Gaze transition entropy and automation trust in a multitasking workspace

Safe flight operation requires visual scanning across multiple displays in a cockpit, which collectively represent the state of the aircraft and supporting automation. Trust is a crucial factor that drives human-automation interaction, and recent work has suggested a relationship between an operator's visual attention and automation trust. One index that captures predictability of eye movements between different areas of interest is gaze transition entropy. The current work reanalyzed data from Sato et al., which examined eye movement patterns and trust in automation associated with the system monitoring task of the Multi-Attribute Task Battery. Results showed credible positive correlations between the entropy measures and performance-based trust, but not for process- nor purpose-based trust. Specifically, higher levels of performance-based trust were associated with more random eye movement. Gaze transition entropy may provide a new window into the relationship between visual attentional resource allocation and automation trust in a multitasking workspace.