Body image is a multi-dimensional concept, which describes how we think, feel, perceive, and act with regard to our bodies. Adolescence constitutes a critical period for the development of a healthy or unhealthy body image [1]. A large number of studies have consistently shown that a negative body image, typically measured as body dissatisfaction, is associated with disturbed eating patterns among adolescents [2–6] and one of the strongest risk factors for the development of eating disorders (ED) [7, 8] and other adverse psychological outcomes such as depression [9–11].

Body image disturbances are key characteristics of eating disorders (EDs) such as anorexia nervosa and bulimia nervosa and encompass distortions in cognition, affect, perception, or behavior related to body weight or shape [12]. They may refer to negative thoughts or negative evaluation regarding one’s own body, negative affect in response to one’s own body, misperception of body-related stimuli, and specific body-related behaviors (e.g., checking or avoidance). In Western societies, body image disturbances including body dissatisfaction are pervasive problems. Particularly among women, the desire for thinness is so prevalent that it is considered a normative discontent [13]. A growing body of evidence suggests that this body-related discontent may apply to a similar extent to children and adolescents. A large number of studies has shown that body image disturbances (e.g., body dissatisfaction, discrepancy between one’s actual and one’s ideal body size, weight and shape concerns) frequently occur even before puberty and are reported by up to 50% of children and adolescents [5, 7, 14–23].

Similarly, a growing body of research suggests that symptoms of disturbed eating behavior are common among youth. In a large German study among 7498 students (11–17 years old), nearly one quarter (21.9%) showed symptoms of EDs (e.g., concerns about loss of control over eating, self-induced vomiting, rapid weight loss in the last 3 months). Girls were significantly more often affected than boys (28.9% vs. 15.2%) [24]. Similarly, a study conducted in the United States among 1739 female students (12-18 years) reported that disordered eating attitudes and behaviors (e.g., dieting, binge eating) were present in 27% [25]. Similar numbers have been reported by other studies [17, 20, 21, 26, 27]. The outcomes of eating-disordered attitudes and behaviors in adolescence are severe. Prospective studies show that body dissatisfaction and early ED symptoms (e.g., body image distortions, weight concerns) predict eating-disordered behavior, onset of ED, depressive symptoms, overweight, and obesity in adulthood [3, 26, 28, 29].

While there is consistent evidence that body image disturbances in terms of dysfunctional cognitions (e.g., body dissatisfaction), negative affect (distress in response to weight or shape), and behavioural measures (e.g., symptoms of disturbed eating behavior) already appear in children and adolescents, few data is available regarding body image perceptions. Recent studies have used pictorial figure rating scales to examine body image perceptions, which typically consist of a series of abstract figures ranging from underweight to overweight (for an overview, see [30]). Up to this point, only a handful of studies have employed figure rating scales displaying naturalistic human bodies [31–35] and only one of these studies has been conducted among children [35]. While this study reported discrepancies between children’s own body image and ideal body image, normative perceptions of human body sizes (e.g., the ability to correctly perceive human bodies in terms of normality) have not been investigated.

Perceptual distortions may play an important role in the development of EDs [8, 36]. Perceptual distortions are considered a type of cognitive bias, which describe systematic errors in the processing of information (i.e., information processing biases). There is accumulating evidence that cognitive biases may influence the onset and maintenance of eating-related pathology in adolescence and early adulthood [37–42]. Cognitive biases may occur in different domains such as attention, perception, or memory and may foster symptoms of mental disorders, because they determine what people notice, attend to, and remember. In ED, perceptual biases related to body weight or shape (e.g., systematic misperceptions or judgement errors) have been proposed to reinforce disturbed body image experiences [43]. For example, underestimating the average body size may result in a larger perceived discrepancy between oneself and the norm, thereby increasing body dissatisfaction and weight and shape concerns.

The present study aimed to examine normative perceptions (perceived average body size) and thin-ideal perceptions (perceived ideal body size) of female bodies¹ among 11–17 year old children and adolescents using a naturalistic photographic figure rating. Furthermore, symptoms of disturbed eating behavior were studied in relation to these perceptions. We hypothesized that children and adolescents would systematically underestimate the average female body size in comparison to the average statistical body size. We also expected that children and adolescents would display a slim female thin-ideal. In addition, we expected that symptoms of disturbed eating behavior would be associated with a bias towards underestimation of female body size and an underweight thin-ideal.

The present study aimed to answer the question how accurate children and adolescents judge body sizes of young females in terms of normality and if there is a general bias towards underestimation of female body size among youth. Using a photograph-rating consisting of sets of naturalistic photographs of young women’s bodies, body image perceptions (i.e., perceived average female body size and perceived ideal female body size) were examined in a large sample of 11–17 year old German students.

The present study is the first to show that children and adolescents considerably underestimate the average female body size when judging naturalistic photographs of young female bodies. On average, they underestimated the average body size of a young woman by more than two BMI-points (i.e., they perceived the average BMI of a young woman to be approximately 20, while the average BMI of the reference population is 22.4). Perceptual biases such as normative misperceptions have been found to play an important role in several health-related behaviors such as uptake of smoking or drinking among youth [52, 53]. Similarly, perceptual body-related distortions may influence eating-related attitudes and behaviors by increasing the perceived discrepancy between oneself and the norm, resulting in body dissatisfaction and weight and shape concerns. Research supports these assertions by showing that women who felt discrepant from the norm show more symptoms of ED [54], which may results in more extreme and maladaptive dieting behaviors to achieve an unrealistic and often unattainable body size.

Furthermore, the present study showed that girls and boys generally held a slim female thin-ideal (i.e., they perceived the ideal BMI of a young woman to be approximately 19.5), which represents the lowest quartile of a healthy BMI range (18.5–25). Yet, a substantial proportion of children and adolescents displayed an underweight thin-ideal (24.9% among girls, 16.8% among boys). The results are in line with previous studies. Connolly, Slaughter, and Mealey [55] showed that already 6-year olds have a systematic preference for underweight body shapes. Similarly, Brown and Slaughter [15] showed that children and adolescents across all age groups rate thin female bodies as more attractive than normal bodies. Schneider and colleagues [21] showed that adolescent girls desired a body shape for themselves corresponding to underweight. Similar strong thin-ideals have been observed in adult women [56–59]. In sum, a large body of research indicates that the sociocultural thin-ideal is internalized by a large proportion of the Western population including children and adolescents. The results of the present study strengthen and extend findings of previous studies using pictorial instead of photographic figure rating scales, which may be limited by methodological shortcomings.

Finally, the present study showed that symptoms of disturbed eating behavior among youth were quite common, especially among female adolescents. Feeling fat, feeling upset about weight or shape, restrictive eating, exercising for weight control, and distress after eating were reported by a quarter to a third of all children and adolescents. Also, a substantial proportion of youth reported unhealthy eating behaviors such as skipping meals or fasting (21.8 and 15.3%, respectively), episodes of binge eating (16%), and perceived loss of control over eating (11%). The results are in line with previous research showing that symptoms of disturbed eating behavior are common among youth [17, 20, 21, 24–27]. Importantly, body image perceptions were associated with disordered eating behaviors among youth. Girls who displayed a strong bias towards underestimation of the average female body size and girls who displayed an underweight thin-ideal were more likely to report harmful dieting behavior (e.g., skipping meals, self-induced vomiting) and psychological distress associated with eating and own body weight (e.g., being terrified of gaining weight, feeling upset about weight or shape, distress after eating). Also, they showed significantly elevated scores on well-established measures indexing risk for ED (i.e., higher levels of shape concerns and a stronger recognition and endorsement of societal appearance standards). These associations indicate that both perceptional biases as well as the internalization of a pervasive thin-ideal may constitute risk factors for the onset and maintenance of ED among youth.

In addition, differences between boys and girls were examined. It is reasonable that both boys and girls hold body images, not only for their own but also for the opposite sex (i.e., ideas about how males and females should look like). With regard to the perceived average female body size, boys and girls did not differ (both underestimated the average female body size to a similar extent). However, with regard to the perceived ideal female body size, girls showed a slightly lower thin-ideal than boys. Previous studies found similar results among adults, showing that men and women differ in attractiveness ratings of female body size, with males being less stringent about female body size than females [60–62]. However, it should be noted that the present study only examined the female body ideal (i.e., thin-ideal), while the male body ideal (i.e., muscular ideal) has not been examined. Therefore, it remains unclear whether females in general are more susceptible than males to adopt and internalize sociocultural body ideals or whether females and males internalize gender-specific sociocultural body ideals to a similar extent. For a comprehensive picture, body ideals of both male and female bodies should be compared between boys and girls.

In addition, differences between developmental groups were examined. Interestingly, the group of early-adolescent girls most often displayed an underweight thin-ideal. Nearly one-third of 13–14 year-old girls perceived a BMI below 18 (underweight) as ideal body size, possibly indicating that early adolescence may constitute a vulnerable developmental period for the onset of disordered eating-related cognitions and attitudes. A potential explanation may be that girls within this developmental phase typically start to experience changes in body composition (i.e., increase in body fat starting with puberty), after a period of typically having a relatively lean body during childhood, which may make this group particularly susceptible for a fear of body fat and the internalization of a pervasive thin-ideal. A general fear of growing or a fear of gaining secondary sex characteristics may also play a role during period and may explain the adoption of a pervasive thin-ideal among early-adolescent girls. With regard to symptoms of disturbed eating behavior, 15–17 year-old girls seemed to be most vulnerable. The results reflect age differences in the onset of different ED. The onset of anorexia (characterized by underweight or severe weight loss) typically lies in early adolescence and the onset of bulimia (characterized by disturbances in eating behavior such as binge eating and inappropriate compensatory behaviors) in late adolescence [63]. The results may reflect a developmental time course, in which cognitive-attitudinal distortions (e.g., adoption of pervasive female thin-ideal) in early-adolescence precede the onset and manifestation of symptoms of disturbed eating behavior during adolescence.

Several limitations should be acknowledged. First, the study has been conducted in a single state of Germany. Although North-Rhine Westphalia is Germany’s most populous state, the findings may not be entirely generalizable to the national population level and do not consider culture-related differences in body perceptions and body ideals. Moreover, body image perceptions and symptoms of disturbed eating behavior were self-reported by youth. It is possible that social desirability or response styles may have influenced the results. In addition, the cross-sectional design of the study does not allow to draw conclusions regarding temporal precedence or causality between study variables. While it is intuitive to assume that perceptual distortions precede the development of symptoms of disturbed eating behavior, it is also possible that children and adolescents with disturbed eating behavior develop perceptual distortions as a correlate of eating-related pathology. Moreover, it should be noted that the present study used single items to measure symptoms of ED, which may have limited psychometric properties. Also, the items did not assess the clinical severity of symptoms of disturbed eating behavior, as no clinical rating nor measures of frequency and severity were applied. In addition, it should be noted that the psychometric validity of the photographic figure rating has not been fully established. Yet, an expert-rating among mental health professionals indicated construct validity and previous studies have shown good test–retest validity and convergent validity of similar photographic figure rating scales [31–34]. Finally, it should be acknowledged that the present study did not control for a general underestimation bias. A body of research suggests that individuals tend to display under- instead of overestimation when asked to make judgements regarding size (e.g., when judging package or portion sizes, cf. Ordabayeva & Chandon [64]). Therefore, underestimation biases may constitute normative, hardwired cognitive errors, at least to a certain extent. The present study, however, shows that a strong bias towards underestimation of body size is associated with symptoms of disturbed eating behavior and psychological distress, indicating that strong perception biases are qualitatively different from common, benign errors. The present study also has several strengths including a large, heterogeneous sample of children and adolescents from all school types in Germany’s most populous state. In addition, the photographic rating, consisting of a variety of real women’s bodies, may have a better ecological validity in the assessment of body image perceptions than figure ratings used in previous studies. As the present rating used a larger number of female body photographs, the risk that a particular confounder was associated with a particular body size is decreased.

The present study suggests several recommendations for future research. First of all, prospective study designs are required to enable conclusions regarding temporal order to improve our understanding of the development and maintenance of ED. Future research may disentangle whether perceptual distortions constitute a risk factor predisposing youth towards the development of ED or merely a symptom of the ED. Furthermore, a better understanding of the frequency and the severity of symptoms of disturbed eating behavior among children and adolescents would be valuable. Future studies may investigate how often these symptoms are experienced by youth and whether they are associated with clinically significant distress or functional impairment. Finally, it would be interesting to investigate if perceptual distortions and symptoms of disturbed eating behavior can be modified by interventions. Possibly, psycho-education and cognitive interventions to modify normative misperceptions and perceptions of the thin-ideal may help to reduce eating-related pathology and prevent the development of ED among youth.

In conclusion, the present study demonstrates that children and adolescents display a considerable perception bias (i.e., bias towards underestimation of female body size). Also, this study suggests the existence of a developmental time course, in which perceptual body-related distortions (e.g., body-related perception biases, internalization of pervasive thin-ideal) in early-adolescence may precede the onset and manifestation of symptoms of disturbed eating behavior during the course of adolescence. However, prospective studies will need to clarify temporal precedence between perceptions, cognitions, and behavior associated with eating-related pathology among youth in the future.