Unravelling the neural dynamics of hypnotic susceptibility: Aperiodic neural activity as a central feature of hypnosis

Abstract

AbstractThe ability for hypnotic responding is marked by great inter-individual variation in the population, while the neural underpinning of this variability remains elusive. The current work leveraged multivariate statistics and machine learning to probe the neural dynamics underlying inter-individual differences in hypnotic susceptibility. We assessed the efficacy of linear classifiers in distinguishing between high and low hypnotic susceptible individuals using neural features from resting-state electroencephalography (EEG) both pre- and post-hypnotic induction. Our focus encompassed both aperiodic and periodic components of the power spectrum, and graph theoretical measures derived from functional connectivity patterns. Several neural features from both pre- and post-induction significantly differentiated susceptibility levels, which underscores the complex dynamics of hypnotic phenomena. Based on model comparisons and feature ranking, we discerned the pre-induction aperiodic exponent as the primary discriminating neural feature, while periodic activity did not differ between groups. This novel finding not only resonates with the increasing emphasis on this neural component in broader EEG research but also promotes the idea that the primary neural distinction in hypnotic susceptibility is evident at baseline, even before hypnosis. Based on prevailing interpretation of aperiodicity in the EEG signal, our findings support the idea that hypnotic susceptibility might be an inherent trait reflected in the balance of cortical excitation and inhibition.Significance StatementHypnotic phenomena reflect the ability to alter one’s subjective experiences based on targeted verbal suggestions. While research has made strides in understanding the cognitive aspects underlying the variability in one’s susceptibility to hypnosis, the brain correlates remain largely elusive. Addressing this gap, our study employs multivariate pattern classification and machine learning to predict hypnotic susceptibility. By recording electroencephalography (EEG) before and after hypnotic induction and analyzing diverse neurophysiological features, we identify several features that differentiate between high and low hypnotic susceptible individuals for both pre- and post-induction periods, which underscores the multifaceted nature of hypnotic phenomena. Our analysis revealed that the paramount discriminative feature is arrhythmic (i.e., non-oscillatory) EEG activity prior to the induction—a novel discovery in the field. This finding, that the chief EEG marker is observed even before hypnosis, aligns with the idea that hypnotic susceptibility represents an inherent trait reflecting the balance between excitation and inhibition in neural activity.