What reported food-evoked emotions may add: A model to predict consumer food choice

• Intrinsic and extrinsic product properties elicit in part different emotions.
• Food-evoked emotions could be organised in two dimensions (valence vs. arousal).
• Food-evoked emotions have added value beyond liking in predicting food choice.

Food-evoked emotions provide information that goes beyond the information from traditional hedonic ratings. The objectives of our study were: (i) to investigate how intrinsic (sensory) and extrinsic (packaging) cues affect consumers’ emotional responses to foods, and (ii) to explore whether emotional responses to these cues combined with liking, predict actual food choice. Participants (n = 103) rated emotional responses to seven products under a blind taste, a package and a package and taste condition using the EsSense Profile™. During the blind taste condition participants also scored liking of the products. Test products were breakfast drinks and desserts. Food choice was measured in two different breakfast sessions reflecting a different choice context. In one choice context, products were presented blind to taste, after which participants chose one out of the seven foods to consume for breakfast. In the other choice context, participants based their choice on the package of the seven foods without tasting them. Results showed that emotions evoked by food products could be organised in a two-dimensional space, representing a valence (pleasantness) and an activation/arousal dimension. Specific emotional profiles generated for products differed across the blind taste, package and the package and taste condition, meaning that intrinsic and extrinsic product properties elicit in part different emotions. Liking and valence together had the strongest predictive value for product choice based on the product’s taste. The combination of liking, valence and arousal had the strongest predictive value for package-based choice. In conclusion, food-evoked emotions add predictive value to solely liking ratings, and may guide consumers’ product choice behaviour.