Abstract
Previous research has shown that gaze closely reflects attentional bias as well as preference during preference-based decision making. For example, prior studies employing a two alternative choice task (e.g., choose between two faces) found that participants initially distribute gaze evenly between both options, but then gradually shifted toward the stimulus that they eventually chose. So far, most studies have examined how participants bias their gaze while presenting choices that are not associated with a strong prior preference. Here, we were instead interested in whether and how gaze biases can represent individuals’ varying degree of prior preferences during a choice task. To address this question, we recruited participants with an alcohol dependence who also smoked cigarettes, because they have already established different degrees of craving between these two substances. First, we identified their most preferred alcohol (e.g., Budweiser) and cigarette items (e.g., Marlboro). Then, participants were exposed to these two preferred items and self-reported cravings for each. Finally, in an eye-tracking task, we presented one each picture of an alcohol, a cigarette, and a water item on a computer screen. We randomly manipulated the relative preference of each item across trials by presenting a most preferred (e.g., Marlboro) or less preferred item (e.g., Newport cigarettes) along with a neural item (e.g., water bottle). We asked participants to select which of the three items they preferred the most by keypress while tracking their gaze. We demonstrated that the relative proportion of time spent looking at alcohol and cigarettes was correlated with self-reported relative craving. Furthermore, we demonstrated that participants fixated at the eventually chosen item for a greater proportion of time, regardless of the relative preference. Taken together, these results reveal a relationship between internal craving, gaze bias and choice during decision making.
Acknowledgement: NSF (BCS 1555006) awarded to J.-H. S.