Understanding the role of emotion, reward and food preference on the perception of sweet-tasting food using a novel classification task.

Abstract

Research demonstrates that humans tend to have behavioural biases toward calorie (kcal)-dense foods, particularly those that are sweet, although cognitive biases toward these foods are less well established. With major public health organisations highlighting the rising concerns surrounding excessive sugar intake, understanding these cognitive biases is of growing importance. Current knowledge about the cognitive and emotional influences on eating preferences is also inconsistent. This study aimed to explore implicit cognitive biases toward various sweet food categories and their connection with explicit behavioural and physiological responses. Using a within-subjects cross-sectional, factorial design, 102 participants (27 males, 74 females and one non-binary, average age 20.65) engaged in a novel food-based cognitive judgment task. Response time and accuracy in identifying sweet versus non-sweet foods were recorded, along with ratings of 'liking' and 'wanting' for each food stimulus. Foods were categorised into high-fat, high-carbohydrate, and high-protein groups. These measures were paired with questionnaire-based assessments of food preference, hunger and thirst, reward-related eating measured through the Reward-Related Eating Drive Scale (RED-13; Mason et al., 2017), and emotional eating tendencies measured through the Emotional Appetite Questionnaire (EMAQ; Nolan et al., 2010). Results revealed cognitive biases: participants were faster to respond to sweet foods, especially those high in fat and protein, and were less accurate in categorising sweet compared to non-sweet foods. Participants also reported greater liking and wanting for sweet foods overall. Furthermore, positive emotional eating was associated with quicker response times to sweet foods, and higher reward-related eating scores were linked to faster responses and broader food preferences. Although liking and wanting were each associated with response time, these effects were analysed separately; therefore, the study does not provide direct evidence of their distinct or independent contributions. The results are consistent with the notion of separate systems for liking and wanting in food reward, but further research using combined modelling approaches is needed to test this more directly. These findings underscore the complexity of emotional and reward-driven eating behaviours and offer insight for future work on cognition, perception, and public health strategies to reduce excessive sugar consumption.