A reproducible EEG-ERP methodology for assessing cognitive decline: integrating P300 latency and resting-state EEG frequency.

Abstract

Early detection of cognitive decline remains one of the major challenges in contemporary neurology. Although event-related potentials associated with cognitive processes, particularly the P300 component, as well as resting-state EEG analysis, have long been investigated as potential functional biomarkers of dementia, their clinical application is often limited by methodological heterogeneity and insufficient transparency of signal processing procedures. In this article, we present a reproducible EEG-ERP methodology based on proprietary software for stimulus generation, synchronization, and deterministic averaging of visual event-related potentials, combined with quantitative analysis of resting-state EEG frequency. The approach integrates precise stimulus timing, transparent offline synchronization using an audio marker recorded on the ECG channel, and a fully deterministic averaging procedure. Pilot clinical data obtained from patients with mild cognitive impairment and dementia are included to illustrate the feasibility of the proposed workflow. • Software-controlled visual stimulation with fixed inter-stimulus intervals and explicit offline synchronization using an audio marker. • Deterministic ERP averaging enabling reliable estimation of P300 latency in routine clinical EEG recordings. • Integration of resting-state EEG frequency and P300 latency as interpretable electrophysiological markers of cognitive decline.