Authors:
(1) Michal K. Grzeszczyk, Sano Centre for Computational Medicine, Cracow, Poland and Warsaw University of Technology, Warsaw, Poland;
(2) M.Sc.; Paulina Adamczyk, Sano Centre for Computational Medicine, Cracow, Poland and AGH University of Science and Technology, Cracow, Poland;
(3) B.Sc.; Sylwia Marek, Sano Centre for Computational Medicine, Cracow, Poland and AGH University of Science and Technology, Cracow, Poland;
(4) B.Sc.; Ryszard Pręcikowski, Sano Centre for Computational Medicine, Cracow, Poland and AGH University of Science and Technology, Cracow, Poland;
(5) B.Sc.; Maciej Kuś, Sano Centre for Computational Medicine, Cracow, Poland and AGH University of Science and Technology, Cracow, Poland;
(6) B.Sc.; M. Patrycja Lelujko, Sano Centre for Computational Medicine, Cracow, Poland;
(7) B.Sc.; Rosmary Blanco, Sano Centre for Computational Medicine, Cracow, Poland;
(8) M.Sc.; Tomasz Trzciński, Warsaw University of Technology, Warsaw, Poland, IDEAS NCBR, Warsaw, Poland andTooploox, Wroclaw, Poland;
(9) D.Sc.; Arkadiusz Sitek, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA;
(10) PhD; Maciej Malawski, Sano Centre for Computational Medicine, Cracow, Poland and AGH University of Science and Technology, Cracow, Poland;
(11) D.Sc.; Aneta Lisowska, Sano Centre for Computational Medicine, Cracow, Poland and Poznań University of Technology, Poznań, Poland;
(12) EngD.
Table of Links
Conclusion, Acknowledgment, and References
Limitations
The main limitation of this work is the small sample size of participants used for both developing the cognitive load detector and evaluating the cognitive effort required to complete the two survey versions. To mitigate this issue, we pre-trained the detector on a similar task, but further improvements in its performance may be possible with a larger sample size. The generalizability of our method can also be impacted by the homogeneous nature of the study participants. The volunteers were healthy individuals from a similar age group. To ensure the robustness of our method, future studies should include a more diverse and bigger sample size. Finally, we conducted experiments in controlled laboratory settings. In the future, we plan to conduct similar studies in real-world conditions[7].
In this feasibility study, we focused solely on developing automatic detection of high cognitive load, which is the first step towards estimating cognitive effort in the real world. This may not be sufficient to distinguish more subtle differences in cognitive demand caused by the mobile application. To identify the game element or combination of such elements that is the most helpful in reducing cognitive burden we would like to conduct an ablation study in the context of well-being questionnaires in our future work. We would also like to include the self-reported cognitive load after survey completion for a better understanding of survey difficulty. Moreover, it would be interesting to collect data during semi-demanding tasks, such as text reading, in order to train a more finely-graded classifier. Finally, we would like to include other tasks requiring a high cognitive load to ensure the efficiency of our method across various cognitively demanding activities.
This paper is available on arxiv under CC BY 4.0 DEED license.