
The beliefs about the outcome of an experiment may be represented by imprecise probability models. On the one hand, when experts aim to make more robust decisions, they may broaden their opinion by making less restrictive (i.e. more imprecise) their beliefs. Distortion models constitute ruled manners to achieve this objective. On the other hand, conflict naturally arise when diverse beliefs are put together. In the literature, conflict quantification and merging methods (often linked) have received quite attention. The most usual approach to solve such conflict is applying an imprecise aggregation rule, i.e. to select an imprecise model that is deemed as representative of the group, on the basis of individual beliefs. Precisely, this manuscript investigates the use of distortion models for imprecise probabilities as a tool for reaching a consensus in situations of global conflict. More concretely, this shall be carried out by forcing individual beliefs to yield and thereafter merging the conceded beliefs. For this purpose, general results on distortion models are provided in first place. Afterwards, it is proposed and analysed in detail an aggregation rule for lower probabilities, called distortion-conjunction, and its usage to solve and measure conflict is studied. Then, the properties of several particular instances of such models are comparatively analysed. Finally, a number of variations of this aggregation rule are proposed and briefly discussed for future research.