Physiologically the middle ear, containing the three ossicles, serves primarily to

2.26.3 The Middle Ear Muscles

The middle ear muscles (MEM) alter the mechanical properties of the middle ear and thus modulate the way sound vibrations are transmitted to the cochlea. Two muscles are involved in this reflex: the stapedius, which attaches to the neck of the stapes, and the tensor tympani, which attaches to the neck of the malleus. When activated, these muscles attenuate sound levels in the middle ear by dampening vibration of the ossicular chain. Specifically, the stapedius stiffens the attachment of the stapes to the oval window of the cochlea and the tensor tympani pulls on the malleus medially, increasing the tension of the tympanic membrane (reviewed by Mukerji et al., 2010). In most mammals, high intensity, low frequency sound elicits contraction of both muscles; however, in humans and monkeys, relevant acoustic stimuli elicit a response mainly in the stapedius (Klockhoff and Anderson, 1960; Stach et al., 1984; Horner, 1986; Guinan and McCue, 1987; van den Berge et al., 1990; Giraud et al., 1995; Clement et al., 2002; reviewed by Mukerji et al., 2010). Tensor tympani contraction in humans can also be elicited by a variety of stimuli, such as swallowing, stimulation of the tongue, orbital air puffs, facial stroking, and self-vocalization (Kamerer and Rood, 1978; Salén and Zakrisson, 1978; Bance et al., 2013).

2.26.3.1 The Middle Ear Muscle Reflex and Its Associated Motoneurons

Each middle ear muscle is innervated by a specific pool of motoneurons in the ipsilateral brainstem: tensor tympani motoneurons are associated with the trigeminal motor nucleus and stapedius motoneurons are associated with the facial nucleus. Each of these motoneuron pools receives input from auditory nuclei, establishing reflex circuits. While these circuits are separate, they are often considered together in the context of a middle ear muscle reflex (Fig. 1). The ascending limb for this reflex is the same for both muscles: an auditory stimulus triggers an action potential that travels from the spiral ganglion to the ventral cochlear nucleus (VCN). Neurons in the VCN then project directly or indirectly to the motoneurons controlling either the ipsilateral stapedius or the ipsilateral tensor tympani (Borg, 1973; Itoh et al., 1986; Ito and Honjo, 1988; Lee et al., 2006; Billig et al., 2007). Sound presentation in one ear results in contraction of both stapedius muscles, so there must also be a projection, directly or indirectly, to the contralateral stapedius motoneuron pool (McCue and Guinan, 1988; Vacher et al., 1989).

Figure 1. Reflex circuits associated with the middle ear muscles. (A) The tensor tympani reflex begins with auditory information entering the brain via the auditory nerve (brown arrow from cochlea). The auditory nerve synapses in the VCN, which then provides input (brown arrow) to the tensor tympani motoneuron pool (purple circles), located ventro-lateral to the trigeminal motor nucleus (Motor V). Motoneurons project to the ipsilateral tensor tympani muscle (attached to the neck of the malleus) via the mandibular branch of cranial nerve V (purple arrow). (B) The stapedius reflex begins with auditory information entering the brain via the auditory nerve (brown arrow from cochlea). The auditory nerve synapses in the VCN, which then provides input (brown arrow) to stapedius motoneurons (blue circles) which surround the facial nucleus. Motoneurons project to the ipsilateral stapedius muscle (attached to the neck of the stapes) via cranial nerve VII (blue arrow).

Stapedius motoneurons surround the rostral portion of the facial nucleus (although there is species variation in their exact location) and their axons leave the brain as part of cranial nerve VII (Lyon, 1978; Shaw and Baker, 1983; Joseph et al., 1985 (cat); Thompson et al., 1985 (squirrel monkey and bush baby); Strutz et al., 1988 (guinea pig); Wong et al., 1992 (chicken)). One report describes these motoneurons as morphologically similar to periolivary cells of the superior olivary complex (SOC; situated medial to the motoneuron group), and dissimilar to adjacent facial nucleus cells (Thompson et al., 1985). Single unit recordings have shown that stapedius motoneurons are spatially segregated in the brainstem based on their response properties (whether they respond to sound in the ipsilateral, contralateral, or both ears), although there is no such organization of fibers contacting the muscle (McCue and Guinan, 1988; Vacher et al., 1989; Wiener-Vacher et al., 1999). Stapedius motoneurons are cholinergic, but can also express calcium gene-related peptide (Joseph et al., 1985; Reuss et al., 2008).

Tensor tympani motoneurons are located laterally and ventro-laterally to the rostral portion of the trigeminal motor nucleus, and their axons leave the brain as part of the mandibular branch of cranial nerve V. Some reports describe these motoneurons as a subgroup within the trigeminal motor nucleus, however most recent reports differentiate them as a separate nucleus; there is species variation on this point (Lyon, 1975; Keller et al., 1983; Shaw and Baker, 1983; Friauf and Baker, 1985 (cat); Mizuno et al., 1982 (guinea pig and cat); Spangler et al., 1982 (rat and guinea pig); Takahashi et al., 1984; Counter et al., 1993 (rabbit); Gannon and Eden, 1987 (cynomolgus monkey); Strutz et al., 1988 (guinea pig); Mukerji et al., 2009 (mouse)). Tensor tympani motoneurons are often smaller than adjacent masticatory motoneurons (Mizuno et al., 1982; Keller et al., 1983; Gannon and Eden, 1987; Strutz et al., 1988; Counter et al., 1993), and can be classified into three groups based on morphology (Mukerji et al., 2009). Their dendrites branch heavily, and Friauf and Baker (1985) suggest that this large membrane area supports the hypothesis that tensor tympani motoneurons receive a diverse array of inputs. Tensor tympani motoneurons are cholinergic and co-express endorphins, with subgroups also co-expressing bombesin, cholecystokinin, enkephalin, and neuronal nitric oxide synthase (Reuss et al., 2009).

2.26.3.2 Synaptic Inputs and Ultrastructure of Middle Ear Muscle Motoneurons

It is widely accepted that stapedius and tensor tympani motoneurons receive inputs other than those from the VCN. This is supported by electron microscopic studies of motoneurons, which show a variety of input types. Most synaptic inputs onto both stapedius and tensor tympani motoneurons can be classified in one of three ways: (1) synapses with large round vesicles and asymmetric synaptic densities (considered excitatory), (2) synapses with small round vesicles and asymmetric synaptic densities (also considered excitatory), and (3) synapses with pleomorphic (round or oval-shaped) vesicles and symmetric synaptic densities (considered inhibitory) (Fig. 2, Fig. 3A and B). Some terminals on both stapedius and tensor tympani motoneurons contain large dense-core vesicles (Fig. 2B, Fig. 3B), indicating the presence of neuropeptide transmission (Lee et al., 2008; Benson et al., 2013). These large dense-core vesicle-containing terminals are more common on tensor tympani motoneurons (Mukerji et al., 2010). For both stapedius and tensor tympani motoneurons, the majority of synapses show excitatory morphology (Lee et al., 2008; Benson et al., 2013). There are two additional, relatively uncommon, types of inputs which contact both groups of motoneurons: (1) synapses with round vesicles of heterogeneous sizes and asymmetric synaptic densities, and (2) synapses with large round vesicles and a subsurface cistern in the motoneuron (Lee et al., 2008; Benson et al., 2013) (Fig. 3A, C). These synapses would also be expected to have an excitatory effect on the motoneurons.

Figure 2. Electron micrographs show examples of the three most common types of synapses on tensor tympani motoneurons (TTMNs). Synaptic junctions are shown between arrowheads. The major distinction is based on the size and shape of the synaptic vesicles, including (A) Large Round (orange), (B) Small Round (green), and (C) Pleomorphic (red). The Small Round terminal in (B) also contains a dense core vesicle (DCV), presumed to contain a neuropeptide. The synapse in (C) contacts a spine (sp). Scale bar = 0.5 μm.

Adapted from Benson et al. (2013), with permission.

Figure 3. Electron micrographs show examples of the five types of synapses on stapedius motoneurons (SMNs). Synaptic junctions are shown between arrowheads. (A) Two common types of synapses are shown: Pleo, with pleomorphic synaptic vesicles (red), and Lg Rnd, with large round synaptic vesicles (orange). A less common type of synapse (Cist, green) is associated with a subsurface cistern in the motoneuron. (B) A common type of synapse is shown which contains small round vesicles (Sm Rnd, yellow). This terminal also contains a dense core vesicle (DCV, black arrow), which indicates the probable presence of a neuropeptide. (C) A less common type of synapse contains round vesicles of heterogeneous sizes (Het Rnd, purple). Scale bar = 0.5 μm.

Adapted from Lee et al. (2008), with permission.

Most synapses on these motoneurons occur on the soma or on dendritic shafts. However, both sets of motoneurons exhibit spines distributed sparsely on the soma and dendrites (Brown et al., 2013a). In most areas of the nervous system, spines are associated with synaptic inputs (often excitatory) on the spine head. In contrast, spines on MEM motoneurons are associated with boutons that engulf the spine (rather than simply abutting the spine head) and form pleomorphic (presumptive inhibitory) synapses (Fig. 4). Moreover, the synaptic junctions are typically at the base of the spine or some distance from the base, but only rarely on the spine shaft or tip, leading to the suggestion that isolation of inputs is not likely a purpose of these spines (Benson et al., 2013).

Figure 4. Electron micrographs show examples of spines (sp) invaginating synaptic terminals (red) on tensor tympani motoneurons (TTMN, top row) and stapedius motoneurons (SMN, bottom row). Sections in a row are consecutive, or a bracket and a number indicate how many sections were skipped between images. Note that spines are completely engulfed by the terminals, and synapses (present between black arrowheads) are present at or near the spine base rather than the spine head. ad - adherens junction. Scale bar = 0.5 μm.

Adapted from Brown et al. (2013a), with permission.

Immunocytochemistry has been used to study the neurotransmitter phenotype of synaptic inputs to MEM motoneurons, although many questions remain unanswered. Boutons containing serotonin, substance P, and neuronal nitric oxide synthase all make close appositions with identified stapedius motoneurons (Thompson et al., 1998; Reuss et al., 2008). Boutons containing serotonin, tyrosine hydroxylase (the rate-limiting synthetic enzyme for catecholamines such as dopamine and norepinephrine), substance P, or neuropeptide-Y all make close appositions with identified tensor tympani motoneurons (summarized in Table 1) (Thompson et al., 1998; Reuss et al., 2009). Observations of neuropeptide signaling are consistent with the reports of dense-core vesicles described above, however which terminal types are associated with serotonergic or dopaminergic terminals is unclear. Additionally, the transmitter phenotype of presumptive inhibitory synaptic inputs is still unknown.

Table 1. Direct inputs to middle ear muscle motoneurons

	Auditory nuclei	Modulators	Peptides		Species	Citation
Ipsi CN	Contra CN	5HT	CAT	SP	NPY	NO
TTMN
	++	+						cat	Itoh et al. (1986)
	++	+						cat	Ito and Honjo (1988)
	+	+						rat	Billig et al. (2007)
			+					bush baby	Thompson et al. (1998)
			+	+	+	+		guinea pig	Reuss et al. (2009)
SMN
			+		+		+	guinea pig	Reuss et al. (2008)

Summary of sources of direct input to middle ear muscle motoneurons. Relative amount of input is indicated by + (some input) or ++ (greater input); lack of symbol indicates no data for such input. Abbreviations: 5HT – serotonin; CAT – catecholamines; CN – cochlear nucleus; Contra – contralateral; Ipsi – ipsilateral; NO – nitric oxide; NPY – neuropeptide Y; SMN – stapedius motoneuron; SP- substance P; TTMN – tensor tympani motoneurons.

The acoustic reflex

The acoustic reflex (AR) (the reflexive contraction of the middle-ear muscles in response to sound stimulation) has a long history of clinical use in defining middle-ear, cochlear, and VIIIth-nerve disorders. The AR and other audiologic tests (e.g., air and bone conduction thresholds, tympanometry, and reflex decay) can help differentiate middle-ear, cochlear, and VIIIth-nerve problems; however, these tests are not able to differentiate auditory nerve from low brainstem involvement. The AR is mediated in the lower pons, and, therefore, reflects only a small portion of the CANS. The AR, however, can provide insight as to central auditory dysfunction in the caudal brainstem.

Accurate measurement of brainstem function using the AR is dependent on normal middle-ear, cochlear, and auditory nerve function. In central (brainstem) dysfunction, the AR threshold usually is elevated or absent. Hall and Johnston's (2007) review of children with CAPD and learning problems showed mixed results across studies, revealing normal, hypersensitive, and in some cases absent or elevated AR. The AR abnormalities occurred in only a relatively small portion of the children in the studies reviewed, suggesting that the central dysfunction underlying their CAPD was rostral to the brainstem. Abnormal AR findings, however, have been reported in a number of different central auditory disorders, such as tumors of the brainstem (Jerger and Jerger, 1977), multiple sclerosis (Jerger et al., 1986), head injury (Hall et al., 1982), and recently in lead poisoning (Counter et al., 2011). In these studies, the sensitivity of the AR was good (approximately 60–80%). Therefore, even though the AR has anatomic limitations in regard to the CANS, it is useful in identifying central disorders that involve the brainstem. The AR, however, has been eclipsed by the auditory brainstem response (ABR) for audiologic diagnosis of brainstem central auditory involvement (see Chapter 29 in this volume).

B Coordinated Activities of Laryngeal (LM) and Middle Ear (MEM) Muscles

In addition to possessing a specialized laryngeal apparatus, a bat also has two highly developed MEM (Henson, 1961, 1965, 1970; Wever and Vernon, 1961). These two groups of muscles (LM and MEM) discharge action potentials not only prior to sound emission but also during acoustic stimulation (Jen and Suga, 1976a, b). Apparently, both LM and MEM are activated coordinately during vocalization and signal reception.

When Myotis lucifugus emitted orientation signals (average duration: 2.9 ± 0.7 msec), the CTM and stapedius muscles (SM) fired impulses of 11.4 ± 2.2 and 8.8 ± 2.2 msec, respectively, prior to vocalization. When it emitted nonorientation sounds (average duration: 31.3 ± 7.1 msec), the CTM and SM discharged impulses of 52.8 ± 4.1 and 38.4 ± 6.7 msec, respectively, before sound emission. Since the other middle ear muscle, tensor tympani (TM), became active 1.9 msec later than the SM during vocalization (Suga and Jen, 1975), the CTM apparently discharged action potentials before both SM and TM.

During acoustic stimulation, the SM and TM discharged impulses with the shortest latency of 3.4 and 4.4 msec, respectively (Fig. 2C and D). This is the acoustic MEM reflex. When recordings were made from the CTM and the inferior laryngeal nerve (ILN), they also fired action potentials (Fig. 2A and B). The shortest latency was 6.2 msec for the CTM and 6.7 msec for the ILN. Since the ILN innervates all extrinsic laryngeal muscles except the CTM, all laryngeal muscles apparently respond to acoustic stimuli. This is called acoustic LM reflex (Jen and Suga, 1976a,b).

All muscle fibers recorded from the two MEM and CTM discharged tonically to acoustic stimuli. The majority (80%) of them had monotonic impulse-count functions, in which the number of impulses increased with the stimulus intensity (Fig. 3A,a-d; B,a-c; C,a,b,d,e). The remaining (20%) fibers showed non-monotonic impulse-count functions, in which the number of impulses increased with the stimulus intensity up to a certain level and then began to decline with a further increase in stimulus intensity (Fig. 3A,e; B,d; C,c).

Triangular threshold curves of 138 CTM fibers were very similar; the fibers were tuned to a small band of frequencies with best frequencies falling within 30–42 kHz (Fig. 4A). Different from the CTM, threshold curves of 67 SM and 46 TM fibers were broadly tuned with best frequencies ranging between 30 and 50 kHz (Fig. 4B and Q.About 62% (41 fibers) of the SM and 39% (18 fibers) of the TM fibers showed a second sensitive peak between 75 and 95 kHz (Fig. 4B,b,e; C,a,b). The Q10-dB (a value used to express the sharpness of a threshold curve and obtained by dividing the best frequency of a threshold curve by the band-width at 10 dB above the minimum threshold; Kiang, 1965) values of CTM threshold curves were mainly between 1.2 and 10, but those of the MEM were between 0.5 and 6.0. The lowest minimum threshold obtained was 40 dB SPL (decibel sound pressure level re 0.0002 dyne/cm2 ) for the CTM fibers and 20 dB SPL for the MEM fibers.

When FM stimuli were used, they usually evoked more vigorous responses from muscle fibers than pure tones did. However, the minimum thresholds of each muscle fiber for these FM stimuli was dependent upon the sweep range and sweep direction of the FM stimuli employed (Jen and Ostwald, 1977; Jen et al., 1978). Among the 66 CTM fibers studied, a 10-kHz range of downward-sweep FM signal on the average was 5 dB more effective than a pure tone, but a 20-kHz range of downward-sweep FM stimulus was only slightly better than a pure tone. The FM stimuli with a 2.5-, 5.0-, or 30-kHz range sweeping either direction were not more effective than pure tones. An upward-sweep FM signal was always less effective than its equivalent downward-sweep signal (Fig. 4D). For example, the threshold for the 10-kHz range of downward-sweep FM stimulus was sometimes as large as 15 dB lower than the threshold for the equivalent upward-sweep FM signal. As for the MEM fibers, more than 95% of the 37 SM and 23 TM fibers studied showed poorer responses to FM stimuli than to pure tones, regardless of the sweep range and direction. The threshold for FM signals was always higher than for pure tones regardless of what types of threshold curves were obtained (Fig. 4E and F).

When a 0.1-msec monophasic electric pulse was applied to the ipsilateral or contralateral SLN, action potentials could be recorded from both MEM and CTM. When the ipsilateral SLN was stimulated, the shortest latency observed in five bats was 4.2 msec in MEM with an average of 5.8 ± 0.97 msec. Stimulation to the contralateral SLN caused a response from the MEM with a latency at least 7.0 msec longer than when the ipsilateral SLN was stimulated. Electrical stimulation of the CTM could also evoke discharges from the MEM, but the latency was always longer than when the SLN was stimulated. This might result from a spread of current to the SLN which causes antidromic firing. When either the SM or TM was electrically stimulated, no action potentials could be recorded from either the ipsilateral or contralateral SLN or CTM. All the data suggest that the neural connection between the CTM and MEM is through a branch of the sensory nerve in the SLN, but sensory nerves of the MEM do not make contact with the CTM.

5.10 Summary

Peripheral hearing loss comes in two broad types, conductive and sensorineural. We describe a diagnostic framework, which includes WAI and MEMR combined with OAEs and ABRs. This may diagnose the type of hearing loss when behavioral audiometry is not possible. Acoustic trauma causes retrograde degeneration of ANFs. It could be secondary when it follows IHC loss or primary as a result of damage of specific ribbon synapses in the IHCs. This primary degeneration can lead to “hidden hearing loss,” characterized by normal audiograms and DPOAEs but great difficulty in understanding speech, especially in background noise. SNHL is accompanied by loudness recruitment, which is likely a central phenomenon. It often coexists with hyperacusis, also a central phenomenon, and so can lead to “over recruitment.” ANP, only identified two decades ago, links to auditory temporal processing disorders and has a strong genetic component. Cochlear implantation very often solves this problem. The differences between ANP and the effects of a vestibular schwannoma are reviewed. Ménière’s disease characterized by intermittent spontaneous attacks of vertigo, fluctuating SNHL, aural fullness or pressure, and roaring tinnitus still remains a mystery. We describe its natural history and the use of ECochG in detail. We conclude with ARHI and describe the epidemiology and its peripheral and cortical substrates.

IX. Summary

The polyvagal theory emphasizes the phylogenetic shifts in the neural regulation of the autonomic nervous system and how this evolutionary shift in neural regulation converged with the regulation of the middle ear muscles to facilitate mammalian vocal communication. The theory emphasizes the different neural circuits that support defensive behaviors (i.e., fight-flight and freeze) and social interactions. According to the theory, during defensive states, when the middle ear muscles are not contracted, acoustic stimuli are prioritized by intensity and during safe social engagement states, acoustic stimuli are prioritized by frequency. During safe states, hearing of the frequencies associated with conspecific vocalizations is selectively being amplified, while other frequencies are attenuated. During the defensive states, the loud low-frequency sounds signaling a predator could be more easily detected and the soft higher frequencies of conspecific vocalizations are lost in background sounds. During social engagement behaviors, an integrated social engagement system regulates a shift in autonomic state to dampen sympathetic activity and to increase parasympathetic tone, while simultaneously increasing the neural tone to the striated muscles of the face and head (i.e., facial expressions, increased “emotional” cueing of the eyes associated with increased eye contact, increased prosody and enhanced listening by contracting the middle ear muscles). During social interactions, the stiffening of the ossicular chain actively changes the transfer function of the middle ear, and functionally dampens low-frequency sounds and improves the ability to extract conspecific vocalizations. However, the selectivity to listen to conspecific vocalizations comes at a cost, and the detection of lower acoustic frequencies generated by predators becomes more difficult. Thus, the identification and construction of safe contexts (e.g., burrows, nests, or houses) plays an important role in enabling the social engagement system to promote prosocial behavior.

11.1 Phantom Sounds

Tinnitus is the conscious perception of sound heard in the absence of physical sound sources external or internal to the body. Sound perceived from physical sound sources inside the body such as blood flow and middle ear muscle twitching is generally called “objective tinnitus”; I will not deal with those here. About 10–15% of adults experience tinnitus. Tinnitus is generally ignited by hearing loss, and very often by NIHL, but most chronic tinnitus is of central origin; that is, it is in the brain and not generated in the ear. A conclusive example is found in patients with one-sided deafness, who often experience tinnitus referred to that ear, yet the tinnitus subsides when that ear is stimulated via a cochlear implant (Chapter 5). The localization of tinnitus to one or both ears is thus likely attributable to a phantom sensation2 and is not unlike that related to sensations or pain experienced after losing a digit or, more severely, a limb. Itch or pain in a no-longer-existing part of the body is truly annoying and so is tinnitus. The pitch of tinnitus corresponds, when there is a hearing loss, to the frequency region of that hearing loss. In case of low-frequency hearing loss the tinnitus is low pitched (“roaring”), but in high-frequency NIHL the tinnitus has a high-pitched ringing or hissing sound. In 1890, MacNaughton Jones3 who studied 260 cases of tinnitus described the sounds of tinnitus as follows:

“The following were the noises I have recorded as complained of by patients. The sound resembling buzzing; sea roaring; trees agitated; singing of kettle; bellows; bee humming; noise of shell; horse out of breath, puffing; thumping noise; continual beating; crackling sounds in the head; train; vibration of a metal; whistle of an engine; steam engine puffing; furnace blowing; constant hammering; rushing water; sea waves; drumming; rain falling; booming; railway whistling; distant thunder; chirping of birds; kettle boiling; waterfall; mill wheel; music; bells.”

As in a true phantom sensation, the brain “hears” the sound of the missing frequencies in one ear, both ears, or inside the head, but describing how it sounds appears to be very personal and typically referred to with known external sounds. Electrophysiological and functional imaging measurements in humans and animals suggest that neural synchrony changes, tonotopic map changes, and increased spontaneous firing rates in the auditory system are potential neural correlates of tinnitus in humans. Tinnitus is likely the result of maladaptive plasticity of the central nervous system. The central nervous system wants to restore its evoked neural activity levels that had been lowered by the hearing loss. This is done by increasing the efficacy (or gain) of its synapses. But this gain also affects the SFR, which occurs in the absence of a physical sound source, and will then generally increase. This is interpreted as sound and called tinnitus. A puzzling aspect is that only 30% of people with hearing loss experience tinnitus, so there must be other purely central nervous system aspects that promote or allow the perception of tinnitus.

Middle Ear Muscles

Two muscles attach to the ossicles of the middle ear: the tensor tympani that attaches to the malleus and the stapedius that attaches to the stapes (Figure 3). The tensor tympani is innervated by a branch of the fifth cranial nerve, whereas the stapedius is innervated by a branch of the seventh cranial nerve. The muscles receive a dense innervation from motor neurons. The cat stapedius receives almost 1200 motor neurons in total, which is about one motor neuron per muscle fiber. Such high innervation ratios are only found in muscles requiring a high degree of control such as the muscles controlling movements of the eye. This innervation ratio implies a high degree of central control on the action of these muscles.

The effect of contraction of the middle ear muscles is generally to reduce transmission of sound through the middle ear. The maximal effect is about 20 dB. While this is generally the same effect as stimulation of the OC neurons, the frequency ranges of the two systems are different and complimentary. Whereas the middle ear muscle contractions cause loss of sensitivity mainly for low frequencies, the action of the MOC neurons is mainly at middle and high frequencies (Figure 4). The middle ear muscles probably have generally similar functions to the OC systems. The middle ear muscle (MEM) reflex is particularly effective at decreasing the masking of auditory-nerve fibers tuned to high frequencies in the presence of low-frequency noise. This is an important situation because much environmental noise contains low frequencies. In addition, activity in the middle ear muscles of some animals precedes their own vocalizations, evidence that this system prevents self-stimulation.

The action of both MOC neurons and MEM motor neurons causes changes in otoacoustic emissions (OAEs). These emissions are sounds that emanate from the ear either spontaneously or evoked by sound stimulation. OAEs are thought to be generated by outer hair cells, which are electromotile. OAEs can be recorded with a microphone placed in the ear canal in animal studies as well as in human subjects. They can hence be used as a ‘window’ to indicate changes taking place in the outer hair cells. In one type of experiment, an OAE is recorded in one ear in response to a low-level sound stimulus and a reflex elicitor is presented to the opposite ear to evoke a reflex via a crossed central pathway (see below). As both MOC neurons (via their direct endings) and MEM muscles (via changing the sound transmission through the middle ear) can influence the outer hair cells, both systems must be considered when changes in OAEs are observed.

15.5 Screening protocols for hearing disorders

The need to identify the overwhelming majority of newborns with hearing loss in the most cost-effective and efficacious approach currently available has spurred the concept of simplified screening protocols (National Institutes of Health, 1993; Berlin et al., 2001; Lee et al., 2001; Dolphin, 2004; Lin et al., 2007). The audiologic test strategy of the Kresge Hearing Research Lab is to triage every new patient with (1) middle ear testing via tympanograms and ipsilateral and contralateral middle ear muscle reflexes (MEMRs); (2) otoacoustic emissions; (3) neuronal integrity testing via ABR. As we have shown in Table 1, this protocol identifies the majority of infants and children with hearing loss, once the child is shown to have a hearing disorder he/she will undergo a full audiological evaluation that will include pure tone audiograms and/or other behavioral testing. The protocol, however, will miss some of the patients with central auditory processing disorders (CAPD). If CAPD is suspected the subject should undergo a full psychophysical audiological testing.

The important lesson to be learned here is that not a single test is sufficient in representing the patient deficit. MEMRs, otoacoustic emissions and ABRs must be cross-validated by behavioral audiograms and only the totality of the evaluation is truly representative of an individual patient’s hearing abilities.

3.10.4.8 Middle-Ear Muscles

The malleus and stapes each have a tendon attached to a tiny muscle, the tensor tympani muscle and the stapedius muscle, respectively. The muscles contract when exposed to high-level sounds, and are part of the middle-ear reflex arc involving the spiral ganglion neurons, the auditory nerve, cochlear nucleus, the superior olive, the facial nerve nucleus, the facial nerve, and the two middle-ear muscles (Margolis, R. H., 1993). This reflex arc can reduce sound transmission through the middle ear at high levels, and may serve to control the dynamic range of the auditory system and to protect the cochlea at high sound levels. The reflex is slow, and thus does not provide protection to the cochlea against sudden impulsive sounds. The time for the stapedius reflex may be on the order of about 20 ms, while the tensor tympani arc is more than ten times slower (Teig, E., 1972).

Two additional functions are attributed to the middle-ear muscle reflex. Low-frequency sounds, particularly when they are high in level, normally tend to mask mid- and high-frequency sounds due to their upward excitation patterns on the BM. One role of the middle-ear muscles is to reduce the level of low-frequency inputs so they do not mask the higher frequency sounds on the BM (Pang, X. D. and Guinan, J. J., Jr., 1997). A second role of the middle-ear reflex is in the reduction of the audibility of self-generated sounds during speech, mastication, yawning, and sneezing (Simmons, F. B. and Beatty, D. L., 1962; Margolis, R. H. and Popelka, G. R., 1975). Because the reflex arc involves so many mechanisms, its measurement is used clinically to diagnose central and peripheral pathologies.

Recently it has been discovered that there are smooth muscle arrays within the peripheral edge of the tympanic membrane, the annulus fibrosus, in all four of the mammalian (bats, rodents, insectivores, and humans) species studied (Henson, O. W., Jr. and Henson, M. M., 2000; Henson, M. M. et al., 2005). The role of this rim of contractile muscle cells in the par tensa region is not clear, but two suggested possibilities are to maintain tension of the tympanic membrane and to control blood flow to the membrane (Henson, M. M. et al., 2005). Measurements indicate that these smooth muscles can exert control over the input to the cochlea as measured by cochlear microphonics (Yang, X. and Henson, O. W., Jr., 2002).

3.24.4.3 Modulation of Sensory Input During Sleep

Significant increases in CM and auditory-nerve CAPs have been found in guinea-pigs during slow wave sleep, as compared to in the awake state and in paradoxical sleep (Velluti, R. et al., 1989). In these experiments the changes in both cochlear potentials could not be attributed to middle-ear muscle reflex, because the middle-ear ossicles and muscles were removed and auditory stimuli were directly delivered to the animal’s bulla. These results imply that during sleep the central nervous system modulates auditory input through OC efferents. A single systemic injection of gentamicin (150 mg kg−1) that blocks MOC efferent effects in unanesthetized guinea-pigs (see Section 3.24.3.4.2.(i)) did not affect the central modulations of cochlear responses produced through OC efferents by sleep/awake state changes (Pedemonte, M. et al., 2004). However, it is possible that the gentamicin dose used in these experiments was too low to block the efferent effect, and a higher dose may be required. This was the case for the slow suppression produced by contralateral-sound activation of OC efferents that was not blocked by a single dose of 150 mg kg−1 gentamicin, but was completely blocked by a 250 mg kg−1 dose (Lima da Costa, D. et al., 1997a).

In relation to humans, Campbell K. B. and Bartoli E. A. (1986) did not find any reliable differences in the amplitude or latency of brainstem auditory evoked potentials recorded during wakefulness and various stages of sleep. A later study, recording TEOAE during sleep in adults found a progressive increase in TEOAE amplitudes through the night that could reach up to 4 dB at the end of the night, but that was not related to the sleep stages (Froehlich, P. et al., 1993). The work also evaluated the effect of contralateral noise on the amplitude of TEOAE and found a decrease in the reduction of TEOAEs by contralateral noise during the night. Although this study suggests a role of auditory efferents during sleep, another factor that could also modify the magnitude of TEOAEs, as middle-ear muscle activity, was not excluded.