Integration of Pupil Size Measures and Self Reports in Understanding the User's Cognitive Load

Abstract

In today's digital world, understanding the cognitive load that digital interfaces require from users is key to designing an optimal user experience. Cognitive load (CL) describes the working memory capacity required to perform a cognitive task and/or interact with a system.

Rating scale techniques operate under the belief that individuals can reflect on their thought processes. NASA Task Load Index stands out among these tools, enabling the assessment of subjective load through the evaluation of six areas, including mental load  [1]. But due to a transient nature of CL, real-time measurement might provide a more accurate reflection of CL changes [1].

Physiological techniques rest on the idea that changes in cognitive function can be detected through physical signs. Mean pupil dilation has been identified as particularly telling of CL levels [2]. As task difficulty rises, so does CL, leading to pupil dilation. When engaged in mentally demanding activities, pupil dilation can occur within 1-2 seconds as a reaction to the mental load and will decrease after the task is completed. It's crucial to accurately identify the specific cause of pupil dilation since factors such as light intensity and changes in lighting conditions can influence the pupillary response [3]. Despite the potential of the method, pupil dilation metrics is not a technology adapted in the field of user experience (UX). This masters thesis aims to implement the method in the field of UX by examining how differences in cognitive load among different interfaces reflect in pupil dilation metrics and assess the possibility of combining subjective reports and pupilary measures in understanding the user's cognitive load. The aim is to develop a framework that would allow a quantitative assessment of the CL imposed by a digital interface.

In the study, 2 different interfaces with different perceived levels of required CL are selected – they are chosen in a way so the same task could be solved on both interfaces. While the individual consecutively solves the same task on the two interfaces, eyetracker monitors the pupil size. After each interface, the participant also gives a subjective feedback of the required mental effort using the NASA-TLX. Trends in pupil size and the correlation between the objective measures of the eye tracker and the subjective reports of the participants are analyzed. The hypotheses of the study predict that pupil size is a reliable indicator of CL, meaning that pupil response will vary across interfaces with different CL required,  and that objective and subjective measures of CL will be positively correlated.

References

[1] M. I. Ahmad, I. Keller, D. A. Robb and K. S. Lohan, "A framework to estimate cognitive load using physiological data," Personal and Ubiquitous Computing, pp. 1-15, 2020.

[2] F. Paas, J. E. Tuovinen, H. Tabbers and P. W. M. Van Gerven, "Cognitive load measurement as a means to advance cognitive load theory," Educational Psychologist, vol. 38, no. 1, pp. 63-71, 2003. [Online]. Available: https://doi.org/10.1207/S15326985EP3801_8.

[3] J. William and R. Murugesh, "Potential Eye Tracking Metrics and Indicators to Measure Cognitive Load in Human-Computer Interaction Research," Journal of Scientific Research, vol. 64, pp. 168-175, 2020. doi: 10.37398/JSR.2020.640137.