Decoding the neural mechanisms of salty peptide perception via electroencephalography and machine learning

Excessive sodium intake poses major public health risks, driving the search for salt substitutes that preserve desirable flavor. Salty peptides have emerged as promising low-sodium alternatives, yet their neural perception, distinct from the ion-dependent mechanism of sodium chloride, remains poorly understood. This study integrated electroencephalography (EEG) and machine learning to elucidate the neurophysiological mechanisms of salty peptide perception. By extracting temporal, spectral, and spatial EEG features and combining them with source localization, the study identified delta-band predominance and a hierarchical cortical activation cascade extending from the prefrontal to the insular and parietal cortices, reflecting both primary gustatory processing and higher-order integration. Among the tested classifiers, XGBoost achieved the highest performance (AUC = 0.815), demonstrating that EEG features effectively distinguish neural responses between salty and non-salty peptides. These findings provide electrophysiological evidence for salty perception and offer a neural framework for reduced-sodium food design.