Cognitive theories of eating disorders implicate Attentional Bias (AB) towards food-related information in the development and maintenance of eating disorders. Empirical evidence for this proposal, however, has been inconsistent, and the measures used to examine AB to food-related stimuli typically showed poor reliability. The aim of the current study was twofold. Firstly, we aimed to examine the psychometric properties of a newly devised eye-tracking task for the assessment of AB in the context of eating disorders. Secondly, we examined the role of Eating Disorder-specific (ED-specific) rumination as a potential moderator of the association between attentional bias to food images and eating disorder symptoms. One hundred and three female students were recruited and completed an eye-tracking task comprising 21 matrices that each contained 8 low-calorie and 8 high-calorie food images. Each matrix was presented for 6 s. First fixation location, first fixation latency, and total dwell time were assessed for low and high-calorie food images and the dwell-time based AB measure showed good reliability based on Cronbach's alpha, McDonald's Omega, and split-half method. In addition, the results revealed that the ED-specific rumination plays the hypothesized moderating role. Specifically, while participants with high levels of ED-specific rumination exhibited a positive association between AB to high-calorie foods and eating disorder symptoms, this association was not present among participants with lower levels of ED-specific rumination. The employed free-viewing task seems a reliable measure of AB to food-related stimuli, and the moderation analysis emphasizes the critical role of ED-specific rumination for eating disorder symptoms. Implications, limitations, and directions for future research are discussed.