Fig. 2.

Thirst scale and physiological indices under different hydration statuses. a Experimental design of our study. Each subject underwent brain magnetic resonance imaging (MRI) scanning and thirst-related physiological measurements three times within 13 h. The first examination was carried out at the beginning of the experiment, where the participant was in a normohydration state without any limitations on food and fluid intake in the past several hours. After that, the participant was prohibited from eating or drinking anything with water content for 12 h to achieve a dehydration status before participation in the second examination. Subsequently, the participant was asked to drink 1.0 L water for rehydration and completed the final examination within one hour. These examinations include brain MRI scanning and physiological status of hydration evaluation. b The VAS showed significant changes among the three statuses in repeated-measures ANOVA. c All physiological indices of hydration were normalized to z scores and then pooled into a principal component analysis (PCA) to generate factors that predicted overall hydration physiological status. Two major components with principal eigenvalue scores >1 were extracted and defined: the dehydration-correlated index (DCI) and the dehydration-anticorrelated index (DAI). Both DCI and DAI showed significant changes among the three statuses in repeated-measures ANOVA. d The radar plot indicates the principal component coefficients for DCI and DAI in each physiological index. (* p value < 0.05, *** p value <0.000, FDR corrected)