Infraslow Activity in Hybrid Brain–Computer Interfaces: Signals, Derived Measures, and Validity Conditions for Wearable State Characterization

Hybrid brain–computer interfaces (hBCIs) combine EEG with optical and peripheral channels and are increasingly worn outside the laboratory for passive state characterization. Activity below 0.5 Hz is largely excluded from such systems, though brain–body coupling is reported to occur there. This review asks what can be rigorously measured in that range. Some of the analytical conventions imported from faster bands assume a stable characteristic frequency that slow signals often lack. We argue that the absence of periodicity does not preclude rigorous tracking and state characterization. Analyses are organized into three tiers by the temporal assumption each makes: stable frequency, recurrence without stable frequency, and neither, with admissibility often established by test for a given recording rather than assumed from a band label. We review candidate channels by generator character, wearability and required stillness. We set out the estimators available without phase, including directional ones. We also specify hardware-related acquisition requirements for two possible approaches to analyze the infraslow band, band-limited envelopes and DC-coupled potentials. In the absence of population norms, we propose individual referencing and identify the reporting minimum a slow-band result requires.