ii Abstract
AbstractThis report describes the use of personalized persuasive messages to influence healthiness of menu choice. The susceptibility to persuasion scale (STPS) was used to assign participants to different strategies. We hypothesized that persuasive messages would have a positive effect on menu choices. A 10-day experiment was setup to evaluate the effect of susceptibility, in which the users were exposed to persuasive messages about healthy eating. Each of the persuasive messages implemented a particular social influence strategy. In total, 324 participants made a daily menu choice out of three presented menus that differed in health score, for ten days in a row. The analysis includes multilevel (regression) models that were fitted to account for the interdependency of the data points. The experiment does not show significant evidence for any of the hypotheses; personalized (tailored) messages did not lead to healthier menu selection than contra-tailored, random or no messages at all. Therefore, it is assumed that food choices are affected more by other factors than the subtle persuasive intervention that was used in this study. However, participants were also asked to estimate the health of each menu. A (negative) effect of persuasive messages was found on the health estimation of the menus. We suggest that the decrease in health estimation is caused by an increase of the health ideal. Furthermore, the results are compared to related work in the field of personalized persuasive messages. In particular with a similar study that has investigated persuasive effects on the reduction of snacking, here researchers found opposite results. The results of this study will help in designing a persuasion engine that will be implemented in the smart kitchen scenario, which has the purpose to create context awareness and empowers people to eat healthy.