Decoding target discriminability and time pressure using eye and head movement features in a foraging search task.

Area of Science:

• Human-Computer Interaction
• Cognitive Science
• Virtual Reality

Background:

• Effective decision-making in military operations relies on rapid visual search.
• Virtual reality (VR) offers a controlled environment to study complex visual search behaviors.
• Eye and head movements are key indicators of attentional and search strategies.

Purpose of the Study:

• To investigate how eye and head movement metrics can infer changes in search behavior.
• To determine the combined utility of eye and head movement data in classifying search conditions.
• To explore the differential contributions of eye and head metrics to understanding task-induced changes.

Main Methods:

• Thirty-one participants performed a visual search task in VR using a head-mounted display (HMD) with eye tracking.
• Task parameters included varying target discriminability (easy/hard) and time pressure (low/high).
• Support vector classifiers were trained using eye and head movement features to predict task conditions.

Main Results:

• Increased discrimination difficulty and time pressure impaired performance (slower response times, reduced d-prime).
• Combined eye and head movement features yielded the most accurate classification of search behavior.
• Eye features (fixation) were strong for discriminability, while head features (kinematics) were important for time pressure.

Conclusions:

• Eye and head movement metrics provide complementary information for understanding visual search.
• Integrating both eye and head movement data enhances the classification of task-induced changes in search behavior.
• This research offers insights into optimizing training and performance in high-pressure visual search environments.