Attitudes and eye movements during vegetable choice tasks

Abstract

Background: Studies have shown that attitudes are related to food choices. While attitudes are usually assessed with questionnaires, computer-based studies revealed an association between eye movements and attitudes, suggesting eye tracking as a potential method to measure attitudes. Recent technological developments now allow to test this assumption in more naturalistic settings by using mobile eye tracking. The present study aimed to compare the relationships between dwell time, attitudes, and food choice in a conventional computer-based task and a naturalistic food buffet task. Methods: Vegetable choice was assessed both using a paired comparison task on a computer and a naturalistic yet standardized food replica buffet from which participants (N = 83) were asked to serve themselves a meal they would typically eat for lunch. During the tasks, the participants’ eye movements were recorded with a mobile eye tracker. In addition, explicit attitudes towards vegetables were measured. Findings: Path modelling revealed that dwell time on vegetables is related to attitudes and choice in both tasks (βs ≥ .21, ps ≤ .025). However, when reaching induced eye movements were excluded, dwell time was significantly related to attitudes and choice in the computer task only (βs ≥ .24, ps ≤ .031). Discussion: The present results suggest that findings from conventional computer-based tasks may not be fully transferable to naturalistic settings due to differences in the overall task structure as well as differing characteristics of the stimuli.