Binge eating

A binge eating episode consists of eating in a short time a larger amount of food than most people would, with a perceived loss of control [1]. Feelings of guilt, shame and distress characterize the symptom, due to the perceived uncontrollability of eating and its consequences [2]. Binge eating episodes are common in different eating disorders [3], often associated with psychological distress, mental diseases and other clinical conditions [4], as reported in the DSM-5 [1]. Overweight and obesity are often associated to binge eating symptoms [2, 3]. Even if dietary and food restriction can be present, in the diagnostic criteria of binge eating disorder no regular food compensatory behaviors—such as fasting—are requested, because they are not as frequent as in anorexia and bulimia nervosa [1].

The risk of onset of binge eating increases during late adolescence [5], and higher rates are found among girls than among boys [6]. Binge eating episodes during adolescence are also a risk factor for the subsequent onset of other eating disorders [6]. The most common antecedent of binge eating episodes is a negative feeling [7]. According to the affect regulation model [8] binge eating is a dysfunctional strategy for coping with negative emotions, providing a temporary relief in absence of more adaptive emotion regulation mechanisms. In line with this perspective, research investigated emotional and cognitive predictors of binge eating, enlightening the role of emotional deficits and compromised cognitive functions, such as inhibition of disorder-related stimuli, on binge eating disorder [9]. A combination of negative emotions—such as feelings of depression and anxiety—and cognitions—such as negative evaluations about oneself—was found to be antecedent of binge eating episodes [10].

Emotional functioning in binge eating

Emotional functioning has two components: emotion regulation, which refers to the capacity of modulating internal states, in accordance with environmental demands, and emotional awareness, that is the ability to recognize and differentiate feelings and emotions [9]. Both the components have been found to be impaired in binge eating [11]. Specifically, binge eaters showed difficulties in impulse control [12], adopted maladaptive emotion regulation strategies, such as emotional suppression [13], and reported difficulties in identifying own feelings [9, 11, 14].

Binge eating has been conceived as an attempt to escape from the awareness of negative internal states focusing on food as an immediate stimulus [15]. The same pattern was found also for avoiding intense positive arousal [16, 17], suggesting an underlying incapacity in facing up to all intense feelings. However, the causal effects of emotion dysregulation on binge eating have not been confirmed [7], whilst recent studies provided evidence of an interaction among cognitive and emotional processes as antecedents of binge eating [5, 10]: Cognitive biases on food indeed influenced suppression and desire thinking in prediction of binge eating [10], and a combination of negative beliefs about oneself and depressive feelings was found to be the trigger of binge eating episodes [5].

Metacognition in binge eating

Metacognition is implied in controlling, modifying and interpreting thoughts [18]. The metacognitive model identified five dysfunctional processes based on negative and positive evaluations of worry and rumination, on attempts to monitor and suppress thoughts, and on low confidence in own cognitive functions [18]. These metacognition processes lead to maladaptive cognitive responses and ineffective coping strategies, contributing to psychological distress [19].

Dysfunctional metacognition was found to be related to eating disorders [20], throughout two dysfunctional thinking styles: worry and rumination [21]. However, among the specific metacognitive processes, only the dimension “need to control thoughts” characterized eating disorders consistently in most of the studies: specifically, it was found to be associated with both anorexia and bulimia nervosa [20,21,22,23].

The need to control thoughts describes obsessive attempts to monitor thinking to avoid negative consequences [24], and was considered an expression of the perceived difficulty in controlling internal states [22]. Both the low perceived control and the striving to control thoughts were found related to eating disorders, suggesting that the lack of control on eating behaviors could be an expression on a general impairment in processes related to control [22].

Aims and hypotheses

The present study aimed to investigate the effects of emotion dysregulation, lack of emotional awareness, and metacognition on adolescent binge eating, and the interaction effects among emotional and metacognitive processes. There is a lack of research indeed about metacognition processes specifically related to binge eating, although it is a common symptom for different eating disorders, and can help in understanding their functioning [3]. Moreover, the interaction effects among metacognition and emotional functioning on binge eating have not yet been investigated.

The need to control thoughts, conceived as a cognitive expression of underlying difficulties in coping with distress, is the metacognitive process most frequently reported in eating disorders [22]. Regarding emotional functioning, the lack of emotional awareness is a well-documented predictor of binge eating [11]. Its role in binge eating could be explained because unknown internal states are perceived as out of control [25], thus behavioral responses are necessary. Binge eating indeed is a behavioral response for avoiding intense internal moods perceived as distressing, with the result of escaping from the awareness of them [16, 26, 27]. Thus, we specifically aimed to verify the interaction effect among the need to control thoughts and lack of emotional awareness, because they could be the cognitive and emotional facets of a common inability to manage internal states.

To our knowledge, this is the first study that aimed to investigate the interaction effects between metacognition and emotional functioning on binge eating adolescents. We hypothesized that the need to control thoughts could be a moderator in the relationship between the lack of emotional awareness and binge eating.

Method

Participants and procedure

For this research, 982 participants from middle to late adolescence were initially recruited in public high schools of Rome. Participants gave their informed consent before participating in the study, and for underage students, written informed consents were obtained also from parents. Data collection was conducted via paper and pencil questionnaires administered at school, under the supervision of trained interviewers. However, 18.1% of the recruited students did not correctly complete some of the self-report instruments, and consequently they were excluded from the data analyses. Thus, the final sample of this study was composed of 804 Italian adolescents (49.7% females) with a mean age of 17.45 (SDage = 1.02; age range, 15–20). Participants were divided into two age groups: 15–17 years old (n = 323, 47.4% females; Mage = 16.38, SDage = 0.64); and 18–20 years old (n = 481; 51.4% females; Mage = 18.17, SDage= 0.44). This study and its procedure were approved by the Ethics Committee of the Department of Social and Developmental Psychology, Sapienza University of Rome.

Measures

Individual data The participants reported their gender, age, height and weight. Subsequently, a BMI score was computed for each participant.

Binge eating The Binge Eating Scale (BES) [28] is a self-report measure composed of 16 items assessing behavioral, emotional and cognitive symptoms of binge eating (e.g., eating large amounts of food in a short time; guilt and fear of being unable to stop eating). For each question, there are three or four weighted answer options referring to different levels of severity of binge eating. The total score of the scale can range from 0 to 46. The scale showed good reliability and validity in the Italian context [29]. In the current study, the Cronbach’s alpha was 0.85.

Emotion regulation The Difficulties in Emotion Regulation Scale (DERS) [30] is a self-report questionnaire composed of 36 items rated on a 5 point-Likert scale from 1 (almost never) to 5 (Always). The DERS evaluates emotion dysregulation either on a total score or on six subdimensions. The scale showed good psychometric properties in the Italian context [31].

As suggested by Bardeen et al. [32] in the present study, we considered the subscale Lack of emotional awareness (6 reverse-scored items; sample item: “I pay attention to how I feel”; Cronbach’s alpha of 0.67) as an independent measure, whereas emotion dysregulation was computed on the total score of five DERS subscales (30 items; Cronbach’s alpha of 0.93).

Metacognition Metacognition was assessed with the Metacognition Questionnaire short form (MCQ-30) [18], a self-report measure composed of 30 items ranged on a 4 point-Likert scale from 1 (disagree) to 4 (agree). The MCQ-30 evaluated dysfunctional metacognitive processes on five dimensions: positive beliefs (6 items; sample item: “I need to worry in order to work well”; Cronbach’s alpha of 0.76); uncontrollability and danger (6 items; sample item: “When I start worrying I cannot stop”; Cronbach’s alpha of 0.75); cognitive confidence (6 items; sample item: “I do not trust my memory”; Cronbach’s alpha of 0.85); need to control thoughts (6 items; sample item: “I should be in control of my thoughts all of the time”; Cronbach’s alpha of 0.64); cognitive self-consciousness (6 items; sample item: “I am constantly aware of my thinking”; Cronbach’s alpha of. 75). The scale has adequate psychometric properties in the Italian context [33].

Data analysis

A factorial ANOVA was run on binge eating, entering gender and age groups as between-subject factors. Correlations among all study variables were then computed. Finally, a moderation regression analysis was conducted to test the hypothesis of the moderation role of the need to control thoughts in the relationship between lack of emotional awareness and binge eating, controlling for the independent effects of gender, BMI, emotion dysregulation, lack of emotional awareness, and metacognition dimensions. Following the suggestion of Cohen et al. [34], the independent variables (BMI, emotion dysregulation, lack of emotional awareness, positive beliefs, uncontrollability and danger, cognitive confidence, need to control thoughts and cognitive self-confidence) were mean centered in the regression equation (except gender which was dummy coded, see Table 2). Then, the centered scores of lack of emotional awareness and need to control thoughts were multiplied to form the interaction term. The moderation effect was then tested in different steps using a hierarchical regression analysis. Considering binge eating as dependent variable, in the first step, gender and BMI were entered as covariates. In the second step, emotion dysregulation and lack of emotional awareness were regressed on the criterion. In the third step, positive beliefs, uncontrollability and danger, cognitive confidence, the need to control thoughts and cognitive self-confidence were regressed on the criterion. In the fourth step, the interaction term was tested.

Since a significant interaction effect was detected, a simple slope analysis was conducted by plotting the predicted values of binge eating as a function of lack of emotional awareness for two different levels of the moderator (need to control thoughts): low level conventionally is represented by one standard deviation below the mean (− 1 SD), while high level is represented by one standard deviation above the mean (+ 1 SD) [34].

Results

Gender and age differences on binge eating

Preliminary analyses verified the normal distribution of the study variables in our sample. The factorial ANOVA detected a significant gender difference on binge eating, F(1, 803) = 53.18, p < .001, η2 = 0.06 with girls showing significantly higher means than boys. Conversely, neither age difference, F(1, 803) = 0.01, p = .90, η2 = 0.00, nor gender–age interaction effect, F(1, 803) = 0.42, p = .51, η2 = 0.00,were found.Footnote 1

Correlations between study variables

The correlations between age, gender, BMI, emotion dysregulation, lack of emotional awareness, metacognition dimensions and binge eating are reported in Table 1. Note that the emotional functioning measures were significantly correlated with many of the metacognitive measures. Emotion dysregulation was negatively correlated with cognitive self-consciousness and was positively correlated with positive beliefs, uncontrollability and danger, cognitive confidence, and need to control thoughts. Similarly, lack of emotional awareness was also negatively correlated with cognitive self-consciousness and positively correlated with cognitive confidence.

Table 1 Correlations among study variables and descriptive statistics
Full size table

Regression moderation analysis

The variance inflation factor scores (VIF) ranged from 1.00 to 1.49, showing no problems with multicollinearity. In the first step of the hierarchical regression analysis, gender and BMI were entered as covariates, accounting for 10.4% of the variance. Both gender and BMI emerged as significant predictors, with females reporting more binge eating than males. In the second step, 26.5% of the variance was accounted for, with a significant increment of 16.1% in the explained variance. Gender and BMI were still significant covariates, but—controlling for their effects—also emotion dysregulation and lack of emotional awareness turned out to be significant predictors of binge eating. In the third step, 28.9% of the variance was accounted for, with a significant increment of 2.3%. Gender, BMI, emotion dysregulation and lack of emotional awareness were still significant predictors, but—controlling for these effects—cognitive confidence and the need to control thoughts, also turned out to be significant predictors of binge eating. Finally in the fourth step, the interaction term between lack of emotional awareness and need to control thoughts was added to the model.Footnote 2 Gender, BMI, emotion dysregulation, lack of emotional awareness, cognitive confidence and the need to control thoughts were still significant predictors, but—controlling for these effects—the interaction term was significant, adding a significant 1% to the explained variance. Overall, the final model explained 29.4% of the variance (see Table 2).

Table 2 Moderation effect of the need to control thoughts on the relationship between lack of emotional awareness and binge eating
Full size table

The simple slope analysis found a significant positive relationship between lack of emotional awareness and binge eating at high levels of need to control thoughts (+ 1 SD) (β = 0.18, p < .001), whereas at low levels of need to control thoughts (− 1 SD), the same relationship became not significant (β = 0.02, p = .62). Thus, the need to control thoughts increased the impact of lack of emotional awareness on binge eating, working as an additional risk factor: specifically, when the need to control thoughts is low, the presence of lack of emotional awareness did not predict binge eating at all. Conversely, only in presence of high need to control thoughts, the lack of emotional awareness predicted binge eating (see Fig. 1).

Fig. 1
figure 1

Simple slopes analysis of the moderating role of need to control thoughts in the link between lack of emotional awareness and binge eating

Full size image

Discussion

This study investigated emotional and metacognitive processes in binge eating adolescents, exploring the effects of gender, age, BMI, emotion dysregulation, lack of emotional awareness, and dysfunctional metacognition dimensions on binge eating. Moreover, we investigated the moderating role of the need to control thoughts on the relationship between lack of emotional awareness and binge eating, controlling for the independent effects of the other variables.

As regards gender and age differences on binge eating, our findings are consistent with literature on eating disorders [26, 27]. About gender differences, our study detected higher binge eating scores for girls, confirming previous studies in which women more frequently than men reported emotional eating, and cognitive and emotional symptoms of binge eating [35, 36]. As expected, binge eating did not differ by age, confirming the evidence that during adolescence the age of onset of the symptom is related to other contextual variables [27]. Our findings also showed that BMI positively predict binge eating, confirming studies on the specific impact of this symptom on body weight [37].

Binge eating was also positively predicted by emotion dysregulation, in accordance with previous findings [37]. Binge eating is a behavioral response to negative emotions, in absence of adaptive emotion regulation strategies, suggesting that emotion dysregulation contributes to the onset and maintenance of the symptom [8, 11]. Also, the lack of emotional awareness predicted binge eating, as expected in line with previous literature [11, 14]. As explained by Fairburn et al. [16], during binge eating episodes the awareness of distressful internal states is avoided by narrowing attention to an immediate environmental stimulus, such as food [15].

About metacognition dimensions, our findings showed that low cognitive confidence and the need to control thoughts predict binge eating, supporting the studies about dysfunctional metacognition as a vulnerability factor for eating disorders [23, 38]. Previous studies indeed demonstrated that negative metacognitive processes predict the severity of binge eating disorder on clinical samples [38]. According to Wells [39], dysfunctional metacognition entails worry and rumination, two maladaptive thinking processes related to eating disorders [40]. Specifically, low cognitive confidence implies constant concerns on own memory and attention [22], and interferes with finding positive solutions for managing distress, eventually increasing worry and rumination [39]. The need to control thoughts instead was conceived as a consequence of worry and rumination, because when emotional symptoms increase, the individuals try to control their thinking [20]. Thus, this dimension can be considered the cognitive expression of underlying difficulties in dealing with internal distress [22].

Finally, our findings showed that the need to control thoughts is a moderator in the relationship between lack of emotional awareness and binge eating, highlighting the presence of an interaction between emotional and metacognitive processes. Specifically, individuals with poor emotional awareness appeared to engage in binge eating only in presence of high need to control thoughts, which worked as an additional risk factor. Conversely when the need to control thoughts was low, the lack of emotional awareness did not predict binge eating at all, thus suggesting the presence of underlying protective factors.

Conclusions

Overall, our results could suggest that the lack of emotional awareness and the need to control thoughts could be, respectively, the emotional and cognitive components of an underlying general impairment in facing up with internal states. The lack of emotional awareness could comport a relevant distress, because unknown internal states are perceived as out of control. However, this distress can be effectively faced in presence of adaptive metacognitive competences, leading the adolescents to be resilient to the development of disorders. According to Davis et al. [41], adequate metacognitive abilities provide effective strategies for managing emotions since childhood.

Conversely, when the lack of emotional awareness is associated with dysfunctional metacognition related to the need to control thoughts, the adolescents appeared to be lacking resources for managing internal states. The absence of both emotional and metacognitive resources could enhance the distress related to unknown internal states, leading the individual to adopt behavioral responses for avoiding them.

Limitations and implications for future research

Limitations of this study can be due to the cross-sectional design, which did not infer causal effects among variables, suggesting that future studies should replicate these results on longitudinal samples. Longitudinal research should follow a cohort of participants from teenage years to early adulthood, to verify the stability across time of this association between emotional and metacognitive difficulties, and binge eating. Moreover, in line with previous literature [9, 10] that suggested how emotional and cognitive impairments are antecedents of binge eating in clinical patients, future longitudinal studies should also control for the presence of different eating disorders symptoms, and should verify if the specific pattern of lack of emotional awareness/need of control thoughts, can also lead to the onset of binge eating disorder or other eating disorders during young adulthood. Another limit is due to the absence of information about bio-psycho-social information of participants: future research should control some aspects related to eating disorders, such as alteration of body image, food restriction and dietary habits, and other clinical and sub-clinical problems that could be associated to the symptom.

Although the study was not conducted with a clinical sample, according to previous literature [42], dysfunctional metacognition could be specifically intensified by psychiatric disorders, such as generalized anxiety disorder and obsessive–compulsive syndrome, as well as by all eating disorders. Moreover, there is evidence in literature that emotion regulation difficulties are enhanced by clinical levels of anxiety, post-traumatic stress, depression and borderline personality traits [43,44,45,46]. Thus, it is conceivable that the emotional/metacognitive pattern we found as predictor of binge eating, could vary in presence of other symptoms able to alter emotional and metacognitive processes.

A further limitation is that participants were from a single geographical context, suggesting the need to replicate these findings across different cultural samples. There is evidence in literature indeed that cultural context can affect the expression and regulation of emotions [47], and the vulnerability to eating disorders [48]. Finally, the self-report instruments may be influenced by individual interpretation, and cognitive and emotional symptoms could have been subjectively evaluated. Future studies should replicate these results using mixed measurement models.

In conclusion, the present study shed light on the emotional and metacognitive patterns that predict adolescent binge eating, addressing the interaction effect between lack of emotional awareness and need to control thoughts, as a risk factor for the development of the symptom. Our findings could contribute to deeply understand the emotional and cognitive correlates of binge eating at an early stage of onset of the symptom, providing implications for research, prevention and treatment. Understanding the emotional and metacognitive functioning leading adolescents to binge eat could support an early screening for at-risk individuals, where poor emotional awareness and poor metacognitive abilities request specific attention. Prevention programs targeted to at-risk adolescents should be focused on strengthening emotional awareness and emotion regulation, and should support the development of adaptive metacognitive processes. Both prevention and treatments for adolescent binge eating could be focused on the metacognition model [39], providing specific attention to the tendency to control thoughts. Clinical and evidence-based research should verify the possible positive effects of strengthening emotional awareness and metacognitive functioning in at-risk adolescents.