Air-conducted ultrasound below the hearing threshold elicits functional changes in the cognitive control network

Abstract

Air-conducted ultrasound (> 17.8 kHz; US) is produced by an increasing number of technical devices in our daily environment. While several studies indicate that exposure to US in public spaces can lead to subjective symptoms such as ‘annoyance’ or ‘difficulties in concentration’, the effects of US on brain activity are poorly understood. In the present study, individual hearing thresholds (HT) for sounds in the US frequency spectrum were assessed in 21 normal-hearing participants. The effects of US were then investigated by means of functional magnetic resonance imaging (fMRI). 15 of these participants underwent three resting-state acquisitions, two with a 21.5 kHz tone presented monaurally at 5 dB above (ATC) and 10 dB below (BTC) the HT and one without auditory stimulation (NTC), as well as three runs of an n-back working memory task involving similar stimulus conditions (n-ATC, n-BTC, n-NTC). Comparing data gathered during n-NTC vs. fixation, we found that task performance was associated with the recruitment of regions within the cognitive control network, including prefrontal and parietal areas as well as the cerebellum. Direct contrasts of the two stimulus conditions (n-ATC & n-BTC) vs. n-NTC showed no significant differences in brain activity, irrespective of whether a whole-brain or a region of interest approach with primary auditory cortex as the seed was used. Likewise, no differences were found when the resting-state runs were compared. However, contrast analysis (n-BTC vs. n-ATC) revealed a strong activation in bilateral inferior frontal gyrus (IFG, triangular part) only when US was presented below the HT (p < 0.001, cluster > 30). In addition, IFG activation was also associated with faster reaction times during n-BTC (p = 0.033) as well as with verbal reports obtained after resting-state, i.e., the more unpleasant sound was perceived during BTC vs. ATC, the higher activation in bilateral IFG was and vice versa (p = 0.003). While this study provides no evidence for activation of primary auditory cortex in response to audible US (even though participants heard the sounds), it indicates that US can lead to changes in the cognitive control network and affect cognitive performance only when presented below the HT. Activation of bilateral IFG could reflect an increase in cognitive demand when focusing on task performance in the presence of slightly unpleasant and/or distracting US that may not be fully controllable by attentional mechanisms.

Validity of the 1984 Interim Guidelines on Airborne Ultrasound and Gaps in the Current Knowledge

“…No substantial scientific advances have been made since 1984 regarding the endpoint of ear damage mediated by the putative frequency down conversion mechanism. However, more data on the distribution of hearing thresholds in the population (up to the 30 kHz TOB) and some data on non-specific symptoms have become available over the years (Henry and Fast 1984; Ashihara et al 2006; Ashihara 2007; Kurakata et al 2013; Rodríguez Valiente et al 2014; Kühler et al 2019; Weichenberger et al 2022). As outlined above, there is some evidence that non-specific symptoms can occur in sensitive individuals upon exposure to ultrasound in the 20 kHz TOB at SPLs below the exposure limit designed to protect from such effects.…”

“…No quantitative evaluation of non-specific symptoms was performed, but almost all of the test subjects that heard the applied ultrasound, described the hearing sensation as displeasing. In a more recent single-blinded experimental study that applied a sequence of sham-exposure, ultrasound exposure 5 dB above the individual hearing threshold level (HTL) and ultrasound exposure 10 dB below the individual HTL (both at 21.5 kHz, the order of ultrasound exposure levels being randomized), statistically significant higher ratings of unpleasant perception upon ultrasound exposure above individual HTL compared to ultrasound exposure below HTL were reported (Weichenberger et al 2022).…”

“…If ultrasound hearing thresholds are measurable, they might be a reasonable proxy for the lower boundary of the possible onset of non-specific symptoms due to ultrasound exposure. To date, there has been only one investigation on how the appearance of self-reported effects (in this study pleasantness/unpleasantness and perceived impact on cognitive tasks) is related to the HTL (Weichenberger et al 2022) and there are no investigations on the impact of psychoacoustic parameters like loudness, sharpness and fluctuation strength, and related sound field properties. In order to validate the working hypothesis that the individual hearing threshold can indeed be used as a proxy for the lowest possible threshold for developing non-specific symptoms, more psychoacoustic research with audible ultrasound is needed, although the lack of blinding will always remain an issue.…”

“…It remains unclear if this finding is due to limited sensitivity of the applied methods. In a follow up to Kühler et al (2019), Weichenberger et al (2022) used fMRI to investigate brain activity during resting-state as well as during cognitive processing in subjects with normal hearing exposed to a 21.5 kHz tone. This study also found no evidence of auditory cortex activation for airborne ultrasound below and above the hearing threshold, but found a higher n-back-task-related activation of the inferior frontal gyrus (IFG) in the below threshold condition compared to the above-threshold condition, as well as an association between higher IFG activation and faster reaction times during the below threshold exposure condition.…”