Published January 26, 2026 | Version v1
Dataset Embargoed

Oopses and Ohs: An Eye-Tracking Study of Human Errors in Virtual Reality

Authors/Creators

  • 1. ROR icon University of Luxembourg

Description

This dataset contains longitudinal eye-tracking recordings collected in a virtual reality (VR) experiment investigating human cognitive errors during a mental arithmetic task. Over 18- and 6-month periods, two participants completed repeated VR sessions while solving multiplication problems, yielding more than 60 hours of eye-tracking data sampled at 120 Hz, collected using a Tobii eye tracker integrated into the HTC Vive Pro Eye VR headset.

Each trial is labeled according to task outcome (Correct = 2, Missed = 1, Wrong = 0) and includes a unique trial identifier encoding participant, session, and trial number (UID with format: PX-##-##), as well as the presented question, the participant’s response, pupil diameter, gaze direction, pupil position, and eye openness.

The dataset was designed to support research on cognitive load, error prediction, and human performance modeling in immersive environments, and accompanies the paper “Oopses and Ohs: An Eye-Tracking Study of Human Errors in Virtual Reality” (GEMINI Workshop at IEEE VR).

Files

Embargoed

The files will be made publicly available on March 21, 2026.