Individual differences in food perceptions and calorie estimation: An examination of dieting status, weight, and gender

Abstract

People frequently place foods into “health” or “diet” categories. This study examined whether (1) evaluations of “healthiness/unhealthiness” influence “caloric” estimation accuracy, (2) people evaluate foods for “healthiness/unhealthiness” or “weight gain/loss” differently, and (3) food evaluations differ by gender, diet status, and weight. Also, undergraduate dieters attempting to lose weight on their own were compared to obese weight loss program participants. Undergraduate students (N=101) rated eight “healthy” and “unhealthy” foods on perceived “healthiness/unhealthiness,” “weight loss/gain capacity” and “caloric” content. Open-ended questions inquiring why a food was “healthy/unhealthy” or would “contribute to weight gain/loss” were coded into independent food categories (e.g., high fat). Results indicate that calories were systematically underestimated in healthy/weight loss foods, while they were systematically overestimated in unhealthy/weight gain foods. Dieters were more accurate at estimating “calories” of healthy foods and more attentive to the foods’ fat, “calorie”, and sugar content than non-dieters. Overweight participants commented more on fat and sugar content than normal weight participants. Undergraduate dieters used fewer categories for evaluating foods than weight loss program participants. Individual difference characteristics, such as diet-status, weight, and gender, influence people's perceptions of foods’ healthiness or capacity to influence weight, and in some instances systematically bias their estimates of the caloric content of foods.

Introduction

The obesity epidemic in the United States and abroad has heightened consumer concerns about foods’ capacity for weight gain and health. However, even with heightened concerns, people often make snap judgments about food choices rather than making informed food selections (e.g., Rozin, Ashmore, & Martwith, 1996). Research suggests that beyond the use of overarching taxonomic food-group categories (e.g., fruits, vegetables, meats), people frequently categorize foods according to health or diet (e.g., healthy, diet, junk food; Ross & Murphy, 1999). These categories provide simplifying strategies (i.e., heuristics; Tversky, 1974) for making food choices. For example, Oakes & Slotterback (2001a), Oakes & Slotterback (2001b), Oakes & Slotterback (2001c), Oakes & Slotterback (2002) provide convincing evidence that foods acquire reputations as healthy or unhealthy. Research suggests that the decision to place a food into a healthy or diet category is likely to be influenced by factors such as the food's perceived fat content (Carels, Harper, & Konrad, 2006; Oakes & Slotterback (2001a), Oakes & Slotterback (2001b), Oakes & Slotterback (2001c), Oakes & Slotterback (2002)).

Interestingly, once a food is categorized as unhealthy, the food is commonly perceived to possess more “calories” than it actually does (e.g., Booth, 1987; Carels et al., 2006) and to contribute to weight gain (Carels et al., 2006). Alternatively, there is some evidence for a similar but opposite relationship with foods categorized as healthy. For example, in a sample of behavioral weight loss program (BWLP) participants, a food perceived as healthy was also perceived to possess fewer “calories” than it actually did (Carels et al., 2006). One goal of this investigation was to examine whether a systematic caloric discrepancy for healthy and unhealthy foods is observed in a more heterogeneous sample (e.g., gender, diet status, weight) not participating in a BWLP. It was hypothesized that people would overestimate the caloric content of foods perceived as unhealthy and underestimate the caloric content of foods perceived as healthy.

Research suggests that the categorization of foods as healthy or as diet foods is likely to influence people's perceptions of those foods. In other words, some characteristics of foods may be perceived to promote weight loss (i.e., satiety) but not health and vice versa. Interestingly, while the factors used to characterize healthy/diet foods or unhealthy/non-diet foods are quite similar (Carels et al., 2006), the factors are not completely interchangeable and subtle distinctions between these food categories are often made. For example, when describing perceived healthy and unhealthy foods, obese BWLP participants were more likely to comment on the food's nutrient content; however, when describing a food's capacity for weight loss or weight gain, participants were more likely to comment on “calorie” content and the ability of the food to provide a feeling of fullness (Carels et al., 2006). Again, this investigation was designed to replicate previous findings observed with obese, BWLP participants in a more heterogeneous sample (i.e., gender, diet status, weight). It is hypothesized that when describing perceived healthy and unhealthy foods, participants would be more likely to comment on the foods’ nutrient content, and when describing a foods’ capacity for weight loss/gain, participants would be more likely to comment on caloric content and the ability of the food to provide a feeling of fullness.

While the health and diet categorization of foods clearly influence perceptions of foods, there is some limited evidence that individual difference factors, such as gender, diet-status, BMI, etc. are associated with perceptions of food healthiness. For example, research suggests that women are socialized to eat differently than men (Rolls, Federoff, & Guthrie, 1991) and that they might view foods differently. Compared to men, women report eating healthier foods, consider themselves to be more knowledgeable about what foods are good, and report reading nutrition labels more often (Oakes & Slotterback, 2001a). Research also suggests that compared to men, women appear to depend more on fat content than other factors when evaluating a food's healthfulness (Oakes & Slotterback (2001a), Oakes & Slotterback (2001b), Oakes & Slotterback (2001c)). Nevertheless, not all studies have noted gender differences in food perceptions (Oakes, 2005; Oakes & Slotterback, 2002). No gender differences were observed when men and women were rating foods’ perceived capacity for weight gain (Oakes, 2005), nor were gender differences observed when middle aged men and women considered the factors most important when judging foods’ healthfulness (Oakes & Slotterback, 2002). The current study used a methodology not utilized in previous research examining men's and women's preceptions of food. Participants were asked open-ended questions about “Why a food was healthy or unhealthy” and “Why the food would contribute to weight gain or loss?” and the responses were later coded into common food-related characteristics such as fat, calories, etc. We hypothesized that women would be more likely to comment on the food's fat content than men when determining a food's healthiness or capacity for weight gain.

Food-related goals, such as losing weight, are also likely to influence people's perceptions of food. For example, at the point of purchase, dieters tend to rate fat content as most important when judging a food's healthfulness, whereas non-dieters rate freshness as most important (Oakes & Slotterback, 2002). Research on dieters versus non-dieter's perceptions of foods’ healthfulness and capacity for weight gain has been limited. We thought it plausible that dieters would be more attentive to the caloric and fat content of foods and thus, better able to estimate the energy in foods. It was hypothesized that dieters would be more concerned about fat and “calories” in foods and more accurate at caloric estimation than non-dieters. Also, in prior research, higher baseline BMI in BWLP participants was associated with greater inaccuracy in “caloric” estimation (Carels et al., 2006). In a sample not restricted to obese BWLP participants, it was hypothesized that individuals with a higher BMI would be less accurate in caloric estimation than individuals with a lower BMI.

Finally, research suggests that older adults utilized a greater number of descriptive characteristics to categorize foods compared to college students (Oakes & Slotterback, 2001b). Similarly, we thought it plausible that older obese BWLP participants would have different food-related concerns than undergraduate dieters since they more likely have greater experience in caloric estimation and diverse exposure to messages regarding important aspects of foods (e.g., high fiber, sugar, etc.) that contribute to health and weight. Therefore, we compared undergraduate dieters to obese BWLP participants (Carels et al., 2006) on food evaluation and caloric estimation. It was hypothesized that obese BWLP participants would be more accurate in their caloric estimation and use a greater number of descriptive categories when discussing a food's healthiness or potential for weight loss than undergraduate dieters.