Detecting Learner Emotions Using Sensors
Author Information
Author(s): Lal Sharanya, Eysink Tessa H. S., Gijlers Hannie A., Veldkamp Bernard P., Steinrücke Johannes, Verwey Willem B.
Primary Institution: University of Twente
Hypothesis
Can physiological and gaze data be used to identify learner emotions in real-time?
Conclusion
The study found that specific physiological and gaze features can effectively distinguish between different learner emotions.
Supporting Evidence
- Skin conductance response peak count and tonic skin conductance levels can distinguish between emotions.
- Fixation count and duration vary significantly across emotional states.
- Saccade metrics also show significant differences based on the learner's emotional state.
Takeaway
This study shows that we can use sensors to understand how students feel while learning, which can help make learning better.
Methodology
The study involved 44 university students completing math tasks while their physiological and gaze data were collected.
Potential Biases
The reliance on self-reported data may introduce subjectivity in emotional assessments.
Limitations
The study was conducted in a controlled environment, which may not reflect real classroom conditions, and the sample was limited to Western European students.
Participant Demographics
Participants were 32 females and 12 males, aged 18-26, from 12 different nationalities.
Statistical Information
P-Value
p<0.001
Statistical Significance
p<0.001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website