Collaborative groups are most effective when the varied stakeholder groups within them understand the risks of wildfire and take proactive steps to manage these risks. Implementing policies for fire risk mitigation and adaptation, however, remains difficult because risks and policy alternatives are not understood or supported uniformly across diverse stakeholders. To facilitate greater understanding and collaboration across diverse groups, we developed a novel approach, based on Fuzzy Cognitive Maps (FCM), in which we systematically collected mental model representations from a range of stakeholders involved in wildfire management in the Ashland, Oregon area to better understand their diverse perceptions of wildfire events, wildfire impacts, and wildfire management and their willingness to support fire management policies. We used the Mental Modeler software in seven stakeholder workshops to facilitate building a group FCM. Mental Modeler helps individuals and communities capture their knowledge in a standardized format that can be used to analyze, through simulation, how the group thinks about management alternatives and what leverage points they see to improve the system under study. The data can also be used to analyze similarities and differences across stakeholder groups.