Ana Krstevska, Sebastjan Kramar, Hristijan Gjoreski, Martin Gjoreski, Junoš Lukan, Sebastijan Trojer, Mitja Luštrek and Gašper Slapničar

Abstract
To mitigate distractions during complex tasks, ubiquitous
computing devices should adapt to the user’s cognitive load.
However, accurately assessing cognitive load remains a significant
challenge. This study aims to present sophisticated, multi-modal
data collection, which can enable accurate estimation of cognitive
load using wearable and contact-free devices. A total of 25
participants participated in six cognitive load-inducing tasks, each
presented at two levels of difficulty. Simultaneously, physiological
and behavioral data were collected from a multi-modal sensory
setup, including: Empatica E4 wristband, Emteq OCOsense
glasses, an eye tracker, a thermal camera, a depth camera and an
RGB video camera. Additionally, participants provided subjective
measures of cognitive load by completing standardized NASA
Task Load Index (NASA TLX) and Instantaneous Self-Assessment
(ISA) questionnaires following each cognitive task. Preliminary
statistical analyses were conducted on participant demographics,
performance metrics, and the perceived difficulty of tasks, as
reported in the completed questionnaires.