Abstract: A considerable number of studies have identified selective attentional biases toward food-related information in restrained eaters. However, to date, the nature of information-processing biases that differentiate successful from unsuccessful restrained eaters has yet to be clarified. Based on Stroebe’s (2008) goal conflict model of eating, the current study was designed to investigate components of attentional bias among successful restrained eaters, unsuccessful restrained eaters, and non-restrained eaters under two conditions: a hedonic goal condition and a restrained goal condition. We hypothesized that within the hedonic goal condition, rapid avoidance would be observed in successful restrained eaters and disengagement difficulties would be displayed among unsuccessful restrained eaters. Conversely, within the restrained goal condition, rapid avoidance of palatable food cues would be present within both categories of restrained eating.
From an initial sample of 907 Chinese undergraduate women, we recruited 60 successful restrained eaters, 60 unsuccessful restrained eaters and 60 non-restrained eaters based on scoring cut-offs from the Dutch Eating Behaviour Questionnaire (van Strien et al., 1986; 2007) to participate in dot probe research. In Experiment 1, a hedonic goal was activated prior to the dot probe task that featured palatable and neutral cues. Results of 3 (Group) × 3 (Prime Location) × 2 (Food Palatability) analyses of variance indicated disengagement difficulties related to palatable food cues among unsuccessful restrained eaters, rapid avoidance of these cues among successful restrained eaters, and no biases in responses to palatability cues among non-restrained eaters. In Experiment 2, a restrained goal was activated subliminally following hedonic goal activation before the dot probe task. Among unsuccessful restrained eaters, disengagement difficulties and rapid avoidance of palatable food cues. In successful restrained eaters, only rapid avoidance was shown.
In conclusion, results indicated that successful restrained eaters showed a tendency to avoid palatable food cues while unsuccessful restrained eaters showed a tendency to approach to palatable food cues. On this basis, researchers should consider the distinction between successful and unsuccessful restrained eating when assessing attentional biases corresponding to restrained eating.

Key words: restrained eating, goal conflict model of eating, attentional bias