As demonstrated by neuroimaging data, the human brain contains systems that control responses to threat. The revised Reinforcement Sensitivity Theory of personality predicts that individual differences in the reactivity of these brain systems produce anxiety and fear-related personality traits. Here we discuss some of the challenges in testing this theory and, as an example, present a pilot study that aimed to dissociate brain activity during pursuit by threat and goal conflict. We did this by translating the Mouse Defense Test Battery for human fMRI use. In this version, dubbed the Joystick Operated Runway Task (JORT), we repeatedly exposed 24 participants to pursuit and goal conflict, with and without threat of electric shock. The runway design of JORT allowed the effect of threat distance on brain activation to be evaluated independently of context. Goal conflict plus threat of electric shock caused deactivation in a network of brain areas that included the fusiform and middle temporal gyri, as well as the default mode network core, including medial frontal regions, precuneus and posterior cingulate gyrus, and laterally the inferior parietal and angular gyri. Consistent with earlier research, we also found that imminent threat activated the midbrain and that this effect was significantly stronger during the simple pursuit condition than during goal conflict. Also consistent with earlier research, we found significantly greater hippocampal activation during goal conflict than pursuit by imminent threat. In conclusion, our results contribute knowledge to theories linking anxiety disorders to altered functioning in defensive brain systems and also highlight challenges in this research domain.