MENTA: how to balance authorial intention and user agency in virtual environments

We aim to create an artificial intelligence based virtual environment to train medical team leaders to rescue injured people after a mass casualty. In this paper, we describe a resilient and adaptive engine, MENTA, to orchestrate dynamically various training situations and various virtual teammates. MENTA is in charge of the narrative control by proposing a set of adjustments that satisfies narrative objectives chosen by the trainer. These adjustments take the form of a prescribed scenario that is generated by MENTA via a planning engine that we have coupled with fuzzy cognitive maps. This approach tackles the problem of a set of objectives that are often contradictory: user agency, authorial intention and resilience.