INTRODUCTION

Misperception of weight is defined as the discordance between an individual's actual weight and the perception of his/her weight status. Weight perception and misperception might influence the healthy or unhealthy behaviors people engage in ¹. In this regard, misperception has repeatedly been documented among overweight and obese adults, and it has been hypothesized that weight misperception among overweight and obese individuals may preclude the adoption of healthful attitudes and behaviors, perhaps as a result of lower weight loss motivation. Overweight and obese individuals who consider their weight healthy, for example, might not try to lose weight and might be less inclined to eat healthfully and to be physically active ². On the other hand, some evidence indicates that weight misperception among overweight and obese individuals might be associated with healthful behaviors (e.g., better diet quality, more physical activity, and less sedentary behavior) ²,³. Perceiving oneself as overweight-obese is relevant given the association between that perception and unhealthy weight-control behaviors ¹.

It seems that overweight misperception varies according to gender (among other variables), females tending to perceive themselves as overweight more than males do, even at the same measured body mass index (BMI) ⁴,⁵-⁶. Misperception of overweight-obesity among adolescents of normal weight might have negative consequences. The combination of overweight-obesity misperception causing body dissatisfaction predicts dieting, and dieting is a clear risk factor for developing different eating disturbances ⁶. In addition, adolescents who have been engaged in dieting and other unhealthy weight-control behaviors have been found to be at risk of weight gain over time ⁷,⁸.

Besides the concept of weight misperception, one's body shape and/or one's body image play a relevant role in different behaviors ⁹. Both weight misperception and poor body image have negative psychological and psychosocial effects (e.g., low self-esteem, anxiety, depression, isolation, discrimination, family conflicts, etc.). It is well known that people engaged in a process of self-evaluation (included body checking) comparing themselves to others who they believe have more desirable sociocultural traits tend to be involved in behaviors aimed to achieve those desired characteristics ¹⁰.

Self-reported physical fitness is another variable to consider as a starting point to different healthy or unhealthy behaviors. Perceived physical ability (i.e., the individual's perception of physical abilities developed over time as a result of cumulative interactions with the environment) and perceived physical competence are two goal-oriented self-perception constructs ¹¹. Recently, it has been shown that body dissatisfaction is a significant mediator of the effect of BMI on perceived physical activity ¹². A large body of research has aimed to validate the idea that exercise improves body image through changes in physical fitness ¹³. However, it has been suggested that improvements in physical fitness play a minor role in changing body image, because the effects of physical exercise and activity on body satisfaction should be mediated by changes in individuals' perceptions of their physical fitness and competence.

As a result, it seems that perceptions (weight, physical fitness) are core constructs to lead to healthy/unhealthy behaviors more than actual weight or actual physical fitness do.

Healthy or unhealthy behaviors as consequence of different perceptions and their corresponding psychological and psychosocial effects lead to the concept of emotional eating among other eating behaviors. Thus, it has been distinguished among restraint eating (conscious restriction of food intake aimed to control body weight and/or to promote weight loss), uncontrolled eating (inability to resist emotional cues, eating as a response to different negative emotions) and emotional eating (tendency to eat more than usual due to a loss of control over intake with a subjective feeling of hunger). Other authors have defined external eating as the tendency to overeat in response to external food-related cues like the sight, smell, and taste of palatable food, regardless of their physical need for food ¹⁴,¹⁵-¹⁶. Moreover, some authors have noted that different types of bingers and dieters may be found: bingers who are engaged in restraint-induced binging, and bingers generally disinhibited; dieters who eventually become disinhibited and overeat, and dieters who maintain the restrictive attitude ¹⁷,¹⁸.

It has been reported that BMI and negative emotional eating are highly related whereas positive emotional eating and external eating loaded onto another factor. In this regard, it is plausible that even though positive emotions may elicit eating, they do not necessarily mirror disordered eating. Eating in response to positive emotions might rather be related to hedonic or external eating ¹⁹,²⁰.

Some models conceptualize eating disturbances as disorders of affect regulation, considering the impairment in the cognitive capacity to process and regulate emotions as the primary regulatory disturbance ²¹. According to these models, some eating behaviors, such as binge eating and compensatory behaviors, as well as restricted food consumption, are interpreted as responses to cope with intense or relatively undifferentiated emotional states ²²,²³,²⁴. These intense emotional states are usually linked to self-perceptions (weight, physical fitness, body shape/body image), especially among adolescents ²⁵,²⁶,²⁷-²⁸.

Based on the above-mentioned previous research, the objectives of this study, focused on preadolescents and adolescents, were: a) to analyze weight self-perception and self-reported physical fitness; b) to explore the possible relationship between these variables and some weight control behaviors (dieting, self-weighing frequency); to explore possible relations among weight self-perception, self-reported physical fitness, dieting, self-weighing frequency and BMI/body image; and d) to analyze the relation between all these variables and different eating behaviors.

METHOD

RESULTS

The sample comprised 336 students, 160 females (47.62%) and 176 males (52.38%), with a mean age of 12.46 ± 2.14. With respect to BMI, the mean was 20.18 ± 3.58. Considering the value of BMI-specific percentiles for age and sex, 57.65% of participants had normal weight, 14.41% had overweight and 26.73%, obesity. Thus, overweight + obesity (BMI ≥ 85th percentile) was 41.14%.

Considering weight self-perception, 66.02% of participants perceived themselves as normal weighted, 19.09% as overweight and 4.21% as obese. This way the perceived total overweight was 23.30%. No significant gender differences with respect to weight self-perception were found (2 = 8.04; p = 0.09). Overall weight misperception was 44.77%. Among those participants who had overweight, 69.23% misperceived their weight, mainly considering that they had normal weight (62.82%). In case of obese participants, 88.63% misperceived their weight, 84.09% of them perceiving themselves as normal weight or slightly overweight. While 49.68% of boys misperceived their weight, in the case of girls that percentage was 39.58%. Nevertheless, among those who perceived themselves as very overweight, 76.92% were girls. On the contrary, among those who perceived themselves as very or slightly underweight, 62.5% and 60% respectively were boys.

Respecting self-reported physical fitness, most of participants reported an average (34.57%) or good (38.27%) physical fitness; poor (2.47%), fair (13.9%) and excellent (10.80%) were the reported physical fitness of the rest. Considering gender differences, while 57.40% of men reported good and excellent physical fitness, in case of women that percentage was 40.52%. On the contrary, 28.40% of men reported average physical fitness while this percentage was 41.83% for women (2 = 14.59; p < 0.01).

Most participants who considered to have average, good or excellent physical fitness reported to be about the right weight (66.79%). When the reported physical fitness was poor or fair, there were more participants who perceived themselves as slightly or very overweight (55.32%). Bearing in mind the actual weight, we found similar results. Figure 1 represents the participants who considered their physical fitness as average, good or excellent and the corresponding percentages related to weight self-perception and actual weight (classified as normal, overweight or obesity). Among boys, considering not weight perception but actual weight, good and excellent physical fitness was reported by 72.27%, 42.10% and 25.92% of participants at normal weight, overweight and obesity, respectively. In case of girls, these percentages were 45.34%, 38.63% and 16.66%.

Figure 1. Actual weight, weight self-perception and self-reported physical fitness (average, good, excellent). 

With respect to dieting, 18.73% of participants were dieting at the moment, and 29% planned to diet in the future. Among those who planned to diet, 30.95% gave aesthetic reasons to do it, 57.14% of them being girls. Considering self-weighing frequency, the following percentages were obtained: several times a day (2.15%), once a day (3.38%), several times a week (5.54%), once a week (20.31%), and occasionally (66.77%). Only 1.85% of participants never self-weighed. The main reason to self-weighing was "controlling my weight" (67.71%), followed by "no fattening" (17.01%) and "feel better" (12.15%). There were no significant gender differences with respect to dieting and self-weighing frequency. Nevertheless, 57.14% of those who self-weighted to "feel better" were girls.

Body image perception was assessed by means of the body silhouettes method. Table 1 shows the results by sex with mention to the BMI linked (approximately) to each silhouette.

Table I. Body image perception by sex 

χ2 = 13.50; p < 0.05.

Bearing in mind the silhouettes which correspond to normal BMI (2,3,4), more girls (60.51%) than boys (51.42%) chose silhouettes 3-4. Overall, no gender differences with respect to body image perception were observed (Fig. 2).

Figure 2. Silhouettes chosen by boys and girls. 

Taking into account actual BMI and body perception, among those who chose silhouettes 2, 3 and 4 (these silhouettes are usually associated to normal BMI), 1.25% had moderate denutrition, 12.5% were overweight and 2.5% were obese in case of silhouette 2; with respect to silhouette 3, 0.95% had severe denutrition, 30.48% were overweight and 2.86% were obese; finally, in the case of silhouette 4, 35.90% were overweight and 20.51% were obese. As a result, when participants identify their body image with silhouette number 2, 16.25% was misperceiving their weight; when participants chose silhouette number 3, the percentage of misperception was 34.29; and, finally, considering the silhouette number 4 the percentage of misperception was 56.41.

With respect to physical fitness, among those who identified themselves with silhouettes 2-4, self-reported physical fitness was "average", "good" or "excellent" in 89.01%. When participants chose silhouette number 1, that percentage was 85.71%. Finally, the percentage was 56.86% when participants identified themselves with silhouettes 5-9. It must be noted that poor physical fitness was considered by 0.78% when participants chose silhouettes 2-4 and 11.76% in case of silhouettes 5-9. Nobody referred poor physical fitness in case of silhouette number 1. Due to the fact that silhouette number 8 has not been chosen and number 9 only was chosen by one participant, these two silhouettes have been removed from Figure 3.

Figure 3. Distribution of self-reported physical fitness by silhouettes. 

Means of eating behaviors as measured by means of TFEQ-Sp and EBBIT are shown in Table 2. It must be noted that the original purpose of the EBBIT was to test the hypothesis that young girls at risk of eating disorders may exhibit problems in several areas. Nevertheless, there are no psychometric reasons not to use this instrument in boys so for this work it was applied.

Table II. Means of TFEQ-Sp and EBBIT by sex 

EBBIT: Eating Behaviors and Body Image Test; TFEQ-Sp: Three-Factor Eating Questionnaire-R18, Spanish version; BIDRE: body image dissatisfaction/ restrictive eating; BEB: binge eating behaviors.

Considering different nutritional states (severe denutrition [SD], moderate denutrition [MD], normal weight [NW], overweight [OW], obesity [OB]), there were no significant differences with respect to restrained eating, emotional eating, uncontrolled eating and binge eating behaviors. Significant differences were found in case of body image dissatisfaction/restrictive eating (BIDRE) subscale (p < 0.0001). Tukey mean-differences test revealed that scores were higher in overweight/obesity than in normal weight (p < 0.0001) and scores were also higher in obesity than in overweight (p < 0.0001) (Fig. 4).

Figure 4. Differences in BIDRE scores considering several nutritional states (SD: severe denutrition; MD: moderate denutrition; NW: normal weight; OW: overweight; OB: obesity). 

Correlational analyses between scores on eating behaviors and BMI revealed a unique significant correlation between BIDRE and BMI (r = 0.47; p < 0.0001; 95% CI = 0.38-0.55), the rest being not significant. With respect to different subscales of EBBIT and TFEQ-Sp, BEB correlated negative a significantly with restrained eating (r = -0.23; p < 0.0.1), uncontrolled eating (r = -0.18; p < 0.01) and emotional eating (r = -0.32; p < 0.01). No significant correlations were found between BIDRE and uncontrolled eating and restrained eating. BIDRE correlated positively with emotional eating (r = 0.19; p < 0.05).

DISCUSSION

Weight misperception may be associated to healthful or unhealthy behaviors ^(1.2.3). Misperception of overweight or obesity among adolescents at normal weight is a clear risk factor for eating behavior disturbances. The sequence weight misperception, body dissatisfaction and dieting is prone to develop negative eating attitudes and finally high risk for different pathologies, eating disorders being the most relevant among adolescents ⁶,⁷,⁸. It has been reported that females tend to perceive themselves as overweight more than males do ⁴,⁵,⁶. In the current study, despite no general differences were found considering the weight self-perception scale, it must be noted that among the participants who considered to be very overweight, 76.92% were girls. On the contrary, among those who perceived themselves as very or slightly underweight, 62.5% and 60% respectively were boys. These results are similar to others previously found in a similar study with a sample of adolescents with a mean age of 16.22, so a bit older ²⁸. Other studies have reported that females tend to perceive themselves as overweight more than men usually do ⁴,⁵,⁶. Generally, the identification with larger silhouettes (e.g., from 2 to 4) increases the percentage of weight misperception.

Body dissatisfaction has been considered as a mediator on the effect of BMI on perceived physical activity ¹², and body image is improved by changes in physical fitness ¹³. Recently, it has been shown that body dissatisfaction is a significant mediator of the effect of BMI on perceived physical activity ¹². A large body of research has aimed to validate the idea that exercise improves body image through changes in physical fitness ¹³. However, Martin and Lichtenberger have suggested that improvements in physical fitness play a minor role in changing body image, because the effects of physical exercise and activity on body satisfaction should be mediated by changes in individuals' perceptions of their physical fitness and competence ¹⁵. This study shows that most participants reported an average or good physical fitness (72.84%). Again, some gender differences appeared since 57.40% of boys reported good and excellent physical fitness, this percentage being 40.52% in case of girls. This difference has been reported previously ²⁸. In view of our results, boys tend to have a better perception of their physical fitness than girls, this result appearing at normal weight, overweight and obesity.

A worse self-reported physical fitness (regardless of actual weight) and a tendency to perceive themselves as overweight could lead girls to higher risk of developing body dissatisfaction. In addition, it must be noted that almost 31% of participants who planned to diet in the future had aesthetic reasons to do it, 57.14% of them being girls. The main reason for self-weighing was "controlling my weight", but when the reason was "feel better", almost 58% of participants were girls. In this regard, a worse self-reported physical fitness, a worse weight perception, aesthetic reasons for planning to diet in the future and considering self-weighing as a way to feel better could yield a constellation of risks to develop both body image dissatisfaction and eating behavior disturbances. In fact, a previous study has reported that dieting for aesthetic reasons, weight misperception, worse self-reported physical fitness and the fact of being female perform a high-risk group of developing eating disorders ²⁸.

Healthy or unhealthy behaviors as a consequence of different perceptions and their corresponding psychological and psychosocial effects lead to the concept of emotional eating, among other eating behaviors. Thus, it has been distinguished among restraint eating (conscious restriction of food intake aimed to control body weight and/or to promote weight loss), uncontrolled eating (inability to resist emotional cues, eating as a response to different negative emotions) and emotional eating (tendency to eat more than usual due to a loss of control over intake with a subjective feeling of hunger) ¹⁴. Other authors have defined external eating as the tendency to overeat in response to external food-related cues like the sight, smell, and taste of palatable food, regardless of their physical need for food ¹⁵,¹⁶. Moreover, some authors have noted that different types of bingers and dieters may be found: bingers who are engaged in restraint-induced binging, and bingers generally disinhibited; dieters who eventually become disinhibited and overeat, and dieters who maintain the restrictive attitude ¹⁷,¹⁸. Also, in the field of eating behaviors, the concept of dietary restraint is relevant, highlighting the regulation of food intake in order to control weight and body shape ³². That control based on restrictions may cause consequent overeating episodes and eating disorders, and overweight and obesity at long-term ¹⁴. Along with dietary restraint, other eating behaviors have been described such as loss of control over intake and overeating as a consequence of emotional distress ³³.

It has been reported that BMI and negative emotional eating are highly related whereas positive emotional eating and external eating loaded onto another factor. In this regard, it is plausible that even though positive emotions may elicit eating, they do not necessarily mirror disordered eating. Eating in response to positive emotions might rather be related to hedonic or external eating ¹⁹,²⁰.

Some models conceptualize eating disturbances as disorders of affect regulation, considering the impairment in the cognitive capacity to process and regulate emotions as the primary regulatory disturbance ²¹. According to these models, some eating behaviors, such as binge eating and compensatory behaviors, as well as restricted food consumption, are interpreted as responses to cope with intense or relatively undifferentiated emotional states ²²,²³-²⁴. These intense emotional states are usually linked to self-perceptions (weight, physical fitness, body shape/body image), especially among adolescents (25-28). Thereby, it has been reported that people with higher weight and those who perceive themselves as overweight usually show higher scores on cognitive restriction ³¹. In fact, the link between higher BMI and higher scores on cognitive restriction has been reported previously ³⁴,³⁵. In the current study we have not found any significant differences with respect to restrained eating, emotional eating, uncontrolled eating and binge eating behaviors when the actual weight has been considered. In this regard, as other studies have shown, cognitive dietary restraint is not consistently linked to body weight-adiposity ²⁹. Nevertheless, scores on body image dissatisfaction/restrictive eating (BIDRE subscale) were different with regards to actual weight (normal weight, overweight and obesity). The fact that the difference is showed in this subscale but not in the BEB subscale indicates that it is possible to maintain a tendency to restrictive attitudes without binge eating behaviors in both obese and overweight participants. With respect to the association, the correlation between BIDRE and BMI was 0.47. In fact, BIDRE and BEB suggest that these two factors might be considered as independent dimensions ¹⁸.

Another point to discuss refers to the instruments which aim to assess eating behaviors. Thus, BEB correlated negative a significantly with restrained eating, uncontrolled eating, and emotional eating. This result seems to indicate that binge eating behavior, restrained eating, uncontrolled eating and emotional eating are different constructs despite having possible shared elements. In addition, no significant correlations were found between BIDRE and uncontrolled eating and restrained eating. It seems that when there is a component linked to body image (e.g., BIDRE), correlations with uncontrolled and restrained eating are not significant but this changes with respect to emotional eating, in this case existing a positive correlation. In this regard, body image dissatisfaction could lead to eating disturbances through emotional more than restrained eating.

This study adds some new results to others previously published in the same field ²⁵,²⁶,²⁸,³². Comparing to these others, we have studied a sample with the lowest age range which permits to study possible links between weight self-perception, body image (and related variables such as dieting or self-weighing) and eating behaviors in a stage of life which could be recognized, to some extent, as a starting point to develop eating disorders. Some conclusions emerge from the results. First, the majority of participants who perceive themselves as obese are girls and the majority of participants who perceive themselves as underweight are boys. Second, boys tend to perceive themselves with a better physical fitness than girls. Third, more girls than boys plan to diet in the future for aesthetic reasons. Fourth, when the reason to self-weighing is feel better, more girls than boys are involved. Finally, BMI is significant and positively correlated to BIDRE. When body image dissatisfaction is controlled for analysis, then BMI correlates with emotional eating.

In view of these results, it would be interesting to study in depth the reasons that lead girls to overestimate their weight and, on the contrary, underestimate it in case of boys.

The current study has some limitations. Different variables are self-reported at an age which could reflect doubts about reliability. Nevertheless, previous studies have followed a similar way to assess some information. Despite that EBBIT was designed for young girls at risk for eating disorders, after analyzing the content of the test no psychometric reasons have been found to avoid that instrument in the current study. Body dissatisfaction has not been assessed by means of a specific instrument apart from the BIDRE subscale of the EBBIT. The body silhouettes method is based on self-reporting where participants must choose the silhouette that most closely resembles the shape of their body. Then, it is possible to analyze differences with respect to self-reported weight or actual weight as well as others measures. In this study, we emphasized some self-reported measures more than body image dissatisfaction scores as usually are measured by several questionnaires. Finally, precocious puberty is a risk factor clearly related to the onset of that puberty. Secondary sexual characteristics may lead to affective and psychosocial adaptive problems, which also lead to behavior disorders and a negative body image. Pubertal timing is a potentially significant factor when assessing psychopathological symptoms. Pubertal timing refers to the timing when pubertal development occurs in relation to peers, i.e., it relates to whether an adolescent is ahead of peers in pubertal development (early pubertal timing), in line with peers (on-time) or behind peers in pubertal development (late pubertal timing). In this regard, not having applied the pubertal stage categorization (e.g., by means of Tanner stages) is a limitation to take into account in future similar studies ³⁶.