Elsevier

Hearing Research

Volume 378, July 2019, Pages 53-62
Hearing Research

Research Paper
Human middle-ear muscles rarely contract in anticipation of acoustic impulses: Implications for hearing risk assessments

https://doi.org/10.1016/j.heares.2018.11.006Get rights and content

Highlights

  • The acoustic reflex and associated MEMC are not conditionable in most people.

  • The MEMC engages quicker for impulse noise than for brief pure tone signals.

  • Implementing an anticipatory MEMC as protective is not supported here.

  • Anticipatory MEMCs are inappropriate to use when calculating auditory injury risk.

Abstract

The current study addressed the existence of an anticipatory middle-ear muscle contraction (MEMC) as a protective mechanism found in recent damage-risk criteria for impulse noise exposure. Specifically, the experiments reported here tested instances when an exposed individual was aware of and could anticipate the arrival of an acoustic impulse. In order to detect MEMCs in human subjects, a laser-Doppler vibrometer (LDV) was used to measure tympanic membrane (TM) motion in response to a probe tone. Here we directly measured the time course and relative magnitude changes of TM velocity in response to an acoustic reflex-eliciting (i.e. MEMC eliciting) impulse in 59 subjects with clinically assessable MEMCs. After verifying the presence of the MEMC, we used a classical conditioning paradigm pairing reflex-eliciting acoustic impulses (unconditioned stimulus, UCS) with various preceding stimuli (conditioned stimulus, CS). Changes in the time-course of the MEMC following conditioning were considered evidence of MEMC conditioning, and any indication of an MEMC prior to the onset of the acoustic elicitor was considered an anticipatory response. Nine subjects did not produce a MEMC measurable via LDV. For those subjects with an observable MEMC (n = 50), 48 subjects (96%) did not show evidence of an anticipatory response after conditioning, whereas only 2 subjects (4%) did. These findings reveal that MEMCs are not readily conditioned in most individuals, suggesting that anticipatory MEMCs are not prevalent within the general population. The prevalence of anticipatory MEMCs does not appear to be sufficient to justify inclusion as a protective mechanism in auditory injury risk assessments.

Introduction

Repetitive exposure to high-level acoustic impulses, such as those from firearms and blast overpressure, increases the risk of hearing loss (Agrawal et al., 2009). Unfortunately, the ability to predict the hearing loss resulting from such exposures is not as straightforward as it is for steady-state noise, due to the brief duration of acoustic impulses (Miller, 1974). The U.S. Department of Defense (DoD) design criteria standard of noise limits for a materiel acquisition, MIL-STD-1474E, includes the choice of two damage-risk criteria (DRC). Either the Auditory Hazard Assessment Algorithm for Humans (AHAAH) software (Price, 2007a, 2007b; Price, 1991; Price and Kalb, 1991) or the LIAeq100 ms variant of the 8 h equivalent noise exposure (LAeq8 model) can be used for calculating the maximum permissible noise levels produced by military systems (DoD, 2015). Each has their own limitation and a medical standard concerning impulsive noise for health hazard assessments (HHA) is still needed.

Significant concerns with the AHAAH have been raised, many of which were summarized in a peer review by the American Institute of Biological Sciences (AIBS, Wightman, et al., 2010). The AHAAH is an electrical equivalence model of the sound transmission pathway to the human inner ear designed to predict the auditory injury caused by the intense pressure changes at the ear resulting from impulsive noise exposure (Price, 1991). Recent studies have focused on investigating specific components of the sound transmission pathway modeled by AHAAH, including responses of the tympanic membrane (Gan et al., 2016; Liang et al., 2016), ossicular chain motion and intracochlear pressures (Greene et al., 2017, 2018), while others have focused on updating model parameters and injury estimates (Zagadou et al., 2016), validating model results against human exposure data (Chan et al., 2001; Patterson and Ahroon, 2004), or investigating alternate approaches to auditory injury prediction (De Paolis et al., 2017; Zagadou et al., 2016). These studies have raised issues about the validation efforts of the AHAAH and concerns about the appropriateness of implementing it as a medical standard in any future updated DRC.

One controversial component in the AHAAH model, which the U.S. Army is required to use, has been the inclusion of the acoustic reflex as a protective mechanism against high-level acoustic impulses (Coles et al., 1967; Kalb and Price, 1987; Murphy et al., 2009; Patterson and Ahroon, 2004; Price, 1991, 2001, 2005). In particular, an assumption that the acoustic reflex can be elicited in anticipation of an acoustic impulse, and thereby provide protection against impulses occurring faster than the acoustic reflex, is implemented, and has been a highly-debated topic of discussion recently (Patterson and Ahroon., 2004; Price & Kalb 2018; Price, 2010; Price et al., 2017; Wightman et al., 2010; Zagadou et al., 2016, 2017). The acoustic reflex refers to a middle-ear muscle contraction (MEMC), which engages bilaterally (Dallos, 1964) and generally involves only the stapedius muscle in humans, in response to high-level acoustic sounds (Feeney and Keefe, 1999; Silman, 2012). Notably, this is in direct contrast with the acoustic reflex in many common laboratory species, where the acoustic reflex may activate both stapedius and tensor tympani MEMCs (e.g. (Forbes and Sherrigton, 1914; Mukerji et al., 2010)), thus results from animal studies may not be directly applicable to exposure estimates.

Contraction of the stapedius muscle stiffens the ossicular chain at the point of the stapes and pulls on the head of the stapes in a direction parallel to the plane of the stapes footplate. The result of a stapedius muscle contraction is a rotation to the footplate within the oval window thereby tensing the annular ligament, which effectively increases the acoustic impedance of sound transmission through the middle ear for frequencies below 1 kHz (Borg, 1968; Feeney and Keefe, 1999; Pang and Peake, 1986; Silman, 2012; Von Békésy and Wever, 1960). The acoustic reflex activates with a sufficiently long latency (Dallos, 1964) that the resulting MEMC would not affect sound transmission for an impulsive noise exposure. Nevertheless, the AHAAH implements an option that includes an anticipatory MEMC as a protective mechanism, if the model user believes the listener can anticipate the sound arrival, presumably by means of classical conditioning or for situations in which there is repetitive and consecutive impulses (DoD, 2015; Patterson and Ahroon, 2004; Price & Kalb 2018; Price, 1991; Price, 2001).

Two main concerns related to the implementation of the MEMC in the AHAAH were raised by the AIBS peer review. First, that evidence in the scientific literature for an MEMC in anticipation of an acoustic impulse exposure is equivocal. To date, there are a limited number of reports demonstrating that an MEMC can be elicited via classical conditioning: in technical reports published by the AHAAH developers (Price & Kalb 2018; Price, 2005), studies are cited as demonstrable evidence that the MEMC can be classically conditioned (Bates et al., 1970; Brainerd and Beasely, 1971; Brasher et al., 1969; Djupesland, 1965; Yonovitz, 1976); however, conflicting findings within these studies suggest that it remains unclear whether MEMCs can be elicited prior to a sound exposure (i.e., a “warned” response). Second, is that the MEMC associated with the acoustic reflex is assumed to be present across the general population, which appears to be inappropriate, as acoustic reflexes are observed in no more than ∼89% of healthy, normal hearing individuals, and at much lower rates in older and hearing impaired populations (Flamme et al., 2017; McGregor et al., 2018). Compounding the low prevalence rates of observable MEMCs reported in these studies, the question of how prevalent an anticipatory MEMC response is remains.

The current study aimed to determine the occurrence of conditionable MEMCs in a group of normal hearing subjects. Specifically, we used laser-Doppler vibrometry to directly measure the time course of acoustic reflex engagement elicited by acoustic impulses, and tested the hypothesis that the acoustic reflex can be classically conditioned to respond in anticipation (i.e., a “warned” response) of sound exposure. Here we attempted to elicit an anticipatory MEMC response by pairing an eliciting acoustical stimulus with a preceding “conditioning” stimulus prior to sound exposure. Three different modalities of preceding stimuli using a light flash, a verbal countdown and or a subject-determined button press were used. This approach allowed us to investigate whether a particular sensory modality or active engagement by the subject could cause an anticipatory MEMC response. As such, these experiments provided a three-pronged approach toward assessing the “warned” middle-ear assumption of the AHAAH. In addition, the use of laser-Doppler vibrometry allows for a more precise measurement of the temporal onset and time constants associated with the engagement of the MEMC over previous techniques using wide-band reflectance. These data can be used to determine appropriate inclusion criteria for MEMCs in future models and DRCs. It is imperative these assumptions be experimentally validated before accurate DRC can be established. For the purpose of a health hazard assessment, an inappropriate implementation of the anticipatory MEMC assumption would result in an underestimation of the risk of acoustic exposure to the auditory system, and would incorrectly predict that some high-level exposures are safe.

Section snippets

Subjects

Sixty-six volunteers (53 males and 13 females, aged 19–53) were consented and underwent audiometric screening to determine if they met the inclusion criteria for the study. First, subjects underwent an otoscopic examination to ensure the ear canals were unoccluded and that the tympanic membrane (TM) could be visualized, and were dismissed otherwise. Pure-tone air-conduction hearing thresholds were measured and no volunteer was included if they had greater than 25 dB HL at any of the octave-band

LDV measurements of MEMCs

Out of the 59 subjects tested, MEMCs were not visible using the LDV described techniques in 9 subject (15%), despite all subjects having detectable acoustic reflexes using a middle-ear analyzer commonly used for clinical assessments prior to testing. This variability may be due either to lack of a measurable MEMC, or due to the LDV technique being unable to detect the MEMC, but either way demonstrates that MEMCs are variable across individuals and highlights the difficulty in measuring the time

Discussion

The current study directly tested whether the MEMC associated with the acoustic reflex could be conditioned to activate prior to the arrival of an acoustic impulse. Using a three-pronged approach, we paired either a visual stimulus (i.e., a light flash), an auditory stimulus (i.e., a verbal countdown) or an active engagement by the subjects (i.e., a button press) with a subsequent acoustic reflex eliciting impulse. Following a conditioning paradigm of several repetitive presentations of the

Conclusions

This study found that although anticipatory MEMC were observable in some individuals, they were certainly not prevalent among the population of the subject tested. This means that the “warned” option is not appropriate to use when calculating auditory injury risk. In addition, the acoustic reflexes reported here did not exhibit the rapid onset and activation slope implemented in the AHAAH for attenuating the later arriving secondary features associated with an impulse noise. The MEMC also

Acknowledgments

We would like to acknowledge the assistance of Mr. Michael Chen and Dr. Bankole Fasanya during the initial development of the current study. We are especially grateful to Ms. Lana Milam who supported data collection and to Dr. Stephanie Karch for her feedback on study protocols. We would like to thank the U.S. Army Medical Research and Material Command (USAMRMC). This work was funded under Task Area A1 for Sensory Injury Prevention and Reduction by the Military Operational Medicine Research

References (56)

  • M.A. Bates et al.

    Attempts to condition the acoustic reflex

    J. Aud. Res.

    (1970)
  • J.A. Beyea et al.

    Laser Doppler vibrometry measurements of human cadaveric tympanic membrane vibration

    Journal of Otolaryngology-Head & Neck Surgery

    (2013)
  • E. Borg

    A quantitative study of the effect of the acoustic stapedius reflex on sound transmission through the middle ear of man

    Acta Otolaryngol.

    (1968)
  • S.H. Brainerd et al.

    Respondent conditioning of the middle ear reflex

    J. Aud. Res.

    (1971)
  • P.F. Brasher et al.

    Middle-ear muscle activity and temporary threshold shift

    Int Audiol

    (1969)
  • P.C. Chan et al.

    Evaluation of impulse noise criteria using human volunteer data

    J. Acoust. Soc. Am.

    (2001)
  • R.R.A. Coles et al.

    Criteria for Assessing Hearing Damage Risk from Impulse-noise Exposure. Technical Memorandum 13-67

    (1967)
  • P.J. Dallos

    Dynamics of the acoustic reflex: phenomenological aspects

    J. Acoust. Soc. Am.

    (1964)
  • R.R. Davis et al.

    Scientific Peer Advisory and Review Services Peer Review on the Human Research and Engineering Directorate (HRED) Method for Assessing the Risk of Auditory

    (2001)
  • G. Djupesland

    Middle ear muscle reflexes elicited by acoustic and nonacoustic stimulation

    Acta Otolaryngol.

    (1964)
  • G. Djupesland

    Electromyography of the tympanic muscles in man

    Int. J. Audiol.

    (1965)
  • DoD

    MIL-STD-1474d Department of Defense Design Criteria Standard: Noise Limits

    (1997)
  • DoD

    MIL-STD-1474E Department of Defense Design Criteria Standard: Noise Limits

    (2015)
  • M.P. Feeney et al.

    Acoustic reflex detection using wide-band acoustic reflectance, admittance, and power measurements

    J. Speech Lang. Hear. Res.

    (1999)
  • M.P. Feeney et al.

    Estimating the acoustic reflex threshold from wideband measures of reflectance, admittance, and power

    Ear Hear.

    (2001)
  • G.A. Flamme et al.

    Acoustic reflexes are common but not pervasive: evidence from the National Health and Nutrition Examination Survey, 1999–2012

    Int. J. Audiol.

    (2017)
  • A. Forbes et al.

    Acoustic reflexes in the decerebrate cat

    American Journal of Physiology--Legacy Content

    (1914)
  • N.T. Greene et al.

    Intracochlear Pressure Measurements during Acoustic Shock Wave Exposure

    (2018)
  • Cited by (0)

    1

    Currently at:Department of Otolaryngology, University of Colorado Anschutz Medical Campus, 12631E 17th Ave. MS B205, Aurora, CO 80045.

    View full text