Abstract. We propose a theory of a-rational persuasion in which we integrate emotional and non emotional strategies by arguing that they both imply reasoning and planning abilities in the two participants. We show some examples of texts from a corpus of persuasion messages in the healthy eating domain and propose a formalism to represent this knowledge. The final goal of our research is to simulate user-adapted persuasion dialogs about healthy eating. 1

Eating habits are influenced by emotional factors. Persuasion to change wrong habits should therefore act on the central and the peripheral route at the same time, by combining rational and emotional strategies appropriately. In this paper, we describe an ongoing research which is aimed at simulating persuasion dialogs in this domain. We describe how we collected and analyzed a corpus of messages and how we compared the persuasion strength of rational vs emotional, negatively vs positively framed discourses. We conclude the paper with a discussion of the knowledge representation methods that are suited to represent these strategies.

This paper presents an overview of communication of uncertainty in genetic counseling patient letters, and its implications for the design of patient-tailored communication systems. A system designer should take into account the prevalence of uncertainty expressions. Many can be analyzed as giving conditional probabilities. Often, the probability values are qualitative or imprecise. Thus, we discuss use of a hybrid Bayesian/Qualitative probabilistic network domain model for this type of system.

In the field of natural language dialogue, a new trend is exploring persuasive argumentation theories. Applying these theories to human-computer dialogue management could lead to a more comfortable experience for the user and give way to new applications. In this paper, we study the different aspects of persuasive communication needed for health-care advising and how to implement them to produce efficient, computer directed persuasion. Our opinion is that a persuasive dialogue will have to combine the current logical approach to persuasion with novel emotional cues to render the dialogue more comfortable to the user.

We describe our preliminary work on analyzing a corpus of clinical genetics patient letters. In this genre, a single letter is intended to serve multiple functions and is designed for audiences with different backgrounds, information needs, and affective responses to the information. The goal of the analysis is to identify stylistically-related features for natural language generation. Our hypothesis is that sets of these features are associated with different points of view.

Eating habits are influenced by emotional factors. Persuasion to change wrong habits should therefore act on the central and the peripheral route at the same time, by combining rational and emotional strategies appropriately. In this paper, we describe an ongoing research which is aimed at simulating persuasion dialogs in this domain. We describe how we collected and analyzed a corpus of messages and how we compared the persuasion strength of rational vs emotional, negatively vs positively framed discourses. We conclude the paper with a discussion of the knowledge representation methods that are suited to represent these strategies.