BACKGROUND: We investigated the association between path-integration (PI) errors related to entorhinal cortex function detectable using a 3D virtual reality (VR) navigation system and various biomarkers to explore its potential as an early AD indicator. METHODS: The PI capabilities of 111 healthy adults were assessed using a head-mounted 3D VR system. Demographic and cognitive assessments, AD-related plasma biomarkers, and apolipoprotein E genotypes were also evaluated. Predictive factors for PI errors were identified using multivariate linear regression, logistic regression, and random forest. RESULTS: PI errors positively correlated with age, plasma levels of glial fibrillary acidic protein [GFAP], neurofilament light, and p-tau181. Multivariate analysis identified plasma GFAP and p-tau181 levels as significant predictors. Random forest analysis and receiver operating characteristic curves underscored plasma p-tau181 levels as the most substantial predictor. DISCUSSION: PI errors, particularly in conjunction with plasma p-tau181 levels, could reflect early AD pathophysiology, highlighting their potential as early biomarkers.