Estimating Subjective Evaluation on Drinking Experiences Using fNIRS Sensing
摘要
Understanding how the body reacts to different types of drinks could help companies develop healthier drinks while keeping customers satisfied. Functional near-infrared spectroscopy (fNIRS) can be used to provide objective information about how neurons respond to taste stimuli. In this study, we collected two datasets using two fNIRS devices to measure hemoglobin concentration changes in the prefrontal cortex and salivary gland regions during the consumption of alcoholic drinks. Machine learning models were utilized to predict the hedonic aspects of participants based on the collected data. The best model achieved 64% balanced accuracy on one of the datasets. The results suggested that the perception of taste is encoded in hemodynamic signals.