Implicit markers of food choice

Abstract

Because of the health risks associated with unhealthy eating and overweight, it is important to better understand the motives underlying (un)healthy food choice. Explicit measures, such as questionnaires and focus groups, are suboptimal because they only tap into that specific part of the motive that people are conscious of and they are highly susceptible to socially desirable responding. Therefore, the aim of this thesis was to establish the potential of implicit markers to (1) predict choice and (2) gain insight into the decision-making processes underlying food choice. To this end, we evaluated markers of cognitive processing (reaction times), visual attention (eye-tracking) and central (brain activation) processing. To determine the potential of brain activation we synthesized results from previous studies with a meta-analysis. Hunger modulated the response to food pictures in reward regions and energy content modulated the response both in reward regions and the visual cortex. To investigate which brain regions respond to preferred products and to what extent everyday food choices can be predicted with brain activation, we performed an fMRI study in which subjects chose between two alternatives of the same product, but with different packaging. Traditional mass-univariate analysis could detect small package-induced differences in product preference and the novel multivariate pattern analysis (MVPA) could predict realistic low-involvement consumer choices from functional MRI data (61% accuracy). Prior research has shown that consumer choice can be influenced by manipulating visual characteristics, but it was unknown whether this was due to increased initial (first fixation) or sustained (fixation duration) attention. In our eye-tracking study we showed that manipulating the location of the first fixation did not influence consumer choice. To gain more insight into the neural processes underlying self-control in weight-concerned women, we performed a combined functional MRI/eye-tracking study. An assumption in self-control theory is that if someone with the long-term goal to limit intake is exposed to a food choice that threatens the accomplishment of this goal, this results in internal conflict. In contrast to this assumption, we found that choices requiring self-control induced no conflict, as demonstrated by lower reaction times, fixation durations, number of gaze switches between snacks, and lower activation of the anterior cingulate cortex. When subjects had to either accept or reject a single high or low energy food we found that choices concerning high energy snacks elicited stronger activation in reward-related brain regions. Highly tasty high energy snacks were more difficult to resist and, accordingly, activation in inhibitory areas was modulated negatively by tastiness. More successful self-controllers showed increased activation in the supplementary motor area during food choices. To conclude, fMRI and eye tracking measures can differentiate between more and less preferred products, and fMRI in conjunction with MVPA can be used to predict choice. Moreover, reaction times, eye tracking measures, and fMRI are valuable tools that give insight into the processes underlying food choice, like self-control, especially when multiple measures are combined.