Table 2 Comparison of cognitive load estimation methods.

From: Deep knowledge tracing and cognitive load estimation for personalized learning path generation using neural network architecture

Estimation method

Data source

Computational complexity

Accuracy assessment

Self-reporting

Questionnaires, rating scales

Low (O(1))

Moderate (r = 0.6–0.7), subject to bias

Physiological

EEG, heart rate, pupil dilation, skin conductance

High (O(n+))

High (r = 0.7–0.85), requires specialized equipment

Behavioral

Eye tracking, mouse/keyboard patterns, response time

Medium (O(n log n))

Moderate to high (r = 0.65–0.8), non-intrusive

Performance-based

Error rates, completion time, secondary task performance

Low (O(n))

Moderate (r = 0.55–0.75), task-dependent

Back to article page