Related Experiment Videos
Preparing computers for affective communication: a psychophysiological concept and preliminary results
Min Cheol Whang1, Joa Sang Lim, Wolfram Boucsein
1Department of Media Technology, Sangmyung University, Seoul, Korea. whang@smu.ac.kr
Human Factors
|April 2, 2004
Summary
This study explored physiological signals to help computers understand human emotions. Findings show electroencephalography and electrodermal activity can differentiate emotional states, paving the way for affective computing.
Area of Science:
- Psychophysiology
- Affective Computing
- Human-Computer Interaction
Background:
- Computers currently lack the ability to recognize and respond to human emotions.
- Understanding emotional states is crucial for developing more intuitive and responsive technology.
Purpose of the Study:
- To identify physiological parameters capable of distinguishing between four distinct emotional states.
- To explore the dimensions of pleasantness/unpleasantness and arousal/relaxation in emotional responses.
Main Methods:
- Induced emotions in 26 undergraduate students using olfactory and auditory stimuli.
- Measured changes in electroencephalographic (EEG) activity, heart rate variability, and electrodermal responses.
- Analyzed physiological data to differentiate emotional states based on valence and arousal.
Main Results:
- EEG differentiated pleasantness from unpleasantness in aroused states; electrodermal parameters were key in relaxed states.
- All measured physiological parameters (EEG, heart rate variability, electrodermal) distinguished arousal from relaxation in positive emotions.
- Electrodermal response latency was the sole differentiator between arousal and relaxation in negative emotions.
Conclusions:
- A psychophysiological approach using EEG, heart rate variability, and electrodermal measures can differentiate emotional states.
- These findings support the development of computer systems capable of affective communication.
- Integrating these physiological insights can lead to computers that better understand and respond to user emotions.