Exploring the relationships between oral sensory physiology and oral processing with mid infrared spectra of saliva

Abstract

Abstract Saliva is a critical component during eating, as it performs an important role in sensory perception and processing of food. However, the individual variations and relationships between oral sensory physiology, oral processing and saliva with food are not well understood. The aim of this study was to analyse as well as to interpret individual variations and relationships between the mid infrared (MIR) spectra of saliva and oral physiology and behaviour (e.g. fungiform papillae, saliva flow, and oral processing time). The fungiform papillae density of the tongue, resting saliva flow rate and oral processing time were measured in participants from a sensory study (n\xa0=\xa052). The MIR spectra of unstimulated saliva samples sourced from participants were collected and used to develop partial least squares (PLS) regression models. These models were utilised to interpret the relationships and individual variations between saliva quantity, oral sensing physiology, and oral processing with the saliva spectra. The coefficient of determination (R2) and the standard error in cross validation (SECV) obtained were for saliva flow rate (R2 0.65–0.83, SECV 0.25–0.31), papillae density of tongue (R2 0.73–0.83, SECV 7.4–11.2), and oral processing time (R2 0.7–0.87, SECV 1.66–3.04). Optimal models were obtained when participants were divided into groups namely young male, young female, and older participants. Different PLS loadings in the MIR fingerprint regions were observed between different groups, which demonstrates that different components of saliva connect with oral sensing physiology and oral processing differences among different demographic groups. These results indicated the potential of the salivary fingerprint spectra to understand the consumer eating experience and food selection more deeply, providing a tool in marketing and sensory experiments.