Salivary Biosensing Opportunities for Predicting Cognitive and Physical Human Performance

Area of Science:

• Biomedical Engineering
• Physiology
• Human Performance

Background:

• Biosensing technologies enable real-time, non-invasive monitoring of salivary biomarkers.
• Salivary analysis holds potential for personalized medicine and public health applications.
• Identifying key salivary analytes is crucial for predicting physiological status and performance.

Purpose of the Study:

• To evaluate the predictive value of 12 salivary analytes for cognitive and physical performance outcomes.
• To assess the utility of salivary biomarkers in military personnel under stressful conditions.
• To determine if salivary analytes can forecast performance in critical tasks like moving, shooting, and communication.

Main Methods:

• Collected time-series data for 12 salivary analytes (hormones, metabolites, enzymes) from 115 military personnel.
• Quantified analyte data using various time-series features.
• Applied multiple regression models (Elastic Net, PLS, Random Forest) to predict performance outcomes.

Main Results:

• Predictive performance was consistently poor across all models, with near-zero R-squared values.
• Limited evidence suggested stable or meaningful predictions from salivary analytes.
• No single analyte reliably predicted performance, although some features like post-peak slopes showed minor frequency.

Conclusions:

• Salivary biomarkers alone are insufficient for robust prediction of cognitive and physical performance.
• Future research should integrate salivary data with other biosensors and contextual variables for improved accuracy.
• The current study indicates limitations in using salivary analytes for real-time performance monitoring in demanding environments.