Framing and Tailoring Prefactual Messages to Reduce Red Meat Consumption: Predicting Effects Through a Psychology-Based Graphical Causal Model

Abstract

Effective recommendations on healthy food choice need to be personalized and sent out on a large scale. In this paper, we present a model of automatic message selection tailored on the characteristics of the recipient and focused on the reduction of red meat consumption. This model is obtained through the collaboration between social psychologists and artificial intelligence experts. Starting from selected psychosocial models on food choices and the framing effects of recommendation messages, we involved a sample of Italian participants in an experiment in which they: (a) filled out a first questionnaire, which was aimed at detecting the psychosocial antecedents of the intention to eat red/processed meat; (b) read messages differing as to the framing of the hypothetical consequences of reducing (gain, non-loss) versus not reducing (non-gain, loss) red/processed meat consumption; (c) filled out a second questionnaire, which was aimed at detecting participants’ reaction to the messages, as well as any changes in their intention to consume red/processed meat. Data collected were then employed to learn both the structure and the parameters of a Graphical Causal Model (GCM) based on a Dynamic Bayesian Network (DBN), aimed to predicting the potential effects of message delivery from the observation of the psychosocial antecedents. Such probabilistic predictor is intended as the basis for developing automated interactions strategies using Deep Reinforcement Learning (DRL) techniques. Discussion focuses on how to develop automatic interaction strategies able to foster mindful eating, thanks to (a) considering the psychosocial characteristics of the people involved; (b) sending messages tailored on these characteristics; (c) adapting interaction strategies according to people’s reactions.

A theory-based and data-driven approach to promoting physical activity through message-based interventions

“…A selective search was then conducted in the space of possible structures, retaining only those variables that effectively improved the accuracy of the network, in terms of combined in-sample and out-of-sample predictive abilities. The structure of the Dynamic Bayesian Network was determined by an extensive automatic search as described in Catellani et al (2022) . Consistent with the intended use, candidate DBN structures were tested by using pre-intervention measures as evidence and then evaluating the conditional probability distribution for the target variable.…”

“…A selective search was then conducted in the space of possible structures, retaining only those variables that effectively improved the accuracy of the network, in terms of combined in-sample and out-of-sample predictive abilities. The structure of the Dynamic Bayesian Network was determined by an extensive automatic search as described in Catellani et al (2022).…”

“…Regular physical activity is one of the cornerstones of a healthy lifestyle, as it has numerous benefits for health, well-being and reduced welfare costs ( Ding et al, 2016 ; Marquez et al, 2020 ; WHO, 2020 ). To date, psychological research has addressed the issue of physical activity engagement using either a theory-based or a data-based approach ( Yarkoni and Westfall, 2017 ; Dijkhuis et al, 2018 ; Aldenaini et al, 2020 ; Zhang et al, 2020 ), while only a small body of research fully combines the two approaches ( Catellani et al, 2022 ). The present research aimed to combine the best of both worlds, combining the great explanatory power of a theory-driven approach with the great predictive power of a data-driven method.…”

Psychosocial drivers influencing local food purchasing: beyond availability, the importance of trust in farmers

“…Moreover, among highly educated people, the perception that local food respects the environment mediates the link between trust and the intention to buy local food. This suggests the importance of communication campaigns in the food sector ( 54 ) and the need to test the effectiveness of various types of content, especially based on the target audience they are intended for ( 55 , 56 ).…”

Legumes or Meat? The Effectiveness of Recommendation Messages towards a Plant-Based Diet Depends on People’s Identification with Flexitarians

“…A longer study could also verify if the changes in the consumption of legumes and meat will modify people’s attitudes towards flexitarians and their identification with them. Fourth, this study mainly captured the “rational” processes of message evaluation and impact, overlooking the possibility that emotional reactions (such as fear or regret) or moral considerations (i.e., ascription of responsibility) might mediate the message effects [ 27 , 66 , 67 , 68 , 69 , 70 , 71 ].…”