Summary
This document is a validation study report for the WebET 3.0 algorithm released by iMotions in May 2023. The study aimed to evaluate the performance of the algorithm using a diverse global sample of 255 participants over 35 days. Participants engaged in a short study involving gifs, images, videos, and surveys, with self-reported parameters such as ethnicity, eye color, and lighting conditions assessed against accuracy. The results indicated that over 50% of participants achieved an accuracy of 2 degrees of visual angle (dva) or lower, while over 90% had an accuracy of 5dva or lower. The study also examined the impact of individual and demographic variables on accuracy, finding that only the presence of glasses significantly affected results. Additionally, the document discusses methodological considerations for webcam-based eye tracking and emphasizes the importance of maintaining participant compliance and optimal study conditions to ensure data quality.
