The purpose of this cross-sectional study was to examine the levels of addiction, eating disorder pathology, memory biases and self-schemas that contribute to, and maintain eating disorders. Participants (N =30) were female and were divided into three different groups (eating disorder, eating disorder recovered and control). Each participant completed a short eating disorder relevant memory task, five self-statements in answer to the question "What do you most like/ dislike about yourself?", the Eating Disorders Inventory (3) and the addiction scale of the Eysenck Personality Questionnaire. Analysis using analysis of variance (ANOVA) showed significant differences between the groups in addictiveness (F (2,27) = 9.5, p = 0.001). A second series of ANOVA's revealed significant differences between the eating disorder group and the other two groups (eating disorder recovered and control) on measures of body dissatisfaction, drive for thinness, bulimia and general psychological maladjustment. An ANOV A revealed no significant differences between the groups in terms of the number of eating disorder relevant words remembered from the memory task. However, inspection of descriptive statistics revealed a trend towards an eating disorder relevant memory bias among all of the groups. Qualitative (content analysis) and quantitative (Kruskal-Wallis) analyses of self-statements revealed differences between the control group and the other two groups (eating disorder and eating disorder recovered) in terms of the number of positive statements made about appearance, with the control group writing significantly more positive statements regarding their appearance. No other significant differences were found. Results indicate that individuals with an eating disorder differ from individuals recovered from an eating disorder and those who have never experienced an eating disorder in terms of their addictiveness and their eating disorder pathology. No differences were found between the groups in the results of the memory test, although differences in terms of self-statements were observed. Further investigations are needed to expand upon results and to investigate each hypotheses in more detail.