This paper presents an exploratory microworld study with the aim to identify individual dierences between participants, and relate those dierence to how well the participant solves the task. Six ocers, rank from lieutenant commander to flotilla admiral, were studied when they commanded a maritime escort mission. The experiment was conducted using a microworld where the participant had to control all own units while the computer controlled enemy and neutral units. Data collection consisted of think-aloud protocols, screen captures of the microworld’s tactical screen, questionnaires, and battle outcomes. Performance was determined using a measure of mission success and a general model of the participants’ decision making process was constructed. This model was used to identify individual dierences and relate those to task performance. The results suggest that there is no correlation between how often the participants perform a certain decision making activity, and how well they perform in the microworld. On the other hand, the results suggest a strong correlation between how well the participants perform in the microworld and how many dierent decision making activities they visit during one coherent reasoning chain. The result seems to suggest that it is more important to consider many aspects of a problem at the same time, and that no decision making activity is more important that another.