Cognition, Behavior, and Memory
Author: Gustavo Juantorena | Email: gjuantorena@gmail.com
Gustavo Juantorena 1°, Francisco Figari 1°, Agustín Penas 1°, Agustín Petroni 2°, Juan Kamienkowski 1°
1° Laboratorio de Inteligencia Artificial Aplicada, Instituto de Ciencias de la Computación, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires – CONICET, Buenos Aires, Argentina
2° University of Gothenburg, Gothenburg, Sweden
3° Maestría de Explotación de Datos y Descubrimiento del Conocimiento, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina.
Recently, there has been growing interest in the development of remote webcam-based eye tracking (ET) prototypes for conducting web-based experiments. These prototypes seek to explore broader and more challenging demographics, with potential applications in telemedicine. However, lower camera quality and the presence of ambient noise have posed new hurdles. To address these challenges, an improved webcam-based remote ET prototype is presented. In particular, enhancements have been introduced to optimize its usability for cognitive and clinical tasks by eliminating the need for constant mouse interaction. The spatio-temporal resolution and reliability of the prototype are evaluated, and its performance is assessed using the anti-saccade task, a well-established cognitive experiment that measures inhibitory control by analyzing eye movement behavior. The results of this experiment closely parallel those obtained with high-quality laboratory ETs, highlighting increased error rates during anti-saccade trials and accelerated response times for incorrect answers. Finally, the prototype demonstrates consistent calibration over time, an important metric for reliability of results.