Introduction Attachment theory is one of the most important theoretical frameworks for understanding human relationships and emotion regulation. Attachment represents permanent emotional bond between the child and primary caregiver (Ainsworth et al., 1978), the purpose of which is maintaining proximity to the attachment figure, especially in stressful situations (Bartholomew & Horowitz, 1991). The attachment figure represents a person who provides emotional and physical care and is consistently and continuously present in the child's life, and who invests emotions in the relationship with the child (Pearce, 2009). The mental representations of self and others that a child develops through early experiences with an attachment figure represents the attachment internal working model (Bowlby, 1973). Internal working models of attachment are the foundations for stable and consistent ways in which children and adults perceive themselves and others, interact, interpret information and regulate emotions. Emotional regulation is defined through processes by which individuals affect the emotions that they will experience, when they experience them and how to express them (Gross, 2014). An individual with a negative sense of self and others is not able to adequately regulate their emotional reactions, which can have a significant impact on the psychopathology of eating disorders. Bulimia nervosa is often described as a syndrome of overeating and purging, consists of episodes of binge eating followed by induced vomiting or misuse of laxatives, to be expelled from the body huge amounts of food (Marcinko, 2013). The disorder occurs in approximately 10 women to one man, usually in late adolescence or young adulthood (APA, 2000), and most often between 14 and 25 years of age (Cavanaugh and Ray, 1999). For bulimic patients the typical psychopathology is associated with poor impulse control, such as auto-aggressive and hetero-aggressive behavior, suicide, sexual promiscuity etc. In this paper we will try to explain in detail the nature of the link between insecure attachment, dysfunctional emotional regulation and bulimia nervosa. Methodology The study was designed as a cross-sectional with clinical and control group of subjects. The clinical group consisted of 50 patients suffering from bulimia nervosa, treated at the University Hospital Center Zagreb. The control group consisted of 50 female subjects that was characteristically consistent with the clinical group subjects. The study used socio demographic profile questionnaire, Eating Disorder Inventory 2 (EDI-2; Garner, 1991), The Experiences in Close Relationships-Revised (ECR-R; Brennan, Clark and Shaver, 1998), Difficulties in Emotion Regulation Scale (DERS; Gratz and Roemer, 2004) and SCOFF (Morgan, Reid and Lacey, 1999). The statistical analysis of results utilised methods of descriptive and inferential statistics, and graphical representations of results. Results The clinical group subjects achieved significantly higher scores on the anxiety and avoidance dimensions than the control group subjects. The regression model showed itself as statistically significant and was able to explain the 24.6% of the variance of BN symptoms, where anxiety is a statistically significant predictor of BN symptoms , and avoidance does not significantly contribute to the explanation of variance in BN symptoms. The subjects of the clinical group showed significantly more emotional regulation difficulties. Difficulties of emotional regulation represent a statistically significant predictor of symptoms BN and explains 41.3% of variance in BN symptoms. Anxiety, avoidance and emotional regulation are individually shown to be statistically significant predictors in the prediction of group membership, but after including all the variables in the regression model, the contributors to anxiety and avoidance have lost their significance, while emotional regulation proved to be a statistically significant predictor in the prediction of group membership. Hierarchical regression analysis showed the existence of mediation influence of emotional regulation in the relationship of dimensions of attachment and severity of symptoms of BN. Discussion The findings are consistent with the assumption that women with BN are expected to show higher level of anxiety and avoidance, suggesting that patients with BN experience close relations through constant vigilance, a sense of fear and worries of possible rejection and abandonment (anxiety), and the discomfort of closeness (avoidance), which is defined as fearful attachment (Hazan and Shaver, 1987). Anxiety is a significant predictor of symptoms of BN, which is associated with lower capacity to regulate emotion, while avoidance in terms of attachment theory is associated with AN, and increased control of emotions. The subjects of clinical group showed significantly more emotion regulation difficulties, and emotional regulation difficulties represent a statistically significant predictor of BN symptoms, which is also consistent with the findings of previous studies suggesting that people who suffer from eating disorders, including BN, reported a higher level of emotional regulation difficulties (Harrison et al., 2010b; Svaldi et al., 2012), as well as in accordance with the theoretical model according to which emotional regulation difficulties contribute to psychopathology of eating disorders, as well as acting as inappropriate strategies to manage their own emotions (Wilde et al., 2010; Haynos & Fruzzetti, 2011; Wonderlich et al., 2014). Anxiety, avoidance and emotional regulation are individually shown to be statistically significant predictors in predicting whether people belong to clinical or control group, but after inclusion of all the variables in the regression model anxiety and avoidance contributors have lost their importance, while emotional regulation proved to be a statistically significant predictor of group membership. Hierarchical regression analysis showed the existence of mediation influence of emotional regulation in the relationship of attachment dimensions and severity of symptoms of bulimia nervosa. The findings are in line with the mediating model (Tasca et al., 2009), which assumes that attachment affects emotional regulation, which in turn influences eating disorders. Conclusion It was found that patients suffering from BN showed significantly higher levels of anxiety and avoidance than control group subjects, suggesting that they experience close relations through constant vigilance, a sense of fear and worries of possible rejection and abandonment (anxiety), and the discomfort of closeness (avoidance), which is defined as fearful attachment. Both dimensions of attachment were significantly positively correlated with the symptoms of BN, but in the same model dimension anxiety emerged as an significant predictor of symptoms BN, while avoidance proved to be an insignificant predictor, indicating that the anxiety is a stronger correlate of BN symptoms. It was also found that subjects who suffer from BN have more difficulties in emotional regulation than women in the control group. Emotional regulation difficulties significantly associated with BN symptoms, and represent a statistically significant predictor of BN symptoms, in the way that those patients with pronounced emotion regulation difficulties show higher levels of BN symptoms. Both dimensions of attachment as well as emotion regulation individually successfully predict whether a person will suffer from BN, but in the same regression model anxiety and avoidance proved to be insignificant predictors, while emotional regulation proved to be a significant predictor. The findings indicate that emotional control correlates stronger with BN, than the dimensions of attachment, and suggest that the anxiety and avoidance are associated with BN because of its association with emotional regulation, while emotional regulation demonstrates a unique relationship with BN. The relationship between variables is explained through mediation. Conducted hierarchical regression analysis determined the mediation influence of emotional regulation in the relations of attachment and BN, whereby attachment influences emotional regulation strategies, which in turn influences eating disorders.