Feature extraction for psychophysiological load assessment in unconstrained scenarios

Area of Science:

• Psychophysiology
• Affective Computing
• Biomedical Signal Processing

Background:

• Psychophysiological measurements are crucial for affective computing and emotion analysis.
• Feature extraction from biosignals, especially cardiovascular data, is an ongoing challenge, particularly when autonomic nervous system stimuli are unidentified.
• Assessing psychophysiological load in unconstrained environments requires robust feature sets.

Purpose of the Study:

• To analyze a novel set of features extracted from multimodal biosignal data for psychophysiological load assessment.
• To evaluate the applicability of these features in unconstrained settings where emotional triggers are unknown.
• To determine if the proposed features can complement traditional psychophysiological measurements.

Main Methods:

• Extraction of features from multimodal biosignal data.
• Experimental evaluation using real-world data from control subjects and subjects with clinical backgrounds.
• Data collection in the context of questionnaire-based clinical history reporting.

Main Results:

• The devised feature set demonstrated promising properties for psychophysiological load assessment.
• The features are applicable in unconstrained settings, addressing the challenge of unknown stimuli.
• The proposed features show potential to enhance traditional measurement techniques.

Conclusions:

• The novel feature set offers a valuable tool for affective computing and emotion analysis.
• This approach advances the assessment of psychophysiological load in real-world scenarios.
• The findings suggest a complementary role for these features alongside existing methods in psychophysiological research.