Multilevel Physiological Responses to the Hypersonic Effect: Concurrent Electroencephalographic, Heart Rate Variability, and Subjective Evaluation in Healthy Adults in a Participant-Blind Crossover Study

Abstract

Background and objective Sounds carrying inaudible high-frequency components (HFCs) above 20 kHz can alter brain activity, autonomic tone, and subjective state, a cluster of responses termed the hypersonic effect. However, only single physiological domains have been examined till now; therefore, whether central, autonomic, and subjective indices move together under one stimulus in the same participants remains unresolved. This study measured electroencephalographic (EEG), heart rate variability (HRV), and subjective responses simultaneously to characterize the layered response to HFC-containing sound. Methods Twenty healthy adult men completed a participant-blind crossover protocol in which the order of the two conditions was randomized and counterbalanced (10 received the wide-band sound first, 10 the audible-only sound first): an audible-band sound limited to ≤20 kHz (low frequency (LF)) and a wide-band sound containing the HFC (low frequency plus high frequency (LFHF)). Each condition comprised five minutes of pre-exposure rest, 10 minutes of sound presentation (analyzed as Audio1 and Audio2), and five minutes of post-exposure rest (Post). EEG centro-parieto-occipital (CPO) Alpha-2 relative power, normalized HRV indices, and 20 semantic-differential (SD) ratings were analyzed as changes from the pre-exposure baseline. Aligned rank transform analysis of variance (ART-ANOVA; condition × time) was used, with epoch-specific Wilcoxon signed-rank condition contrasts (Bonferroni correction, m = 3); effect sizes are the rank-biserial correlation (r). Results EEG Alpha-2 showed a trend-level main effect of condition (ART-ANOVA F = 3.59, p = 0.074), with no effect of time or condition-by-time interaction. The condition-by-time interaction was not significant; in pre-specified epoch comparisons, the condition contrast was significant during the second half of presentation (Audio2: Bonferroni-corrected p = 0.025, r = 0.657, indicating a large effect) and showed a trend after presentation (Post: corrected p = 0.089, r = 0.552). For HRV, relative low-frequency power (LF%) was significantly lower under LFHF at Post (corrected p = 0.025, r = −0.657, indicating a large effect; median difference −10.1 percentage points, 95% confidence interval −17.0 to −2.9), indicating a reduction in the relative low-frequency component of normalized HRV. None of the 20 SD items differed between conditions (all p > 0.10). Conclusions Under these conditions, HFC-containing sound was associated with a trend-level increase in posterior Alpha-2 activity, most evident in pre-specified comparisons during the second half of presentation, and a lower relative LF component of normalized HRV after exposure, without significant differences in subjective ratings. These preliminary findings suggest a possible central-to-autonomic pattern that can be quantified using noninvasive physiological recordings and support further study of HFC-containing sound as an environmental, non-pharmacological approach.