**1. Introduction**

Emotion regulation (ER) is defined as the sum of techniques applied to manage the variety, intensity, and duration of emotions [1]. Such strategies range from the putatively less adaptive, such as dissociation, avoidance, or suppression, to the supposedly more adaptive, e.g., cognitive reappraisal or problem-solving. Difficulties in ER are a transdiagnostic feature among multiple mental disorders and may explain high comorbidity rates (e.g., with anxiety, depression, or borderline personality disorder) [2]. Accordingly, ER is proposed as a transdiagnostic target for treatment [3].

ER allows one to cope with aversive emotions, is a core feature of self-regulation, and has a profound influence on food intake behaviors [4]. Difficulties in ER are present across all types of eating disorders (ED) [5–9]. In some studies, anorexia nervosa (AN) and bulimia nervosa (BN) do not seem to significantly differ with regard to most domains of ER. Patients with binge-eating disorder (BED) show less severe ER difficulties than patients with AN or BN [5,10], although there are also studies claiming that patients with binge-eating episodes (BED, BN, and AN/binge-eating purging subtype) present more ER difficulties compared to patients with AN/restrictive subtype [11] and others have shown no differences across ED types [12]. Nevertheless, ED are associated with other behaviors linked to ER difficulties, such as substance abuse and self-harm [9,13–15].

In addition, it is unclear to which degree difficulties in ER in ED may be seen as etiopathogenetic/vulnerability or as a maintenance factor contributing to the perpetuation of the disorder. In AN, starvation and low body weight reduce the susceptibility for emotional stimuli in the short-term and are thought to serve as dysfunctional strategies to regulate aversive emotions [16,17]. Patients suffering from AN are known to have difficulties in identifying emotional states in themselves and in others (i.e., alexithymia) and may, in part, be reversed parallel to weight gain during the course of treatment [18]. This is of clinical relevance, since difficulties in identifying emotions in others are associated with difficulties in one's own ER skills [19,20]. Relatedly, the interrelation between ER and binge-eating behavior postulated in different models. According to the affect regulation theory [21], binge-eating episodes in BN are used to relieve states of negative affect. In their meta-analysis, Haedt-Matt and Keel [22] showed that negative affect immediately before an episode of binge-eating is higher than a day's average affective content and higher than the dominant affect immediately prior to an unobtrusive eating behavior. In opposition to the affect regulation theory, the aversive emotional state does not resolve immediately after the binge-eating episode, but after an apparent delay of several hours [22]. Following a binge episode, compensatory behavior in BN may prevent a further increase in negative affect. In addition, a prior study analyzing ER in female ED patients before and after treatment found that emotional dysregulation can be modified as an effect of symptomatological ED improvement [23]. With these controversial results in mind, the question of whether the emotional dysregulation is a vulnerability factor for ED, a factor that maintains and worsens with the ED or both, is still open. At present, several manualized therapies for ED focusing on ER have been published [24,25].

Nevertheless, as in the vast majority of ED studies, females are overrepresented in studies on ER in ED [7,8], and males with ED are not researched as a whole. Although females greatly outnumber males with respect to diagnosed ED, it stands to reason that ER could also play a role in eating pathology in males, as it does in females.

With regard to gender-related differences in ER, evidence is scarce and mostly derived from studies in nonclinical community samples of males and females. In the study by Hayaki and Free [26], difficulties in ER predicted disordered eating in both male and female undergraduate students. Whereas some studies have shown no global differences between genders in nonclinical cohorts [27], others have shown gender-specific affective responses to high-calorie visual cues [28]. Significantly higher levels of rumination have also been identified in females, which, as an ER strategy, mediated the relationship between gender and disordered eating [29]. Difficulties in ER were identified as important determinants of body dissatisfaction and disordered eating in a study with only undergraduate males [30].

In a recent study in a cohort suffering from ED, difficulties in ER were found to be more strongly associated with cognitively oriented ED symptoms than with behavioral symptoms, such as binge eating, purging, driven exercise, non-suicidal self-injury, or suicide attempts. However, no gender comparisons were undertaken [31]. So far, studies investigating gender-related ER differences in clinical cohorts show no relevant gender-specific differences with regard to negative affect, emotional instability, and interpersonal dysfunction in an ED cohort consisting of *n* = 251 females and *n* = 137 males [32] or with regard to emotional overeating in a BED cohort comparing *n* = 172 females and *n* = 48 males [33]. There are also divergent results showing no differences in complex emotion recognition

between males with ED (*n* = 29) and healthy controls (HC) (*n* = 42) [34]. However, none of these studies in males made ER-specific instruments. Instead, the studies used subscales from a personality questionnaire as indirect measures to assess both negative affect and interpersonal dysfunction. Others have solely applied a specific measure of overeating in response to emotions, or analyzed only emotion recognition, but not ER strategies or emotion difficulties In addition, no study published to date, to our knowledge, analyzed ER in males using the different DSM-5 ED types, either because the sample size did not allow for it or because they only analyzed one ED type.

Personality traits and ER appear to be intertwined, with evidence showing links between the two in a number of studies [35,36]. For instance, difficulties in ER are implicated in the diagnostic criteria for some personality disorders (e.g., borderline personality disorder) [37]. ED are also associated with specific personality traits, including harm avoidance and low self-directedness in all ED diagnostic types, high novelty seeking in BED and BN, and high reward dependence and persistence in AN [38,39]. Males suffering from ED scored significantly lower than females with ED on harm avoidance, reward dependence, and cooperativeness, had less body image concerns, and lower general psychopathology [40]. In addition, dysfunctional personality traits are associated with higher ED severity, general psychopathology, self-harm behaviors, and worse therapy response and prognosis [41–43]. In a previous study by our group, we showed that ER difficulties mediated the relationship between personality traits (i.e., high harm avoidance and low self-directedness) and ED severity [44]. Thus, personality traits may increase vulnerability to ED pathology through ER difficulties. As these aspects were not studied in males with ED before, we incorporated an examination of the interplay between ER, personality traits, ED severity, and ED-related and general psychopathology in males with ED as further objectives of the present study.

Taking into account all the aforementioned gaps in the literature, primarily the lack of studies with clinical samples of males with ED, we aimed to examine ER in a large sample of consecutively recruited male and female patients with ED and HC, considering different DSM-5 ED types. Based on a previous research carried out at our Unit [23], which found how ER strategies improved along with improvements in eating symptoms after cognitive behavioral therapy (CBT), we analyzed the relationship between ED severity, general psychopathology, specific personality traits, and ER. In addition, assessment of the associations between ER and other behaviors commonly used to alleviate aversive emotional states, such as non-suicidal self-injury (NSSI), (reduced) interoceptive awareness, binge-eating, and purging behaviors were part of the study protocol.
