Detecting Early Cognitive Decline Using Eye-Tracking Metrics in a Virtual Reality Mini-Mental State Examination

This dataset contains eye-tracking metrics and performance data collected during a Virtual Reality Mini-Mental State Examination (VR-MMSE) designed to detect early cognitive decline. The study included 60 participants across three groups: Alzheimer's disease (n=20), mild cognitive impairment (n=20), and healthy controls (n=20). 

Data includes:1. Demographic information (age, education level, gender)2. Traditional MMSE and VR-MMSE performance scores3. Eye-tracking metrics including fixation duration, saccade velocity, and gaze enter count data for predefined regions of interest4. Classification analysis results comparing diagnostic accuracy

Data was collected using a Pico 4 Pro head-mounted display with integrated eye-tracking capabilities (90 Hz refresh rate, 0.5° angular accuracy, sampled at 60 Hz). The VR environment was developed using Unity3D, adapting traditional MMSE components to an immersive setting.

This dataset supports research on multimodal approaches to early detection of cognitive impairment, demonstrating how integrated performance measures and physiological data can enhance diagnostic sensitivity.