Neural variability reliably and selectively encodes pain discriminability

Abstract Neural activity varies dramatically across time. While such variability has been associated with cognition, its relationship with pain remains largely unexplored. Here, we systematically investigated the relationship between neural variability and pain, particularly pain discriminability, in five large electroencephalography (EEG) datasets (total N = 489), collected from healthy individuals (Datasets 1–4) and patients with postherpetic neuralgia (PHN; Dataset 5) who had received painful or nonpainful sensory stimuli. We found robust correlations between neural variability and interindividual pain discriminability. These correlations were (1) replicable in multiple datasets, (2) pain selective, as no significant correlations were observed in nonpain modalities, and (3) clinically relevant, as they were partly disrupted in patients with PHN. Importantly, variability and amplitude of EEG signals were mutually independent and had distinct temporal and oscillatory profiles in encoding pain discriminability. These findings demonstrate that neural variability is a replicable and selective indicator of pain discriminability above and beyond amplitude, thereby enhancing the understanding of neural encoding of pain discriminability and underscoring the value of neural variability in pain studies. Competing Interest Statement The authors have declared no competing interest.