https://orcid.org/0000-0002-2000-5357
Figures
Abstract
Hearing protection devices (HPDs) remain the first line of defense against hazardous noise exposure and noise-induced hearing loss (NIHL). Despite the increased awareness of NIHL as a major occupational health hazard, implementation of effective hearing protection interventions remains challenging in at-risk occupational groups including those in public safety that provide fire, emergency medical, or law enforcement services. A reduction of situational awareness has been reported as a primary barrier to including HPDs as routine personal protective equipment. This study examined the effects of hearing protection and simulated NIHL on spatial awareness in ten normal hearing subjects. In a sound-attenuating booth and using a head-orientation tracker, speech intelligibility and localization accuracy were collected from these subjects under multiple listening conditions. Results demonstrate that the use of HPDs disrupts spatial hearing as expected, specifically localization performance and monitoring of speech signals. There was a significant interaction between hemifield and signal-to-noise ratio (SNR), with speech intelligibility significantly affected when signals were presented from behind at reduced SNR. Results also suggest greater spatial hearing disruption using over-the-ear HPDs when compared to the removal of high frequency cues typically associated with NIHL through low-pass filtering. These results are consistent with reduced situational awareness as a self-reported barrier to routine HPD use, and was evidenced in our study by decreased ability to make accurate decisions about source location in a controlled dual-task localization experiment.
Citation: Snapp HA, Millet B, Schaefer-Solle N, Rajguru SM, Ausili SA (2023) The effects of hearing protection devices on spatial awareness in complex listening environments. PLoS ONE 18(1): e0280240. https://doi.org/10.1371/journal.pone.0280240
Editor: Hussain Md Abu Nyeem, Military Institute of Science and Technology, BANGLADESH
Received: May 12, 2022; Accepted: December 26, 2022; Published: January 12, 2023
Copyright: © 2023 Snapp et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results presented in the study are available from a public repository https://scholarship.miami.edu/esploro/outputs/991031721719402976.
Funding: This work was supported by the University of Miami Laboratory for Integrative Knowledge Grant and the University of Miami Clinical and Translational Science Institute (CTSI). Miami CTSI is supported by National Center For Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR002736 (HS,BM,NS,SR). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
I. Introduction
Noise-induced hearing loss (NIHL) is the second most common cause of hearing loss after age-related hearing loss [1], with noise exposure now recognized globally as the leading cause of acquired hearing loss in adults. In fact, 24% of hearing difficulty among U.S. workers is caused by occupational exposures making hearing loss the most common occupational impairment in the U.S. [2]. Despite increasing awareness of NIHL as a major occupational health problem, implementation of effective hearing protection interventions remains challenging [3–5]. Reduction of situational awareness has been reported as a primary barrier to hearing protection device (HPD) use in at-risk populations such as the fire service and military [3, 6–10].
Accurate sound localization contributes to situational awareness and influences rapid decision making. HPDs both reduce overall access to sound and disrupt access to auditory cues responsible for key spatial orientation tasks such as localization accuracy and front-back distinction [11–15]. Occupational NIHL typically reduces access to high-frequency spectral cues, which may serve to further disrupt localization abilities [16–19]. Disruption of these cues may not only compromise localization, but also disorient the listener and affect reaction times. The ability to quickly locate and orient to targets of interest is essential to safety and job performance for several emergency response workers (ERWs) focused on public safety such as police officers, correction officers, firefighters, and emergency medical response workers who regularly respond to life-threatening calls to action under stressful and hazardous conditions.
Reduced situational awareness has been identified as one of the major causes of workplace injuries and fatalities [20–25]. Moreover, the importance of spatial perception for worker safety has been emphasized by the Occupational Safety and Health Administration for several occupations [26–29]. Precise and timely analysis of the environment is critical in emergency response occupations. ERWs work in challenging, hazardous situations and must manage multiple tasks simultaneously. This is further complicated by the need to detect and identify the location of a sound source, most often with interfering signals which create poor or even negative signal-to-noise ratios (SNRs). For this reason, HPD use continues to be perceived by workers as increasing risk for operational injuries and fatalities [10, 30].
A large body of research points to the importance of auditory localization for situational awareness and orientation, and previous studies have shed light on potential factors which may influence localization abilities in emergency response situations. This includes findings that stimuli are better localized when they are broadband [31, 32], longer in length [33], and salient [34, 35]. While prior studies have explored the impact of HPDs and NIHL on spatial perception, an important limitation is the common use of impulse or short duration stimuli [15, 36–43], which lacks ecological validity for many non-combat related emergency response situations. Public Safety operations often entail communication tasks that require the listener to monitor, process, and respond to two or more simultaneous speech streams [44]. Continuous monitoring of multiple channels of communication is an essential duty of ERWs, making performance in multi-talker environments of critical importance for those working in public safety. For example, listeners will employ active head movement to make use of the interaural difference cues required for localizing an auditory event in space [45, 46]. Competing noise is known to negatively affect localization accuracy [47]. In multi-talker environments, uncertainty of the target signal [48], would be expected to affect localization accuracy. At the same time, there is an expected benefit of the spatial separation of the target from the masker. In competing noise, listeners may re-orient themselves to improve the SNR, and in turn, enhance their performance [49, 50]. However, re-orienting as a strategy is typically ineffective for short duration stimuli [50, 51]. Moreover, studies show that the greatest effect of spectral cue disruption on localization is at +/- 90 degrees azimuth and that localization is minimally affected by NIHL for targets directly in front of the listener [16]. Re-orienting to the target may require greater effort, but may also allow for improved accuracy. Further, there is little information on the effect of other competing tasks on auditory localization, inherent in essential duties of ERWs.
We are interested in studying behaviors that listeners may employ to improve their performance under challenging listening scenarios where they may be required to perform more than one task simultaneously. The objective of the current study was to investigate the effect of HPDs and NIHL on spatial perception in a dual-task experiment. Here, we use both spatial and speech paradigms to investigate localization abilities across task modality. In addition to HPDs, we employed low-pass filtering to simulate NIHL in normal hearing listeners to allow for within-subject comparison of how limited frequency bandwidth would affect localization, speech intelligibility, and reaction times.
II. Methods
This research was conducted according to the principles expressed in the Declaration of Helsinki and was approved by the Institutional Review Board at the University of Miami (IRB#20210223).
A. Participants
Ten normal hearing adults participated in this study. The participants ranged in age from 20 to 46 years (M = 35.8, SD = 6.9). There were 2 males and 8 females. All participants had normal hearing bilaterally, as determined by air conduction hearing thresholds of < 20 dB HL across the independent standard audiometric test frequencies, 250–8000 Hz, and no prior history of hearing impairment. Written consent was obtained for all participants.
B. Experimental setup
The experiment took place in a sound attenuated auditory booth (4.3 x 4.3 x 2m). During the task, listeners sat in the center of a 24-speaker circular array (Mix Cube, Avantone, NY, USA) spanning 360° with a radius of 1.3m (Fig 1). Sound presentation was driven by a 24-channel sound card (MOTU 24Ao, Cambridge, MA, USA), connected to three amplifiers of eight channels each (Crown CT 875, Los Angeles, CA, USA). Stimuli presentation and the analysis of the responses were implemented using custom script written in MATLAB (ver. R2020b, The MathWorks, Natick, USA), the Psychophysics Toolbox extension, and the Lab Streaming Layer library for device time synchronization (https://github.com/sccn/labstreaminglayer). A lightweight, non-intrusive, custom head tracker with laser pointer and 9-axis inertial measurement unit fixed to a three-dimensional printed spectacle frame was used to obtain head rotations in response to stimuli.
A) Participants were seated in the center of a 24-speaker array. The 12 speakers used to present the stimuli are indicated by their location in azimuth at a speaker-to-speaker distance of 30°. A single talker was presented from each quadrant for every trial, indicated by the grey dotted lines. B) The 6 listening conditions (normal hearing, with low-pass filtering, and with over-the-ear (OTE) or in-the-ear (ITE) HPDs, see section E) and the low-pass filtering applied to simulate NIHL are visualized.
C. Stimuli
Stimuli consisted of target sentences from the coordinated response measure (CRM) corpus (36). All CRM sentences have the same structure: “ready (call-sign) go to (color) (number) now”, and consist of eight possible call signs (Arrow, Baron, Charlie, Eagle, Hopper, Laker, Ringo, Tiger), four colors (blue, red, green, white), and eight numbers (1–8). These combinations yield 256 different sentences recorded for eight different talkers giving a total of 2048 sentences in the corpus. The CRM stimuli are broadband in nature and embedded in multiple competing speech streams, allowing for the investigation of how listeners may leverage available spatial perception cues in complex environments while managing more than one task simultaneously (i.e., target identification, target localization, and speech intelligibility).
D. Procedures
Four speech streams, randomly selected from the CRM corpus [36], were presented simultaneously to the listener through four speakers in a 360° horizontal array, with each quadrant represented in every trial (Fig 1A). Twelve of the 24 visible speakers separated by 30° were used to present the stimuli (Fig 1A). Participants were not made aware of which speakers the stimuli were presented from or the number of speakers used in the experiment.
Listeners were instructed to listen for the target talker, which was identified by the presence of the call sign “Hopper” amidst three different interfering talkers. All talkers were male. Target stimuli were randomly presented at +5 and +10 SNR with the interfering talkers presented at 70dBA. Pilot data collected before the primary experiment indicated these SNRs appropriately minimized performance ceiling and floor effects. The stimulus and competing speech levels were verified at the ear level of the listener with signals calibrated individually to ensure the designated SNR as measured at the location of the listener. The sentences in the speech corpus were synchronized at their onsets (“Ready”) [52].
The experimental task required the listener to 1) locate the source of the target talker in azimuth, and 2) report (identify and recall) the associated color/number (e.g., “Blue” “One”). During the task, participants sat comfortably in a chair and were instructed to point a head-mounted LED towards the perceived sound location via head movement. Head orientation and velocity of the head movement were recorded during sound presentation and analyzed offline to obtain localization accuracy. Prior to each trial, subjects had to fixate at 0° azimuth directly in front of the participant to ensure proper head orientation (calibrated using the head tracker).
E. Listening conditions
Listeners were tested using both over-the-ear (OTE) and in-the-ear (ITE) HPDs. E-A-Rsoft™ FX™ foam earplugs (3M™ New Zealand Pty, Limited), with a Noise Reduction Rating (NRR) of 33 dB were used for the ITE listening conditions. At the tolerance of the participant, deep insertion of the plug was obtained by ensuring the outside edge of the earplug was flush with the entrance of the ear canal [53]. PELTOR™ Optime™ 105 Earmuffs (3M™ New Zealand Pty, Limited), with an NRR of 30 dB were used for the OTE listening conditions.
Since individuals with NIHL tend to experience hearing loss at 3000 Hz and above, the signal was low-pass filtered using a cutoff frequency of 2.5 kHz with an order of 100 so that the audibility of the filtered speech was representative of individuals with impaired high-frequency hearing arising from noise exposure. The experiment was conducted with six listening conditions using a within-subjects design (shown in Fig 1B): 1) normal-hearing (NH), 2) NH with low-pass filter (NH-LPF), 3) OTE, 4) OTE with LPF (OTE-LPF), 5) ITE, and 6) ITE with LPF (ITE-LPF). Each listening condition consisted of 96 trials for a total of 576 trials, with 288 presented at +5 dB SNR and 288 presented at +10 dB SNR randomly in an interleaved manner. All listeners started in the NH listening conditions. HPD listening conditions were counterbalanced (i.e., half the participants began with the OTE listening conditions and half with the ITE listening conditions). The LPF conditions were always anchored to their associated hearing condition (i.e., OTE to OTE-LPF) so as not to introduce intra-subject variability by disrupting the positioning of the HPD and thus attenuation at the ear.
F. Data analysis
Speech intelligibility was calculated as the percentage of color and number stated by the target talker that were correctly repeated by the participant. For analysis, percent correct scores were converted to rationalized arcsine units (RAU) [54]. The rationalized arcsine transform produces a scale with units that have almost the same size as percentages while satisfying the assumptions of statistical procedures that are used to analyze the scores. The arcsine transform is used to express percent correct scores in radians (Eq 1) and the rationalized arcsine transform adjusts the scores into units for analysis (Eq 2):
(1)
(2)
where X is the total number of colors and numbers that were correctly repeated by the participant and N is the number of trials performed.
Head-orienting responses were characterized by a saccadic profile (i.e., rapid, step-like movement of monophasic velocity). Head-orienting responses were automatically detected using a custom-made MATLAB script that identified head velocities exceeding 20°/s (Fig 2A). This analysis determines onset (vel. > 20°/s, start of head movement) and offset (vel. < 20°/s, when the head is static again). These markers were visually checked offline by the experimenter on a trial-by-trial basis. The perceived target’s speaker location, or end point, was determined by the offset head’s position (in degrees, Fig 2). The overall measure of the localization response accuracy was processed by computing the mean absolute error (MAE) across trials as follows:
(3)
where R is the response and T is the target (in degrees). The reaction time (RT) relative to the stimulus presentation is defined by the onset of the head rotation (in sec, Fig 2). For quantitative analysis, RT data were transformed to its reciprocal, known as the response promptness (in s-1), which is shown to follow a Gaussian distribution [55].
A sample target stimulus from the CRM corpus is presented with its associated waveform. The target call sign used was “hopper” and the associated color/number to be repeated are indicated in green. The upper panel shows the head position in azimuth (degrees) and lower panel shows the velocity of the head movement (degrees/second). RT is calculated as the time between the onset of the stimulus and initiation of the head movement. The endpoint of the head movement is set once its velocity profile falls below 20°/s (grey shaded bar).
Dependent variables included speech intelligibility, localization, and response promptness. Results were analyzed using repeated measures analysis of variance (RM-ANOVA) with three within-subject factors (listening condition [NH, NH-LPF, OTE, OTE-LPF, ITE, ITE-LPF], SNR [+10 dB, +5 dB], and hemifield presentation [front, back]). A significance level of α = 0.05 was applied. Pairwise comparisons with Bonferroni corrections for multiple comparisons were used to follow-up significant main effects. Correlation analysis was also performed to evaluate the predictive value of the independent variables. All analyses were performed using SPSS® software (version 26.0; New York: IBM Corp®).
III. Results
A. Speech intelligibility
Listeners performance for the speech intelligibility task is presented in Fig 3A. Analyses revealed significant main effect of hemifield [F(1,9) = 22.86, p < .001, ηp2 = .718], a significant main effect of SNR [F(1,9) = 424.19, p < .001, ηp2 = .979], and a significant main effect of listening condition [F(5,45) = 6.99, p < .001, ηp2 = .437] on listeners’ ability to correctly identify the color and number stated by the target talker. Specifically, there was a significant difference in performance when signals were presented from the front compared to rear presentation (Mean difference = 15.1 RAU, p < .01), and between +10 dB SNR and +5 dB SNR (Mean difference = 27.3 RAU, p < .001), with performance in the rear and at +5 dB SNR being significantly worse. Pairwise comparisons revealed a mean difference of 16.1 RAU between the NH condition and the ITE-LPF condition (p = .01). There was a significant interaction between hemifield and SNR [F(1,9) = 6.86, p < .05, ηp2 = .437], suggesting that the effect of hemifield on speech intelligibility was greater at the reduced SNR, and that the effect of signals presented from behind on speech intelligibility was greater at the reduced SNR. There were no interactions of hemifield and listening condition (p = .09), SNR and listening condition (p = .41), or hemifield by SNR by listening condition (p = .60). Pairwise comparisons revealed a mean difference of 20.0 RAU between the NH condition and the ITE condition (p < .001) and 20.8 RAU between the NH condition and the ITE-LPF condition (p < .05) when signals were presented in the front at +5 dB SNR.
Here, lower MAE values indicate better localization performance. Performance for targets presented to the front is shown in blue and to the back in red. Shaded areas show standard error of the mean.
B. Localization
Fig 3B presents the localization performance. Analyses revealed a significant main effect of hemifield [F(1,9) = 10.55, p = .01, ηp2 = .54], a significant main effect of SNR [F(1,9) = 63.68, p < .001, ηp2 = .876], and a significant main effect of listening condition [F(5,45) = 4.94, p < .05, ηp2 = .354]. In general, listeners’ localization of the target talker was better (i.e., lower MAE) for signals presented to the front (hemifield Mean difference = 19.42°, p = 0.01) and at the more favorable +10 dB SNR (Mean difference = 12.17° p < .001). This result extends the existing literature on localization abilities where best performance occurs for stimuli located in the front hemifield [56, 57], and at higher SNRs [35, 47]. There was also a significant interaction between hemifield and listening condition [F(5,45) = 3.33, p = .01, ηp2 = .270]. Pairwise comparisons revealed significant differences in localization abilities for signals presented to the front from the NH to OTE listening condition (p = .001), NH to OTE-LPF listening condition (p < .05), and NH-LPF to OTE listening condition (p < .01) at +10 dB SNR; NH to OTE-LPF listening condition (p < .01), NH-LPF to OTE listening condition (p < .01), and NH-LPF to OTE-LPF listening condition (p < .05) at + 5 dB SNR. No significant differences by listening condition were observed for signals presented from the rear. Furthermore, there were no significant interactions between hemifield and SNR (p = .227), SNR and listening condition (p = .91), or hemifield by SNR by listening condition (p = .41).
C. Promptness
Response promptness (the reciprocal of RT) for SNR and listening condition is shown in Fig 4. Analyses revealed a significant main effect of SNR [F(1,9) = 26.389, p < .001, ηp2 = .746] and hemifield [F(1,9) = 16.606, p < .01, ηp2 = .649]. There was no significant effect of listening condition [F(5,5) = .808, p = .59, ηp2 = 447]. RTs slowed when signals were presented from the rear compared to the front, and at +5 dB SNR compared to +10 dB SNR. There were no significant two-way interactions between hemifield and SNR (p = .121), hemifield and listening condition (p = .194), SNR and listening condition (p = .549), or three-way interaction among hemifield, SNR, and listening condition (p = .553).
Shaded areas show standard error of the mean. Participants with measurable RTs are indicated by circles. Those participants whose average RTs across listening conditions exceeded a cutoff of 1.5 s are indicated by squares.
We identified a subset of subjects (n = 3) that showed lower overall values (slower responses) than other respondents. The overall mean CRM sentence length was ~1.8 s. Closer review of the data revealed that these subjects waited until the color/number of the sentence was presented to then orient the head towards the perceived target location. This behavior was irrespective of condition, SNR, or hemifield. Fig 4 presents this difference in behavioral pattern for those subjects having an overall reaction time <1.5s (circles, Fig 4) and >1.5s (squares, Fig 4). As expected, there is no difference in promptness of the response for slow subjects for hemifield, SNR, or any conditions. In those who were fast, a significant difference in mean response was observed for hemifield (.25 s, p < .01) and SNR (.21 s p < .001), but not for any of the listening conditions, consistent with the overall results.
D. Influencing factors on performance
Although hemifield and SNR influenced speech intelligibility, Pearson correlation analysis revealed that localization accuracy did not influence speech intelligibility. There were also no significant correlations between the promptness of the response and localization, or promptness of the response and speech intelligibility. Combined, these results suggest that speech intelligibility is not influenced by how fast or accurate listeners are in identifying and locating the target. Yet, a closer review of the response patterns revealed different behavioral strategies between listeners.
Fig 5 shows a representative example of 3 listeners’ strategies for speech intelligibility, localization, and promptness. Interestingly, the listener whose promptness increased (see circles, Fig 5) as conditions became more challenging (i.e., lower RTs) also had the smallest localization error (MAE) and the best speech intelligibility (% Correct) in the +10 dB SNR condition. This does not hold true at +5 dB, suggesting that this strategy may not be effective in increasingly poorer SNRs. This effect has been observed in other domains where individuals who take longer to initiate a response outperform those who are quick to respond [58]. Longer processing times have been attributed to “information gathering”, which services to improve accuracy [59]. The second participant’s behavior (see squares, Fig 5) compromised speech monitoring and decreased promptness (increased reaction times), revealing decreased ability to manage dual tasks when compared to the other listeners and in the NH condition. The strategy of the third participant (see triangles, Fig 5) indicates a tradeoff between reaction time and accuracy, and prioritization of accuracy of the word task over localization.
(A) accuracy for color and number identification, (B) MAE (localization performance), and (C) promptness (inverse of RT) are shown for participant 1 (O), 2 (□), and 3 (Δ). Behavior in +10 dB SNR is shown in the left column and +5 dB SNR in the right column of the figure.
IV. Discussion
This study explored the impact of HPDs and NIHL on spatial perception. Results indicate that listeners are highly accurate at both localization and speech monitoring under the NH condition. Localization performance was negatively affected by HPDs, with significant effects observed in the OTE and OTE-LPF listening conditions when compared to the NH condition. Previous studies have demonstrated increased localization errors with loss of spectral cues either through the use of a low-pass filter [60–62], or through the use of HPDs [6, 11, 13, 14, 37, 39]. NIHL was simulated by removal of spectral information from the speech signals using a high order low-pass filter set at 2.5 KHz. While performance decreases under simulated NIHL conditions using low-pass filtered speech streams, the change is not significant for any measure and marginal compared to the changes observed with the OTE HPDs. This may be attributed to use of a longer speech stream compared to short noise bursts or single words [15, 36–43], which allows the listener to apply a search strategy to locate the target. Head movements are known to assist the listener in resolution of ambiguities in source location [50, 61–63]. The length of the CRM stimuli provided listeners with ample time to initiate head movements to disambiguate the spatial acoustic cues for localization of the target. Listeners are thereby able to employ inherent behavioral strategies to combat cue disruption arising from occlusion, loss of spectral cues, or competing sound sources. Investigating the role of different ecologically valid sources of interest provides new insights into the barriers of HPD use in emergency response operations, and how the disruption of spatial perception cues through HPDs or NIHL affects task performance. Specifically, impulse stimuli may not accurately reflect performance or spatial perception abilities when continuously monitoring multiple channels of communication.
This experiment employed a dual-task design where the listener had to actively locate the target signal and repeat the associated color and number from the target sentence. Target location was presented at random, unlike in previous studies where the target for speech-in-noise was always presented from a fixed and known location directly in front of the listener [42, 43]. The resulting cues available to the listener varied with each stimulus presentation, as the target could be presented anywhere in the 360° array. The ability to move the head during the stimuli presentation in the present experiment, likely allowed the listeners to leverage available interaural cues to facilitate locating the target and improve the SNR to optimize speech intelligibility.
As seen in Fig 3, the use of HPDs seems to be more disruptive than the low-pass filter alone (NH-LPF). There was no significant effect of filtering out high frequency spectral information from the speech stimuli on localization abilities in the azimuthal plane for any of the conditions. These findings suggest that the disruption caused by OTC HPDs may exceed that which arise from NIHL under certain scenarios, specifically at favorable SNRs, and/or for longer stimuli such as speech streams. Targets presented in elevation may have been more disrupted by the low-pass filtered speech, although head movement has been shown to reduce errors in elevation as well [64, 65]. Additionally, while overall MAE increases in the ITE HPD conditions, the change in MAE from the NH condition did not reach significance. Others have found that OTE HPDs are more disruptive to localization abilities than ITEs [13, 39, 66], and can be attributed to the additional loss of pinna cues which provide direction-dependent filtering for high frequency signals. Pinna cues are essential for localization in elevation [67], although, our findings present a clear affect in azimuth. Others have also demonstrated disruption of localization in the azimuthal plane when spectral cues are perturbed [68]. Here, even when the signal is longer in duration, allowing for the listener to re-orient to improve their performance, the ability to make accurate decisions about the location source in azimuth is negatively affected by HPDs. This affect is amplified as the competing signals create greater interference with the target (i.e., +5 dB SNR).
Results indicate that OTE HPDs are disruptive to localization abilities. This is consistent with that reported by ERWs [3, 6–8]. However, targets presented in the front hemifield at a favorable SNR are fairly well located when considering the complexity of the tasks. The MAE for signals presented to the front is < 30° (Fig 3), which may allow for sufficient performance on tasks such as search and rescue, particularly when integrated with other inputs (i.e., audiovisual integration). That is that the degree of error, although insufficient for the localization of impulse stimuli such as gunfire [43], is such that a listener could presumably locate the region or relative area of the source when presented in the front hemifield. Seemingly, a search to locate strategy could then be effective in quickly locating a target. The timing and effort required for signals presented from the rear, however, is perhaps too great for head orientation to facilitate localization of acoustic streams. Interestingly, there were no significant findings with ITE HPDs. This may be due to the small sample size in combination with the large variance in MAE across these conditions.
The ability of listeners to correctly identify and recall the associated color and number spoken by the talker using the call sign “Hopper” decreased when using HPDs, with significant decrease in performance for the ITE-LPF condition (Fig 3A). Speech intelligibility was also observed to be consistently better when the target originated in front of the talker compared to the rear. Larger differences in speech intelligibility were observed in the +5 dB SNR when compared to the +10 dB SNR across the various listening conditions, possibly pointing to the increase in cognitive demand that comes with making the SNR more difficult. These results indicate that disruption of spectral cues has a much greater effect on localization and speech intelligibility when the SNR is poor. Although the experiment was not designed to directly measure cognitive demand, it is possible that increased cognitive resources were required to perform the dual-task of localizing the target while also identifying and recalling the color and number stated by the talker. Here the target signal is both reduced in spectral detail while also masked by competing talkers, thereby requiring an increase in cognitive resources [69]. Collectively, this may also elucidate the findings observed in listeners on the localization and speech intelligibility task.
Changes in reaction times may represent disruption in situational awareness and have been suggested to reflect an increase in listening effort required for a given task. Bolia and McKinley showed increased reaction times for listeners under HPD conditions [36]. Smalt and colleagues showed that the use of HPDs increased listening effort, and suggested that HPD use led to cognitive fatigue in noisy environments [42]. In our cohort, promptness was highly variable across listeners, but as a group did not change significantly with HPD use.
Listeners performed a series of spatial and verbal tasks of increasing difficulty in a complex acoustic scene. Results demonstrate that HPDs impact spatial perception such that low SNRs were more susceptible to disruption of target identification and speech intelligibility relative to higher SNRs, with negative impacts to both target location accuracy and speech intelligibility. However, when the target signal was more salient and positioned in the front hemifield, listeners were better able to locate the talker and identify the talker content. There was no additive effect when the signal was low-pass filtered to simulate NIHL, suggesting that the negative effects HPDs on spatial perception may exceed the loss of high frequency cues associated with NIHL for monitoring of speech streams.
Because the sample used in this study consisted of normal hearing participants and used low-pass filtering to simulate NIHL, the findings from this study may not directly translate to ERWs. In addition, the design is not able to fully represent real world scenarios that ERWs may encounter. Future studies should include specific metrics to further explore the potential effects of NIHL and HPD use on cognitive resources.
V. Conclusions
Use of HPDs can disrupt spatial perception in different ways, with the effect varying with the type of HPD. OTE HPDs significantly impacted localization performance, while ITE HPDs significantly impacted monitoring of speech signals. These results are consistent with the self-reported barrier to HPD use of reduced situational awareness and decreased task performance, evidenced by decreased ability to make accurate decisions about the location source in a controlled, dual-task localization experiment. Occlusion of the ear results in perceptible acoustic disruption of sound, and reduced access to spatial cues. The results further suggest that behavioral strategies are ineffective for improving performance in increasingly adverse listening scenarios. The disruption of spatial perception cues associated with HPD use exceeds that observed under simulated NIHL listening conditions through low-pass filtering.
References
- 1. Nelson DI, Nelson RY, Concha-Barrientos M, Fingerhut M. The global burden of occupational noise-induced hearing loss. Am J Ind Med. 2005;48(6):446–58. pmid:16299704
- 2.
Centers for Disease Control and Prevention C. National Occupational Research Agenda. National Institute for Occupational Safety and Health: DHHS (NIOSH) Publication Number 96–115; 2014.
- 3. Broyles G, Kardous CA, Shaw PB, Krieg EF. Noise exposures and perceptions of hearing conservation programs among wildland firefighters. J Occup Environ Hyg. 2019;16(12):775–84. pmid:31658434
- 4. Hong O, Chin DL, Ronis DL. Predictors of hearing protection behavior among firefighters in the United States. Int J Behav Med. 2013;20(1):121–30. pmid:22161219
- 5. Hong O, Samo D, Hulea R, Eakin B. Perception and attitudes of firefighters on noise exposure and hearing loss. J Occup Environ Hyg. 2008;5(3):210–5. pmid:18213534
- 6. Talcott KA, Casali JG, Keady JP, Killion MC. Azimuthal auditory localization of gunshots in a realistic field environment: effects of open-ear versus hearing protection-enhancement devices (HPEDs), military vehicle noise, and hearing impairment. International journal of audiology. 2012;51 Suppl 1:S20–30.
- 7. Patil M, Breeze JJBMH. Use of hearing protection on military operations. 2011;157(4):381–4. pmid:22319982
- 8. Clasing JE, Casali JG. Warfighter auditory situation awareness: effects of augmented hearing protection/enhancement devices and TCAPS for military ground combat applications. International journal of audiology. 2014;53 Suppl 2:S43–52. pmid:24564692
- 9.
Millet B, Solle N, Snapp H, Rajguru S, Khan U. Firefighters’ Perceptions of Occupational Noise Exposure and Associated Health Outcomes. 12th International Conference of Applied Human Factors and Ergonomics; July 25–29, 2021; New York, NY2021.
- 10. Millet B, Harris K, Rajguru S, Snapp H. Noise exposure and hearing health among US Firefighters. Safety and Health at Work. 2022;13:S281.
- 11. Vause NL, Grantham DW. Effects of earplugs and protective headgear on auditory localization ability in the horizontal plane. Human factors. 1999;41(2):282–94. pmid:10422534
- 12. Brungart DS, Kordik AJ, Simpson BD, McKinley RL. Auditory localization in the horizontal plane with single and double hearing protection. Aviat Space Environ Med. 2003;74(9):937–46. pmid:14503671
- 13. Simpson BD, Bolia RS, McKinley RL, Brungart DS. The impact of hearing protection on sound localization and orienting behavior. Human factors. 2005;47(1):188–98. pmid:15960096
- 14. Abel SM, Armstrong NM. Sound localization with hearing protectors. J Otolaryngol. 1993;22(5):357–63. pmid:8283505
- 15. Fostick L, Fink N. Situational Awareness: The Effect of Stimulus Type and Hearing Protection on Sound Localization. Sensors (Basel). 2021;21(21). pmid:34770351
- 16. Brimijoin WO, Akeroyd MA. The Effects of Hearing Impairment, Age, and Hearing Aids on the Use of Self-Motion for Determining Front/Back Location. Journal of the American Academy of Audiology. 2016;27(7):588–600. pmid:27406664
- 17. Smith-Olinde L, Koehnke J, Besing J. Effects of sensorineural hearing loss on interaural discrimination and virtual localization. The Journal of the Acoustical Society of America. 1998;103(4):2084–99. pmid:9566330
- 18. Noble W, Byrne D, Lepage B. Effects on sound localization of configuration and type of hearing impairment. The Journal of the Acoustical Society of America. 1994;95(2):992–1005. pmid:8132913
- 19. Lorenzi C, Gatehouse S, Lever C. Sound localization in noise in hearing-impaired listeners. The Journal of the Acoustical Society of America. 1999;105(6):3454–63. pmid:10380669
- 20. Deshaies P, Martin R, Belzile D, Fortier P, Laroche C, Leroux T, et al. Noise as an explanatory factor in work-related fatality reports. Noise Health. 2015;17(78):294–9. pmid:26356371
- 21. Girard SA, Leroux T, Courteau M, Picard M, Turcotte F, Richer O. Occupational noise exposure and noise-induced hearing loss are associated with work-related injuries leading to admission to hospital. Inj Prev. 2015;21(e1):e88–92. pmid:24639292
- 22. Morata TC, Themann CL, Randolph RF, Verbsky BL, Byrne DC, Reeves ER. Working in noise with a hearing loss: perceptions from workers, supervisors, and hearing conservation program managers. Ear and hearing. 2005;26(6):529–45. pmid:16377991
- 23. Teizer J, Allread BS, Fullerton CE, Hinze J. Autonomous pro-active real-time construction worker and equipment operator proximity safety alert system. J Automation in construction. 2010;19(5):630–40.
- 24. Gaydos SJ, Harrigan MJ, Bushby AJ. Ten years of spatial disorientation in US Army rotary-wing operations. Aviation Space Environ Med. 2012;83(8):739–45.
- 25.
Houser A, Jackson BA, Bartis JT, Peterson D. Emergency responder injuries and fatalities: An analysis of surveillance data. Santa Monica, CA: National Personal ProtectiveTechnology Laboratory; 2004.
- 26. Arslan M, Cruz C, Ginhac D. Semantic enrichment of spatio-temporal trajectories for worker safety on construction sites. Pers Ubiquitous Comput. 2019;23(5):749–64.
- 27. Choi M, Ahn S, Seo J. VR-Based investigation of forklift operator situation awareness for preventing collision accidents. Accid Anal Prev. 2020;136:105404. pmid:31927452
- 28. Teizer J, Cheng TJAiC. Proximity hazard indicator for workers-on-foot near miss interactions with construction equipment and geo-referenced hazard areas. 2015;60:58–73.
- 29.
Arslan M, Cruz C, Ginhac D, editors. Exploiting Semantic Trajectories using HMMs and BIM for Worker Safety in Dynamic Environments. 2018 International Conference on Computational Science and Computational Intelligence (CSCI); 2018: IEEE.
- 30. Abel SM. Barriers to hearing conservation programs in combat arms occupations. Aviat Space Environ Med. 2008;79(6):591–8. pmid:18581943
- 31. King RB, Oldfield SR. The impact of signal bandwidth on auditory localization: Implications for the design of three-dimensional audio displays. Human factors. 1997;39(2):287–95.
- 32. Wightman FL, Kistler DJ. The dominant role of low-frequency interaural time differences in sound localization. The Journal of the Acoustical Society of America. 1992;91(3):1648–61. pmid:1564201
- 33. Vliegen J, Van Opstal AJ. The influence of duration and level on human sound localization. The Journal of the Acoustical Society of America. 2004;115(4):1705–13. pmid:15101649
- 34. Lorenzi C, Gatehouse S, Lever C. Sound localization in noise in normal-hearing listeners. The Journal of the Acoustical Society of America. 1999;105(3):1810–20. pmid:10089604
- 35. Abouchacra KS, Emanuel DC, Blood IM, Letowski TR. Spatial perception of speech in various signal to noise ratios. Ear and hearing. 1998;19(4):298–309. pmid:9728725
- 36. Bolia RS, McKinley RL. The effects of hearing protectors on auditory localization: evidence from audio-visual target acquisition. Int J Occup Saf Ergon. 2000;6(3):309–19. pmid:11082547
- 37. Bolia RS, D’Angelo WR, Mishler PJ, Morris LJ. Effects of hearing protectors on auditory localization in azimuth and elevation. Human factors. 2001;43(1):122–8. pmid:11474758
- 38. Killion MC, Monroe T, Drambarean V. Better protection from blasts without sacrificing situational awareness. International journal of audiology. 2011;50 Suppl 1:S38–45. pmid:21288067
- 39. Zimpfer V, Sarafian D. Impact of hearing protection devices on sound localization performance. Frontiers in neuroscience. 2014;8:135. pmid:24966807
- 40. Abel SM, Boyne S, Roesler-Mulroney H. Sound localization with an army helmet worn in combination with an in-ear advanced communications system. Noise Health. 2009;11(45):199–205. pmid:19805929
- 41. Borg E, Bergkvist C, Bagger-Sjoback D. Effect on directional hearing in hunters using amplifying (level dependent) hearing protectors. Otology & neurotology: official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology. 2008;29(5):579–85. pmid:18520633
- 42. Smalt CJ, Calamia PT, Dumas AP, Perricone JP, Patel T, Bobrow J, et al. The Effect of Hearing-Protection Devices on Auditory Situational Awareness and Listening Effort. Ear and hearing. 2020;41(1):82–94. pmid:31045653
- 43. Brown AD, Beemer BT, Greene NT, Argo Tt, Meegan GD, Tollin DJ. Effects of Active and Passive Hearing Protection Devices on Sound Source Localization, Speech Recognition, and Tone Detection. PLoS One. 2015;10(8):e0136568. pmid:26313145
- 44. Koskan PD, Millet B. Method and system for providing alerts for radio communications. Google Patents; 2018.
- 45. Begault DR, Wenzel EM, Anderson MR. Direct comparison of the impact of head tracking, reverberation, and individualized head-related transfer functions on the spatial perception of a virtual speech source. J Audio Eng Soc. 2001;49(10):904–16. pmid:11885605
- 46. Iwaya Y, Suzuki Y, Kimura D, technology. Effects of head movement on front-back error in sound localization. Acoust Sci Technol. 2003;24(5):322–4.
- 47. Good MD, Gilkey RH. Sound localization in noise: the effect of signal-to-noise ratio. The Journal of the Acoustical Society of America. 1996;99(2):1108–17. pmid:8609294
- 48. Shinn-Cunningham BG. Object-based auditory and visual attention. Trends in cognitive sciences. 2008;12(5):182–6. pmid:18396091
- 49. Grange JA, Culling JF. The benefit of head orientation to speech intelligibility in noise. The Journal of the Acoustical Society of America. 2016;139(2):703–12. pmid:26936554
- 50. Goossens HH, van Opstal AJ. Influence of head position on the spatial representation of acoustic targets. J Neurophysiol. 1999;81(6):2720–36. pmid:10368392
- 51. Middlebrooks JC, Green DM. Sound localization by human listeners. Annual review of psychology. 1991;42:135–59. pmid:2018391
- 52. Brungart DS, Simpson BD. Effect of target-masker similarity on across-ear interference in a dichotic cocktail-party listening task. The Journal of the Acoustical Society of America. 2007;122(3):1724. pmid:17927432
- 53. Munro KJ, Agnew N. A comparison of inter-aural attenuation with the Etymotic ER-3A insert earphone and the Telephonics TDH-39 supra-aural earphone. British journal of audiology. 1999;33(4):259–62. pmid:10509860
- 54. Studebaker GA. A "rationalized" arcsine transform. Journal of speech and hearing research. 1985;28(3):455–62. pmid:4046587
- 55. Carpenter RH, Williams ML. Neural computation of log likelihood in control of saccadic eye movements. Nature. 1995;377(6544):59–62. pmid:7659161
- 56. Makous JC, Middlebrooks JC. Two-dimensional sound localization by human listeners. The Journal of the Acoustical Society of America. 1990;87(5):2188–200. pmid:2348023
- 57. Oldfield SR, Parker SP. Acuity of sound localisation: a topography of auditory space. I. Normal hearing conditions. Perception. 1984;13(5):581–600. pmid:6535983
- 58. Dickman SJ, Meyer DE. Impulsivity and speed-accuracy tradeoffs in information processing. Journal of personality and social psychology. 1988;54(2):274. pmid:3346814
- 59. Drugowitsch J, Moreno-Bote R, Churchland AK, Shadlen MN, Pouget A. The cost of accumulating evidence in perceptual decision making. The Journal of neuroscience: the official journal of the Society for Neuroscience. 2012;32(11):3612–28.
- 60. Musicant AD, Butler RA. The psychophysical basis of monaural localization. Hearing research. 1984;14(2):185–90. pmid:6746432
- 61. Wightman FL, Kistler DJ. Resolution of front–back ambiguity in spatial hearing by listener and source movement. J Acoust Soc Am. 1999;105(5):2841–53. pmid:10335634
- 62. Perrett S, Noble W. The contribution of head motion cues to localization of low-pass noise. Percept Psychophys. 1997;59(7):1018–26. pmid:9360475
- 63. Brimijoin WO, McShefferty D, Akeroyd MA. Auditory and visual orienting responses in listeners with and without hearing-impairment. The Journal of the Acoustical Society of America. 2010;127(6):3678–88. pmid:20550266
- 64. Kato M, Uematsu H, Kashino M, Hirahara T. The effect of head motion on the accuracy of sound localization. Acoust Sci Technol. 2003;24(5):315–7.
- 65. Perrett S, Noble W. The effect of head rotations on vertical plane sound localization. The Journal of the Acoustical Society of America. 1997;102(4):2325–32. pmid:9348691
- 66. Abel SM, Hay VH. Sound localization. The interaction of aging, hearing loss and hearing protection. Scand Audiol. 1996;25(1):3–12. pmid:8658023
- 67. Hebrank J, Wright D. Spectral cues used in the localization of sound sources on the median plane. The Journal of the Acoustical Society of America. 1974;56(6):1829–34. pmid:4443482
- 68. Snapp HA, Ausili SA. Hearing with One Ear: Consequences and Treatments for Profound Unilateral Hearing Loss. J Clin Med. 2020;9(4). pmid:32260087
- 69. Pichora-Fuller MK, Kramer SE, Eckert MA, Edwards B, Hornsby BW, Humes LE, et al. Hearing Impairment and Cognitive Energy: The Framework for Understanding Effortful Listening (FUEL). Ear Hear. 2016;37 Suppl 1:5S–27S. pmid:27355771