style font. Bold typeface was used for item names and normal typeface was used for the descriptions. See Appendix A for menu stimuli. Plating stimuli. Two different versions of presentation were employed for testing during this study. In the refined presentation condition the cupcake was served on a small, flat, s quare, fake mother of pearl plate. Each plate was adorned with a chocolate garnish to create a more ornate presentation experience. In the crude condition the cupcake was served on a folded white napkin. Though napkins as plates are not commonly used in ac tual restaurants, they were employed in this study to emphasize a more unrefined presentation experience. Participants in both conditions were also given a
18 white square napkin and a 12oz clear plastic cup filled with water of which refills were allowed. Se e Appendix B for pictures of each presentation condition. Measures Three versions of a short questionnaire were also developed for this study. Questions were adapted from measures used in another study with permission of the researcher and altered to accom modate the focus of the current study (Wansink et al., 2002; B. Wansink, personal communication, February 16, 2012). Each questionnaire included four Likert scale items. Items on the first questionnaire included: "This item seems appealing"; "This item see ms to be of high quality"; "This item is something I would consider ordering from a menu"; and "This item seems appetizing." Each of these items was followed with a 5 point Likert scale ranging from "strongly disagree" (1) to "strongly agree" (5). The seco nd questionnaire only differed from the first in a few aspects of wording to emphasize the presence of the food item as opposed to its potential; for example: "This item is appealing" compared to "This item seems appealing" and "This item is something I wo uld be satisfied with if I had ordered it from a menu" compared to "This item is something I would consider ordering from a menu." The third questionnaire also differed in this respect but with emphasis placed on post consumption evaluation; for example: This item was appealing" and "This item was appetizing." The third survey also varied from the others in that the item concerning ordering the item from a menu was replaced with "My expectations for this item were" followed by three options: (1) not satisf ied, (2) met, and (3) exceeded. Each of the three questionnaires was presented at different phases of testing.
19 An additional questionnaire was also included to collect demographic information. Items included questions about age, gender, and year in school A single additional Likert scale item was also included concerning cooking habits. This item asked, "How often do you cook and/or bake?" followed by five possible responses: (1) never, (2) rarely, (3) sometimes/on occasion, (4) regularly, and (5) often. See Appendix C for complete measures. Procedure Participants were tested on campus in a dorm study room. This room included a table, chairs, and a blocked off area behind which food items were kept out of site from participants to avoid premature exposur e. Each participant was tested individually. Menu and plating conditions were randomly assigned to participants prior to testing. When they arrived, participants were seated at the table across from the researcher. A computer screen was positioned between the researcher and the participant to avoid the feeling of constant, intense observation as the participants answered questions. Participants were given a consent form to sign at the beginning of testing. Once this was collected, participants were given o ne of the two menu stimuli (based on random condition assignment) and the first questionnaire. Once the participant finished answering the items, the questionnaire and menu were collected by the researcher and put away out of sight. After this, participant s were presented with the food item plated according to their assigned condition and instructed to fill out the second questionnaire without tasting the cupcake. Again, once the participant finished this, the researcher collected the questionnaire and pla ced it aside with the previous materials. Finally, participants were instructed to taste as little or as much of the food as they wanted (though they were
20 required to at least taste the item once) and then asked to fill out the third and fourth questionnai res. After all materials were collected, participants were debriefed and any questions they had were answered. All questionnaires were kept anonymous and coded with an identification number to ease with data entry and to help keep track of the assigned con dition. Results The following hypotheses were tested: (1) that the descriptive menu condition would receive higher ratings than the plain one; (2) that the refined plating condition would receive higher ratings than the crude one; (3) Overall final ratings will be as follows in descending order: descriptive/refined, plain/refined, plain/crude, descriptive/crude. Descriptive Statistics Sixty participants (40 females, 20 males) ages 18 to 23 years ( M = 19.9, SD = 1.34) were tested. The majority of participan ts were first and second year students (28% and 30% respectively), followed by third year students (22%), and then fourth years and beyond (20%). The majority of participants had at least some familiarity with cooking and/or baking with 73% of responses in dicated that they cooked and/or baked either sometimes/on occasion, regularly, or often (Fig 1). Menu Descriptions The researcher hypothesized that the descriptive menu condition would receive higher scores than the plain menu condition. T tests were co nducted to assess differences between the two conditions for each item on the first questionnaire. Results show that the descriptive menu condition received higher ratings across all items. For "This item seems
21 appealing," significant differences were not reached between the two conditions, t = 1.88 p = 0.0651. Though this test did not reach significance, it did approach significance at a .05 level. The descriptive menu condition received a mean(standard deviation) score of 4.8(0.41) while the plain menu c ondition received a mean score of 4.63(0.61). In regards to "This item seems to be of high quality," significant differences were found between the two conditions, t = 2.28, p = 0.0264, with the descriptive menu condition receiving a mean score of 4.43(0.8 1) and the plain menu condition receiving a mean score of 4.2(0.68). Significance differences were also observed concerning "This item is something I would consider ordering from a menu," t = 2.48, p = 0.0163, with the descriptive condition receiving a mea n score of 4.53(0.57) while the plain condition received a mean score of 4.37(.76). For the final item, "This item seems appetizing," the descriptive menu condition had a mean score of 4.47(0.63) and the plain condition received a mean score of 4.6(0.72), t = 1.83, p = 0.0718, revealing non significant differences between the two. The hypothesis not was confirmed for all survey items, but the descriptive menu condition did receive higher scores than the plain menu condition, significantly so for two items. Plating Aesthetics It was hypothesized that the refined plating condition would receive higher scores than the crude plating condition. A general linear model (GLM) was conducted for each survey item, as indicated by F tests GLM statistics tested the com plete model, dependent variable = menu + plate + menu*plate, then main effects and interactions are reported. Although no significant results were observed, interesting trends in the means occurred.
22 Appeal. Concerning the survey item "This item is appealin g," significant differences were not found between the conditions, F (3, 56) = 0.7, p = 0.5572. No significant main effect was found concerning menu, F (1 56 ) = 01.5 p = 0.2265, nor plating, F (1, 56) = 0.06, p = 0.8077 and there was no significant interacti on between the two F (1, 56) = 0.54, p = 0.4661. The DR condition pairing received the highest scores with a mean(standard deviation) of 4.87(0.352), followed by DC with a mean of 4.73(0.458), then PC with a mean of 4.67(0.617), and finally PR with a mean o f 4.6(0.632). Results show that for this item, regardless of plating condition, those exposed to the descriptive menu condition allotted the food item slightly higher ratings, though not significantly so. Quality Concerning t he item "This item is of high quality," significant differences were not found between the conditions, F (3, 56) = 0.74, p = 0.5354. No significant main effect was found concerning menu, F (1 56 ) = 1.44 p = 0.235, nor plating, F (1, 56) = 0.03, p = 0.8644 and there was no significant int eraction between the two F (1, 56) = 0.74, p = 0.3948. Rankings also did not follow the same pattern as before. For this item, the PC pairing received the highest mean of 4.53(0.516), followed by PR with a mean of 4.33(0.816), then finally the DR and DC pai rings, with means of 4.27(0.704) and 4.13(0.915) respectively. Pertaining to inquiries of quality, the condition pairings involving the plain menu received slightly higher ratings, not significantly so. Appetizing characteristic. Concerning the item "This item is appetizing," significance was not reached F (3, 56) = 0.28, p = 0.8386. No significant main effect was found concerning menu, F (1 56 ) = 0.56, p = 0.4563, nor plating, F (1, 56) = 0.14, p = 0.709 and there was no significant interaction between the two F (1, 56) = 0.14, p = 0.709.
23 For this survey item, rankings followed the same pattern as above. The PC pairing received the highest scores with a mean of 4.67(0.816), followed by the PR pairing with a mean of 4.53(0.64), and finally the DC and DR pairin gs received the lowest scores both with a mean of 4.47(0.634), though these differences were not significant. Satisfaction Concern ing the item This item is something I would be satisfied with if I had ordered it from a menu," significance differences we re not found between the conditions, F (3, 56) = 0.61, p = 0.6112. No significant main effect was found concerning menu, F (1 56 ) = 0.9 p = 0.3475, nor plating, F (1, 56) = 0.9 p = 0.3475 and there was no significant interaction between the two F (1, 56) = 0. 04, p = 0.8504. Again, the PC pairing received the highest scores with a mean of 4.6(0.507), followed by the PR and DC pairings both with a mean of 4.47(0.634), and then finally the DR pairing received the lowest scores, with a mean of 4.27(0.884). Signifi cant differences between these groups, however, were not found Post consumption Ratings and Satisfaction of Expectations It was hypothesized that overall ratings would be affected by condition pairings in the following descending order (such that the fi rst pairing will receive the highest ratings and the last will receive the lowest ratings for each questionnaire item): DR, PR, PC, and finally DC. A general linear model (GLM) was conducted for each survey item, as indicated by F tests GLM statistics tes ted the complete model, dependent variable = menu + plate + menu*plate, then main effects and interactions are reported. Although no significant results were observed, interesting trends in the means occurred. Appeal. Concerning the item "This item was ap pealing," significant differences were not found between the pairings, F (3, 56) = 0.16, p = 0.9203. No significant main
24 effect was found concerning menu, F (1 56 ) = 0.25, p = 0.6221, nor plating, F (1, 56) = 0.25, p = 0.6221 and there was no significant int eraction between the two F (1, 56) = 0, p = 1. However, the DR condition pairing did receive the highest scores as predicted with a mean of 4.73(0.458), followed by DC and PR with means of 4.67(0.617) and 4.6(0.488) respectively, and finally PC with a mean of 4.6(0.507). Results suggest that for this item, the descriptive menu/refined plating condition pairing did receive the highest final ratings as predicted, though not significantly so. Quality Concerning t he item "This item was of high quality," signif icant differences were not found between the condition pairings, F (3, 56) = 0.83, p = 0.4821. No significant main effect was found concerning menu, F (1 56 ) = 0.34 p = 0.565, nor plating, F (1, 56) = 0.34 p = 0.565 and there was no significant interaction b etween the two F (1, 56) = 1.82, p = 0.1822. Also, condition pairings also did not follow the predicted pattern. The PC pairing received the highest mean score of 4.6(0.632), followed by DR with a mean of 4.4(0.632), then finally the DC and PR groups both w ith a mean of 4.26(0.704), though not significantly so. Appetizing characteristic. Concerning the item "This item was appetizing," significance was not reached, F (3, 56) = 0.3, p = 0.827. No significant main effect was found concerning menu, F (1 56 ) = 0.0 5 p = 0.8292, nor plating, F (1, 56) = 0.45, p = 0.5182 and there was no significant interaction between the two F (1, 56) = 0.45, p = 0.5182. Though the condition pairings did not show significant differences for this item, the pairings did follow the predi cted pattern. The DR pairing received the highest scores with a mean of 4.73(0.458), followed by the PC and PR pairings both with a mean of
25 4.6(0.632), and finally the DC pairing received the lowest scores with a mean of 4.53(0.64). Expectations Concernin g the final item on the third questionnaire, "My expectations for this item were" with the possible responses of (1) "not satisfied," (2) "met," or (3) "exceeded", significant differences were not found, F (3, 56) = 0.98, p = 0.4072. No significant main eff ect was found concerning menu, F (1 56 ) = 2.83 p = 0.0978, nor plating, F (1, 56) = 0.06, p = 0.8108 and there was no significant interaction between the two F (1, 56) = 0.06, p = 0.8108. The frequency of responses across condition pairings, however, showed an interesting trend in the data that strayed from the predicted pattern. Response frequencies show a tendency for participants in either of the descriptive menu condition pairings (DR and DC) to respond that their expectations had been "met", while both p lain menu condition pairings (PR and PC) showed more frequent "exceeded" responses (Fig 2). Discussion Implications The effects of menu descriptions and plating aesthetics on ratings for the first two phases of testing. Two research hypotheses were ge nerated concerning the effects of menu descriptions and plating aesthetics: (1) the descriptive menu condition would receive higher scores than would the plain menu condition, and (2) the refined plating condition would receive higher scores than would the crude plating condition. The first of these two hypotheses, which addressed the menu conditions, was partially supported. The descriptive menu received higher scores than did the plain menu across all survey items
26 for the first questionnaire, significantl y so concerning quality and intent to order the item from a menu. These results show that participants reading the descriptive menu thought the listed item was more appealing and appetizing, as well as being of higher quality than participants reading the plain menu. Participants in the descriptive condition were also more likely to consider ordering this item off of a menu than were their plain condition counterparts. This confirms previous findings that attest to how influential menu descriptions can be o ver consumer ratings, showing preference towards more embellished labels and descriptions ( Gueguen & Jacob, 2010; Wansink & Park, 2002; Wanksink et al., 2005, 2007) The data did not, however, support the second hypothesis. Though significant differences w ere not observed, more often than not the pairings that included the crude plating condition received slightly higher ratings than did the refined plating condition on the second questionnaire, contrary to expectations. Although, this hypothesis was not co nfirmed, some interesting trends in the data were found. Mean ranks showed that for the survey item "This item is appealing", condition pairings involving the descriptive menu received slightly higher ratings than did the pairings that involved the plain m enu, with the pairings that involved refined plating conditions receiving slightly higher ratings within the menu condition separation (from highest to lowest: DR, DC, PR, PC). This observation suggests that when considering how appealing a food item is to consumers pre consumption, perceptions created during the menu stage of an eating experience could possibly hold more influence over ratings than manipulation of plating aesthetics, though mean ranks suggest that the refined plating condition was slightly preferred over the crude condition.
27 Trends were the opposite for the items during the second round of testing concerning ratings for level of quality, how appetizing the item was, and whether or not the consumer would be satisfied if they had ordered the item from a menu. For these inquiries, though not significantly so, the plain menu/crude plating condition pairing received the highest ratings for all items, followed by the plain menu/refined plating condition pairing, the descriptive menu/crude plating pairing, and finally the descriptive menu/refined plating, which usually received the lowest ratings. A possible explanation for the observed trend could be that those exposed to the plain menu had lower expectations, which were then exceeded by the visual stimuli of the food item itself (not including any plating aesthetics). At this point it seems necessary to mention the ceiling effect observed for the cupcake. Concerning all responses to items that utilized a 5 point Likert scale (excluding the demogra phics questionnaire) only 15 individual data points out of 660 were ratings that suggested dislike for the item (lower than three/neutral), and only two out of those 15 occurred after exposure to the cupcake. This observed fondness may have overshadowed th e influence of plating manipulation during the second phase of testing and it is also a possible explanation for the higher ratings concerning the pairings involving the plain menu and the crude plating (for the quality, appetizing, and order satisfaction survey items on the second questionnaire). Again if lower expectations were formed during the menu phase, and then exceeded by the food item itself, it is logical that ratings could be higher for pairings that involved the plain menu. Concerning the plati ng conditions, it is possible that liking of the food item itself was only emphasized by the contrast of the crude plating (a folded napkin). Different trends in the data may have been observed if
28 the food item was not as well received on its own regardles s of possible outside influences. Post consumption ratings and the role of expectations It was predicted that the final post consumption ratings would be as follows (from highest to lowest): descriptive/refined, plain/refined, plain/crude, descriptive/cru de. This hypothesis was not supported by the results, as the pairings did not follow the expected order of ranks nor were any groups found to be significantly different for any of the survey items. For the survey items concerning appeal and how appetizing the cupcake was, the DR condition pairing did receive the highest ratings as predicted though not significantly so. Trends in the data concerning responses to the satisfaction of expectations item ("My expectations for this item were" with possible respons es of (1) not satisfied, (2) met, or (3) exceeded), were not significantly different, however, they are interesting and worth discussing. Frequency distributions suggest that it is possible that the plain menu created lower expectations, which were then ex ceeded by the actual food item (regardless of plating), compared to the descriptive menu, which set expectations high only to be met instead of exceeded. This is a possible explanation as to why both condition pairings that included the descriptive menu re ceived responses that were lower than the pairings that included the plain menu. To reiterate previous statements, this ranking hypothesis was based on the notion that sensory p erceptions are created and can be altered over the course of exposure to a meal (Cardello, 1994; Sorensen, Moller, Flint, Martens, & Raben, 2003). What might have happened in this case was that the presence of the cupcake itself was considered highly pleasant, as previously mentioned with the observed ceiling effect, and the manipula tion of plating had minimal influence (though within the pairings that
29 involved the plain menu condition, PC and PR, responses that indicated participant expectations had been exceeded were slightly more frequent for the pairing that included exposure to the crude plating condition, PC, than for its refined counterpart, PR). These findings suggest that if a food item itself is liked, it is possible that inducing lower expectations during pre consumption phases through the utilization of a less descriptive menu and more simplistic plating could yield higher post consumption satisfaction ratings. Limitations There are a few limitations that should be addressed concerning this study. First and foremost being the previously discussed ceiling effect. As mention ed before, only 15 out of 660 data points revealed lower than neutral ratings and only two out of those 15 occurred after exposure to the cupcake. This high level of liking for the food item itself creates the potential to skew results and may be why signi ficant differences were not found between condition pairings for the second and third phases of testing. It is possible that liking for the food item outweighed any manipulation attempt made by the researcher to increase or decrease liking. The chosen t esting environment should also be acknowledged here. This study was conducted in a dorm study room, which is normally utilized for group meetings and homework. This is not an environment that one would normally choose to consume food compared to a cafeteri a, mock restaurant, food lab, or even dorm kitchen. Weber, King, and Meiselman (2004) have discussed how environment can affect preference and expectations concerning food items and considering that this is was abnormal setting in which to consume food, ra tings may have been influenced by this factor.
30 The final limitations to be discussed are the generalizability of the current sample as well as any suggestions made from the data beyond the effectiveness of the descriptive menu. Given the population from w hich the sample was taken, findings from this study should not be generalized beyond that of 18 to 23 year old undergraduate college students. Also, considering that all analyses after the first phase of testing on the two menu conditions showed insignifi cant differences, any of the minor trends that were found here were only pointed out because they were interesting. These trends should be taken as possible suggestions for future inquiries, and not as actual statements of consequence. Future research Fut ure research endeavours should continue to investigate the relationship between menu descriptions and plating aesthetics, as well as how these two things interact to create pre consumption expectations which then affect post consumption ratings. It would b e wise to use a more generalizable population with greater variance in the age of participants. A better testing setting is also recommended. It would probably be best to conduct this type of study in a food lab or mock restaurant, or, if the resources are available, in a professional eating establishment. This final option would allow for a more realistic gauge of consumer ratings. Future researchers should also find a more neutral food item to test in an attempt to avoid the observed ceiling effect. For example, an item that isn't as favourable as a dessert, possibly a side item like a vegetable, pasta dish, or salad. Another suggestion is to use a Likert scale larger than 5 points. There would likely be more variance in
31 response if the participants had m ore options to choose from concerning the level of agreement with the statements. Conclusion Though only one of the generated hypotheses was supported, overall, this study is a hopeful first step towards a better understanding of how menu descriptions and plating aesthetics influence consumer ratings and expectations. It was confirmed that a menu item with a more embellished description that relies on the heavy use of adjectives receives higher initial ratings than a simpler, plain menu description. Many o ther possible implications were also suggested here and they should be used as consideration for future research. The goal of this study was to identify easily manipulated factors of a meal experience that could increase sales and consumer ratings. The use of descriptive menus has shown to receive higher initial consumer ratings and this finding suggests a possible way to increase sales. Also, although no consistent or significant patterns were illuminated, possible implications concerning how lowered expec tations can increase consumer satisfaction if the food item itself is liked give way to numerous paths of inquiry and further investigations could possibly identify ways in which to not only increase the sale of food items in professional establishments bu t also increase post consumption satisfaction ratings, which was the other goal of this research.
32 Figure 1 Frequencies of responses to the item "How often do you cook and/or bake?"
33 Figure 2 Frequencies of responses to the item "My expectations for this item were."
34 Appendix A Descriptive Menu stimuli : Hi Hat Royale A scrumptious chocolate cupcake topped with fresh Florida strawberry whipped cream drenched in a decadent chocolate shell Plain Menu stimuli : Chocolate Strawberry Cu pcake Chocolate cupcake topped with strawberry whipped cream and chocolate
35 Appendix B Refined Plating stimuli: Crude Plating stimuli:
36 Appendix C Please circle the response that represents your attitudes towards the following statements (A) This item seems appealing: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree This item seems to be of high quality: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree This item is something I would consider ordering from a menu: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree This item seems appetizing: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree
37 Please circle the response that represents your attitudes towards the following statements (B) This item is appealing: 1 2 3 4 5 strongly disagree disagree neutral ag ree strongly agree This item is of high quality: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree This item is something I would be satisfied with if I had ordered it from a menu: 1 2 3 4 5 stron gly disagree disagree neutral agree strongly agree This item is appetizing: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree
38 Please circle the response that re presents your attitudes towards the following statements (C) This item was appealing: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree This item was of high quality: 1 2 3 4 5 strongly disa gree disagree neutral agree strongly agree This item was appetizing: 1 2 3 4 5 strongly disagree disagree neutral agree strongly agree My expectations for this item were: 1 2 3 not satisfie d met exceeded
39 Please answer the following questions What is your age? ___________ With which gender do you most comfortably identify? ___________ What year are you? (circle one) First year Second year Third year Fourth year (or above) How often do you cook and/or bake: 1 2 3 4 5 Never Rarely Sometimes/On occasion Regularly Often
40 References Cardello, A. V. (1994). Consumer expectations and their role in food acceptance. In H. MacFie & D. Thomson (Eds.), Measurement of Food Preferences London: Blackie Academic. Drewnowski, A. (1997). Taste Preferences and Food Intake. Annual Review of Nutrition, 17, 237 253. Gueguen, N., & Jacob, C. (2010). T he effect of menu labels associated with affect, tradition, and patriotism on sales. Food Quality and Preference, 23 (1), 86 88. Hurling, R., & Sheperd, R. (2003). Eating with your eyes: Effect of appearance on expectations of liking. Appetite 41, 167 174. Koster, E. (2003). The psychology of food choice: Some often encountered fallacies. Food Quality and Preference 14 359 373 Lyman, B. (1989). A psychology of food: more than a matter of taste New York: Van Nostrand Reinhold Co. Meiselman, H.L. J ohnson, J.L., Reeve, W., & Crouch, J.E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35 (3), 231 237. Namkung, Y. & Jang, S. (2007). Does food quality really matter in restaurants? Its impact on custome r satisfaction and behavioral intentions. Journal of hospitality & tourism research 31 (3), 387 409. Ophuis, P., & Trijp, H. (1995). Perceived Quality: A market driven and consumer oriented approach. Food Quality and Preference, 6 177 183. P erkins, K., Epstein, L., Stiller, R., Fernstrom, M., Sexton, J., & Jacob, R. (1990).
41 Perception and Hedonics of Sweet and Fat Taste in Smokers and Nonsmokers Following Nicotine Intake. Pharmacology Biochemistry & Behavior 35, 671 676. Pneau, S., Ho ehn, E., Roth, H.R., Escher, F., & Nuessli, J. (2006). Importance and consumer perception of freshness of apples. Food Quality and Preference, 17(2), 9 19. Piqueras Fiszman B. Alcaide J. Roura, E., & Spence, C. (2012) Is it the plate or is it the food? Assessing the inuence of the color (black or white) and shape of the plate on the perception of the food placed on it. Food Quality and Preference, 24, 205 208. Sorensen, L.B., Moller, P., Flint, A., Martens, M., & Raben, A. (2003). Effect of s ensory perception of foods on appetite and food intake: A review of studies on humans. International Journal of Obesity, 27, 1152 1166. Stroebele, N., & Castro, J. D. (2004). Effect of ambience on food intake and food choice. Nutrition 20 821 838. Wansink, B., & Park, S. (2002). Sensory suggestiveness and labeling: Do soy labels bias taste? Journal of Sensory Studies, 17, 483 491. Wansink, B., Payne, C., & North, J. (2007). Fine as North Dakota wine: sensory expectations and the intake of compan ion foods. Physiology & Behavior, 90 712 716. Wansink, B., van Ittersum, K., & Painter, E. (2002). How descriptive menu labels influence attitudes and repatronage. Advances in Consumer Research, 29, 168 172.
42 Wansink, B., van Ittersum, K., & Painter, E. (2005). How descriptive food names bias sensory perceptions in restaurants. Food Quality and Preference, 16, 393 400. Weber, A., King, S., & Meiselman, H. (2004). Effects of social interaction, physical environment, and food choice freedom on consu mption in a meal testing environment. Appetite, 42, 115 118. Zellner, D. A., Lankford, M., Ambrose, L., & Locher, P. (2010). Art on the plate: Effect of balance and color on attractiveness of, willingness to try and liking for food. Food Quality and Preference, 21, 575 578. Zellner, D A, Siemers, E, Teran, V, et al. (2011). Neatness counts. How plating affects liking for the taste of food. Appetite, 57(3), 642 648.