Hidden Markov model analysis for space shuttle crewmembers' scanning behavior

The paper describes the application of a novel analysis technique employing hidden Markov models (HMMs) to analyze complex sequences of visual fixations and understand the supervisory monitoring strategies of space shuttle cockpit crewmembers. While HMM analysis has been shown to be useful for understanding the scanning behavior of air craft pilots engaging in manual flight control, it has not yet been applied to the supervisory monitoring context. The paper presents a proof-of-concept demonstration by applying HMM analysis to the eye-movement data of a veteran Commander astronaut performing simulated ascent-phase operations. The paper also demonstrates how the resulting HMM parameters can be used to detect deviations from regular scan patterns.