This study compares the feasibility of physiological signals acquired from wearable sensors to assess building occupants' thermal comfort. Field experiments were conducted to acquire electrodermal activity (EDA), heart rate (HR), skin temperature (SKT), and thermal comfort ratings from 18 subjects in actual office settings using wristband-type wearable biosensors based on thermal preference votes. Multilevel modeling outcomes indicate that (1) there are significant differences in electrodermal level (EDL) and SKT between the “Want Warmer” and “No Change” and between the “No Change” and “Want Cooler” states across subjects; (2) while the thermal comfort significantly affects subjects' EDLs, the effects are not significant on SKT after associations with other physiological signals are accounted for. Given the diversity of building occupants and environmental conditions, findings from the comparative analysis are expected to provide a foundation for the development of a general thermal comfort model using physiological signals.
