Arguing about preferences and decisions

Making a rational decision requires the decision maker's preferences. However, in many situations it is not clear what to prefer. For example, the outcome of a decision may never been encountered before, there may be are very different aspects that matter, or it may be difficult how to weigh short-term and longterm consequences. Furthermore, artificial agents may be required to justify and discuss their decisions to others. Designers must communicate their wishes to artificial agents. Existing research does not address how to reason about what to prefer. Therefore, this paper addresses how to reason about preferences. For this, argumentation is used because it allows to justify and attack preference statements. To be able to justify and question preferences, we propose a qualitative model of preferences that is inspired by multiattribute utility theory.