Presentation
Refining Mathematical Task Difficulty for Accurate Mental Workload Estimation
DescriptionAccurately assessing mental workload (MWL) is essential for optimizing task performance. Although mathematical operations are widely used to induce MWL due to their scalability, few studies provide clear criteria for defining difficulty levels. This study proposes a systematic classification method based on cognitive demand, defined by the number of interim values stored during mental calculation stages. Six math difficulty levels were developed and tested in a controlled experiment with 26 participants. Task performance and eye-tracking metrics were recorded. Results showed that the number of digits and carrying operations significantly influenced performance, particularly for simpler problems. However, for complex problems, increased operations did not yield distinguishable MWL levels, suggesting cognitive resource saturation. Four distinct difficulty levels emerged based on performance, while eye-tracking data revealed fewer significant differences. The average number of interim values stored per stage better predicted MWL than total interim values, forming three cognitive demand tiers. These findings suggest that not all mathematical problems elicit distinguishable MWL differences and highlight the importance of task structure in MWL classification. This work contributes to refining MWL assessment methodologies and supports the development of better cognitive workload modeling using structured math problems.
Event Type
Lecture
TimeThursday, October 16th4:50pm - 5:10pm CDT
LocationGrand Hall M/N
Augmented Cognition