Unsupervised Clustering Uncovers Two Distinct Types of Fixational Eye-Movements in Dynamic Environments

We investigated eye-movement control in a dynamic, time-constrained task environment where participants steered a spaceship to avoid obstacles under varying levels of motor control uncertainty. Using unsupervised clustering, we identified two distinct fixation types: Type 0 fixations, which were longer in duration and centrally located near the spaceship, and Type 1 fixations, which occurred farther from the agent and were directed toward more open areas of the screen. Linear mixed modeling revealed that with increasing input noise, Type 0 fixations became shorter and more tightly focused, while Type 1 fixations became longer in duration and were allocated even farther from nearby obstacles. These patterns suggests adaptive gaze strategies, with Type 1 fixations possibly supporting predictive tracking of future target locations under high control uncertainty. Our findings contribute to understanding how eye movements flexibly support action in complex, real-time environments.