Human escape follows a structured movement pattern shaped by threat and context

Evading danger is critical to survival. In non-human animals, escape strategies are shaped by neural, biomechanical, and ecological constraints, resulting in species-specific patterns. In humans, ethical and practical constraints have, until recently, hindered investigation of escape movements. To fill this gap, we used wireless virtual reality and simulated various biologically relevant threats including apex predators as well as aggressive non-predators and conspecifics. We identified a distinct motor sequence in human escape to shelter: a head turn toward the threat, followed by body rotation in the same direction until facing away from threat, and propulsion with the ipsilateral foot to run forward. Infrequent variants included turning away from threat to escape, backward movement, and misdirected flight. Backward escape occurred more often at close shelter distance and with small-to-medium sized threats. Key kinematic features – such as backward movement, delayed escape initiation, and delayed acceleration – reduced escape success. At close threat distances, pre-encounter adjustment – such as widening the foot stance – contributed to successful escape. Individuals exhibited stable, threat-specific movement preferences that were not predicted by behavior in non-threatening circumstances. These findings establish a framework for dissecting the neural control of escape, quantifying modulatory influences, and exploring its potential disruption in clinical populations.