Neuroscience 283 (2014) 64–77

REVIEW LOUDNESS MODULATION AFTER TRANSIENT AND PERMANENT HEARING LOSS: IMPLICATIONS FOR TINNITUS AND HYPERACUSIS P. FOURNIER, a,b,c M. SCHO¨NWIESNER b,d AND S. HE´BERT a,b,c*

Contents Introduction 64 Neural coding of stimulus intensity 65 Distinguishing the coding of intensity and loudness in humans 65 Normal-hearing listeners 65 Hearing-impaired listeners 66 Reversible short-term deprivation/stimulation: neural modifications 66 Normal-hearing animals: deprivation 66 Normal-hearing animals: stimulation 69 Hearing-impaired animals: deprivation and stimulation 69 Behavioral measures 70 Bi-directional modification of loudness by transient (reversible) deprivation and stimulation 71 Normal-hearing listeners 71 Behavioral measures 71 Physiological measures 71 Hearing-impaired listeners 72 Behavioral measures 72 Physiological measures 73 Conclusions 73 Implications for hearing pathologies 73 Tinnitus 73 Hyperacusis 73 Sound therapies 73 Conclusions 74 Acknowledgments 75 References 75

a

School of Speech Pathology and Audiology, Universite´ de Montre´al, Montre´al, Que´bec, Canada b International Laboratory for Research on Brain, Music, and Sound (BRAMS), Universite´ de Montre´al, Montre´al, Que´bec, Canada c Centre de recherche de l’Institut Universitaire de Ge´riatrie de Montre´al (CRIUGM), Montre´al, Que´bec, Canada d Department of Psychology, Universite´ de Montre´al, Montre´al, Que´bec, Canada

Abstract—Loudness is the primary perceptual correlate of sound intensity. The relationship between sound intensity and loudness is not fixed, and can be modified by shortterm sound deprivation or stimulation. Deprivation increases sound sensitivity, whereas stimulation decreases it. We review the effects of short-term auditory deprivation and stimulation on the auditory central nervous system of humans and animals, and we extend the discussion to permanent auditory deprivation (hearing loss) and auditory pathologies of loudness perception. Although there is sufficient evidence to conclude that loudness can be modulated in normal hearing listeners by temporary sound deprivation and stimulation, evidence is scanter for the hearingimpaired listeners. In addition, cortical effects of sound deprivation and stimulation in humans, which may correlate with loudness coding, are still largely unknown and should be the target of future research. This article is part of a Special Issue entitled: Brain compensation. For good? Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved.

INTRODUCTION Loudness is an attribute of an auditory sensation that can be ordered on a scale from quiet to loud (Moore, 2003). Loudness is determined mainly by physical sound intensity (i.e., pressure), although it is also affected by other stimulus attributes such as duration, frequency, bandwidth, and monaural versus binaural presentation (for a review, see Glasberg and Moore, 2002) as well as psychological factors such as mood and emotional exhaustion (He´bert et al., 2012). Considerable evidence from previous studies suggests that the relationship between sound intensity and loudness is not static in time, but can be modified by sound deprivation or stimulation. Deprivation increases sound sensitivity, whereas stimulation decreases it. The neurological bases of those loudness shifts remain largely unknown. A hypothetical

Key words: loudness, deprivation, stimulation, tinnitus, hyperacusis, auditory physiology.

*Correspondence to: S. He´bert, E´cole d’orthophonie et d’audiologie, Faculty of Medicine, Universite´ de Montre´al, C.P. 6128, succursale Centre-Ville, Montre´al, Que´bec H3C 3J7, Canada. Tel: +1-514-3436111x2594. E-mail address: [email protected] (S. He´bert). Abbreviations: 2-DG, 2-deoxycglucose; AAF, anterior auditory field; ABR, auditory brainstem responses; AEPs, auditory-evoked potentials; AI, primary auditory cortex; ASR, acoustic startle reflex; ASSR, auditory steady-state responses; AVCN, anteroventral cochlear nucleus; CAPs, compound action potentials; DPOAEs, distortion products of otoacoustic emissions; fMRI, functional magnetic resonance imaging; HL, hearing level; IC, inferior colliculus; MG, medial geniculate; MSO, medial superior olive; SPL, sound pressure level. http://dx.doi.org/10.1016/j.neuroscience.2014.08.007 0306-4522/Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved. 64

P. Fournier et al. / Neuroscience 283 (2014) 64–77

central gain process, which regulates supra-threshold sensitivity somewhere in the auditory system, is one of the most commonly proposed mechanisms. We examine the current evidence for such a mechanism and others, by reviewing the neural changes following auditory deprivation and stimulation in normal-hearing humans and animals along the auditory pathway. In addition, we examine whether similar changes exist in hearingimpaired listeners, both humans and animals. Finally, we discuss the clinical implications of the results for loudness pathologies including tinnitus and hyperacusis and potential treatments.

NEURAL CODING OF STIMULUS INTENSITY Intensity is first coded at the periphery of the auditory system by the basilar membrane: increased sound pressure increases the amplitude of the basilar membrane displacement, which increases the spike rates in primary afferent auditory nerve fibers and the number of responding fibers (for a review, see Philips, 1987). The large dynamic range (120 dB) of intensity coding in the auditory nerve is achieved by fibers with different spontaneous firing rates and saturation thresholds (Liberman and Kiang, 1978), which project, at least partly, to different target cells in the cochlear nucleus, as well as different thresholds. Rate-level functions are not static, but dynamic. Auditory neurons in the inferior colliculus (IC) of guinea pigs can adapt to the time-average sound level by shifting their thresholds to approach the average (Dean et al., 2005). Within a single neuron, this shift occurs only when the average sound level is higher than the neuron’s sound level threshold. This shift happens rapidly, within about 160 ms, and appears to reduce the maximum spike rate and slope of the neuron’s rate-level function, particularly at high sound levels (Dean et al., 2008). The underlying mechanism is unknown. This shift has been observed in the auditory nerve of cats (Wen et al., 2009), in the IC of guinea pigs (Dean et al., 2005), and in the auditory cortex of ferrets (Rabinowitz et al., 2011) and marmoset monkeys (Watkins and Barbour, 2008). Auditory deprivation and stimulation using, for instance, earplugs and noise generators change the average sound level and thus, assuming a similar mechanism in humans, may change intensity coding in the auditory central nervous system. The effect of hearing loss on the adaptation of rate-level functions to average sound level has not been studied.

DISTINGUISHING THE CODING OF INTENSITY AND LOUDNESS IN HUMANS In this section, we briefly summarize the current knowledge on neural correlates of loudness in humans. We distinguish between neural coding of intensity and neural correlates of loudness, because these are distinct concepts: one refers to a physical attribute of sound, whereas the other refers to its perception. However, because they usually co-vary it is often difficult to dissociate their neural correlates.

65

Normal-hearing listeners Neuroimaging studies have shown increased neural activity with increasing sound intensity in cortical and subcortical areas: the cochlear nucleus, superior olive, IC, medial geniculate (MG) body, and auditory cortical areas (Sigalovsky and Melcher, 2006), including Heschl’s gyri and the planum temporale (Hart et al., 2002, 2003). To better dissociate intensity from loudness neural coding, Ro¨hl and Uppenkamp (2012) used inter-individual differences of loudness ratings from 45 normal hearing subjects. They correlated subjective loudness ratings for a fixed presentation level of 80-dB sound pressure level (SPL) sound to their associated functional magnetic resonance imaging (fMRI) bold signals. They found a significant correlation between the slope of the fMRI signal change and the subjective loudness ratings in the auditory cortices, but not in the IC or MG body. The authors concluded that loudness coding was completed at the level of the auditory cortex because activity at lower levels was more related to sound intensity. The same group also showed that increasing loudness caused by increasing stimulus bandwidth (loudness spectral summation) correlates with activity in the primary auditory cortices, but not in the auditory brainstem (Ro¨hl et al., 2011). Sound-level-evoked activity in the human auditory system has also been studied using auditory brainstem responses (ABR; Picton et al., 1981), auditory-evoked potentials (AEPs) such as N1-P2 (O’Neill et al., 2008), evoked gamma-band responses (Schadow et al., 2007), and auditory steady-state responses (ASSR; Picton et al., 2003). Conflicting results have emerged from ABR studies: loudness growth was found to correlate with ABR amplitude measures (Silva and Epstein, 2010) and with wave V latency in some studies (shorter latency equaled higher loudness; Serpanos et al., 1997), whereas others reported no correlation with wave V latency (Howe and Decker, 1984). There is an extensive literature on the loudness dependence of AEPs (for a review, see O’Neill et al., 2008). The amplitude of the N1-P2 complex increases with increasing sound level (Wutzler et al., 2008; Min et al., 2012; Park and Lee, 2013; Wyss et al., 2013). However, most of these studies did not measure loudness, as mentioned by O’Neill et al. (2008). The relationship between intensity and AEP amplitude appears to be related to the serotonin level in the primary auditory cortex (AI) (Wutzler et al., 2008). A decrease in serotonin concentration increases the intensity dependence of the AEPs, whereas an increase in serotonin decreases it. Two studies have investigated the relationship between the loudness growth function and ASSR in normal hearing subjects (Me´nard et al., 2008; Zenker Castro et al., 2008). In both studies, behavioral loudness growth was measured prior to ASSR recordings at a high-frequency modulation rate, such that neural activity within the brainstem was measured (Picton et al., 2003). Both studies reported a correlation between ASSR amplitude and intensity, and between ASSR and loudness. However, Me´nard et al. (2008) found that statistically loudness had better predictive value than intensity for ASSR amplitude, at least for two out of six measured

66

P. Fournier et al. / Neuroscience 283 (2014) 64–77

loudness categories: loud and comfortable/loud. These findings appear to conflict with the conclusion of Ro¨hl and Uppenkamp (2012), whose findings argue in favor of a cortical, not brainstem, neural correlate of loudness. However, in the Rho¨l and Uppenkamp study, the stimulus level was fixed at 80-dB SPL such that only loudness ratings varied with fMRI bold signal change. In contrast, Me´nard and colleagues did not have such methodological control. Rather, in their study intensity and loudness strongly co-varied (both dB and loudness ratings varied for each participant) such that the respective contributions of loudness and intensity in the ASSR amplitudes are difficult to disentangle. Since the findings of Rho¨l and Uppenkamp seem more robust, the contribution of brainstem to loudness is still uncertain. Hearing-impaired listeners The most common type of hearing impairment is sensorineural hearing loss. It is associated with damage to the hair cells and/or damage to the auditory nerve. Hearing-impaired listeners are not able to hear low-level sounds perceived by normal listeners, whereas highlevel sounds are perceived as loud for both types of listeners. This discrepancy is known as loudness recruitment. In loudness recruitment, the increase in stimulus intensity required to produce a given increase in loudness is smaller than in normal ears, and consequently the loudness function is steeper for lowlevel sounds. Sound transduction by the inner hair cells and conduction of spikes along the auditory nerve are the first steps of intensity coding and damage to either structure will modify intensity coding and loudness. The steeper loudness growth function in hearingimpaired ears was used by Langers et al. (2007) to distinguish fMRI activity related to intensity and loudness. They compared the change in fMRI activity with increasing sound intensity between normal-hearing listeners and listeners with high-frequency sensorineural hearing loss and loudness recruitment. Low- and high-frequency stimuli were used. Low-frequency stimuli increased in intensity steps, while high-frequency stimuli increased in equalloudness steps as matched with the low-frequency stimuli. In normal-hearing listeners, auditory cortical activity grew linearly with both increasing intensity and loudness. In hearing-impaired listeners, however, auditory cortex activity increased more strongly with stimulus intensity compared to normal-hearing listeners for high-frequency tones. However, the rate of activity increase was similar in both groups. Thus, in all subjects, cortical activity increased significantly with increasing loudness, but not with increasing intensity. Although the stage when loudness coding began was not specified, these findings indicate that auditory cortex activity is more closely interrelated to loudness than to intensity. To summarize, there is some evidence that the auditory cortices play a major role in the neural coding of loudness. The role played by the brainstem is less clear and could be linked more to the neural coding of intensity than to loudness per se. For hearing impairment, current evidence suggests that the relationship between intensity and its associated neurophysiological response is

modified without altering the relationship between loudness and its associated physiological response, at least for listeners with moderate to severe highfrequency sensorineural hearing loss (Langers et al., 2007). Investigation of participants with different degrees of hearing loss and hearing loss of different etiologies should be carried out. Overall, in light of these results, particular attention should be paid to cortical-evoked responses when investigating neural responses related to loudness following deprivation and stimulation.

REVERSIBLE SHORT-TERM DEPRIVATION/ STIMULATION: NEURAL MODIFICATIONS This next section will focus on neural modifications in the auditory central nervous system following auditory deprivation and stimulation in animals. These studies are very useful because they provide insights into possible alterations in neural activity such as glucose activity, neurotransmitter receptors expression, protein expression, and evoked activity. It is important to note that animal studies diverge in terms of species (rats, mouse, guinea pigs), animal age (young vs. adults), type of measures (physiological, anatomical, behavioral), deprivation methods (plugs, surgical intervention, environmental deprivation), stimulation methods (broadband noise, tone pips), deprivation depth (in dB), duration, and recovery time. Although these differences make direct comparisons difficult, several conclusions emerge. Normal-hearing animals: deprivation Studies investigating anatomical modifications following deprivation have shown reduced cell size in the anterior ventral cochlear nuclei under monaural but not binaural deprivation (Coleman and O’Connor, 1979; Feng and Rogowski, 1980; Webster, 1983; Trune and Morgan, 1988; Pasic and Rubel, 1989), and during infancy but not adulthood (Blatchley et al., 1983) (see Table 1). Studies investigating physiological modifications following binaural deprivation have found decreased cfos expression compared to controls in the cochlear nucleus and IC during and immediately after deprivation (Keilmann and Herdegen, 1997; Harrison and Negandhi, 2012) (see Table 2). C-fos is a transcription factor expressed within some neurons following depolarization. It is used as a marker of neuronal activity following peripheral stimulation. Enhanced c-fos expression was seen after 1 week of restoration in the ventral and dorsal cochlear nucleus and in the IC, but was significant only in the IC (Keilmann and Herdegen, 1997). After 2 weeks of restoration, no differences were found between c-fos expression in the experimental group compared to controls. Over-expression of c-fos in the IC may still be apparent after several weeks if hearing restoration is incomplete and there is residual hearing loss (Sun et al., 2011). These results suggest that binaural deprivation decreases neural activity during and immediately after occlusion removal. They also suggest a compensatory process in the IC, which ceases after 2 weeks of complete

67

P. Fournier et al. / Neuroscience 283 (2014) 64–77 Table 1. Anatomical modification following deprivation Studies Blatchley et al. (1983)

Animals

25 Sprague–Dawley rats Equal numbers of animals in one of the following groups: Operated at P10 P16 P24 P36 Or control animals Coleman and 12 Sprague–Dawley rats O’Connor (1979) 2 experimental groups: Rats operated at P10 2 monaurally operated 2 controls Rats operated at P16 3 monaurally operated 3 binaurally operated 2 controls Feng and Rogowski Rats (P12) (1980) 3 groups (4–5 rats per group) Monaural occlusion Binaural occlusion Controls

Webster (1983)

Trune and Morgan (1988)

Pasic and Rubel (1989)

Sie and Rubel (1992)

Hearing loss inducer

Monaural/ Duration binaural

Type of measures

Reported effect

Spherical cells Spherical cells reduction Ligation of the Monaural Animals size in the AVCN for every deprivation but external meatus were most prominent for P10 tested at and P16 P70. 60, 54, 46, 34 days of deprivation

Spherical and Monaural: spherical cell Ossicle removal Monaural Animals fusiform cells reduction in the deprived and were binaural tested at size of the AVCN side, more pronounced at P10 than P16 P50. Binaural: no reduction in 40 and spherical cell size 34 days of Binaural and monaural: no deprivation effect of cell size of fusiform cell Silicone rubber Monaural 48 days of Distribution of Monaural: shift in neuron cement in the and deprivation neurons in the distribution toward more external auditory binaural MSO neurons to the open ear meatus side Binaural: no effect on neuron distribution in the MSO No significant effect on External ear Binaural 45 days of Cell size of Three groups of CBA/J mice cell size deprivation globular cells, with 7 animals per group (P45) sutured shut Environmentally cells of the 1 group bilaterally sutured deprived nucleus of the external meatus trapezoid body, 1 group environmentally deprived or inferior 1 control group colliculus CBA/J mice (P8) Meatal occlusion Monaural 37 days of Cellular Cytoplasm reduced 6 Unilaterally deprived deprivation involvement of Mithocondria smaller 6 aged matched controls deprivation on Less active metabolically AVCN cells Auditory nerve endbulbs and bouton seen less and non-auditory boutons more common Tetrodotoxine Monaural None AVCN cell size Cochlear ablation and Mongolians Gerbils (P21-P42) Cochlear and And TTX treatment ablated the 14 animals for physiological ablation binaural Auditory ABR response studies brainstem Monaural: Decrease in 12 animals for anatomical studies response (ABR) AVCN cell size 4 experimental groups: Binaural: No significant Monaural TTX modification in cell size Monaural cochlear ablation between ears within Sham surgery subject, but increase cell Binaural: unilateral TTX and size compared to shamcontralateral cochlear ablation operated bilaterally Mongolians gerbils (P14) Tetrodotoxine Monaural 1 h, 6 h, Protein Decreased protein Monaural TTX Cochlear 18 h, 48 h synthesis in synthesis by AVCN Monaural cochlear ablation ablation AVCN neurons by 1 h Sham surgery Decrease in cell size Sham TTX observable after 18 h after TTX and 48 h after cochlear ablation

hearing restoration or remains active if restoration is incomplete. However, the timeline of the compensatory process is inconsistent with loudness changes observed immediately after deprivation in humans.

Decreased neural activity following binaural deprivation has also been found in connection with auditory-evoked responses (Batkin et al., 1970; McGinn et al., 1973; McGinn and Henry, 1974). Two studies

68

P. Fournier et al. / Neuroscience 283 (2014) 64–77 Table 2. Time-course of c-fos expression following bilateral reversible conductive deprivation

VCN DCN SO IC AC

During deprivation

Immediately after restoration

1 week after restoration

2 week after restoration

4–6 weeks of partial restoration

;1 ;1

;2 ;2 =2 ;2

"2 "2 =2 "2

=2 =2 =2 =2

=3 =3

;1

"3 =3

;, Significant decrease; ", Significant increase; =, no difference to controls; [;"], Not significant decrease or increase. 1. Harrison and Negandhi (2012), 2. Keilmann and Herdegen (1997), 3. Sun et al. (2011).

(McGinn et al., 1973; McGinn and Henry, 1974) found higher AEP thresholds in adult and infant C57BL/6J mice immediately after hearing restoration following 4 days of binaural deprivation. Moreover, higher AEP thresholds persisted at least 24 h after occlusion removal (McGinn and Henry, 1974). They also investigated peak-to-peak amplitude of the evoked response for sound intensities up to 80 dB by 10-dB steps. They found that AEP amplitudes increased gradually for all intensities immediately and up to 3 days after occlusion removal. They also found a steeper AEP peak-to-peak (lV) per stimulus intensity (dB) function at one, two, and three days after hearing restoration. Because AEPs are known to reflect activity within the inferior colliculi of mice, these results suggest diminished evoked activity within this structure following binaural deprivation and a gradual increase up to 3 days after hearing restoration. It is unclear whether these changes are compensatory because no control animals were included and no AEP measures were assessed before the earplugging. There is also evidence that binaural deprivation affects auditory cortical areas. Batkin et al. (1970) found higher cortical-evoked thresholds in 20 infant albino rats after 8 months of sound-attenuated environment compared to 20 matched controls. The ambient level in the sound-attenuated environment was reported to be constant at around 15-dB SPL compared to the control room, where the level fluctuated from 40- to 80-dB SPL. The animals had their toes amputated and mothers were made aphonic by cauterization of the larynx in order to eliminate animal-produced noise. Cortical-evoked thresholds returned to levels similar to those for controls in a subgroup of deprived animals after 3 weeks of exposure to ambient noise following the 8 months of deprivation.

Studies investigating physiological modifications in monaural deprivation have found decreases in 2-deoxycglucose (2-DG) uptake in the deprived ear pathway, including the anteroventral cochlear nucleus (AVCN), IC, MG, AI, and anterior auditory field (AAF) compared to control animals, but no alteration in the medial superior olive (MSO) (Hutson et al., 2008, 2009) (see Table 3). Glucose utilization in the auditory pathway of the open ear (contralateral to the deprived ear) has shown a different pattern, with decrease in the IC, MG and AI, more utilization in the MSO, and no modification in the AVCN and AAF (see Table 3). Additionally, up-regulation of GluR3 AMPA receptors and decreased expression of GlyR1 subunits in the bushy cells and fusiform cells of the VCN and DCN have been reported on the ipsilateral side after 1 day of unilateral ear plugging (Whiting et al., 2009). All modifications were fully reversible 1 day after unplugging. Similar results were also reported after 7 days of monaural ear plugging for both sides (Wang et al., 2011). Evoked activity in the brainstem, midbrain, and auditory cortex has also been investigated following monaural deprivation (Popescu and Polley, 2010). Sprague–Dawley rats were deprived monaurally starting post-natal day 14, 28, or 140 for a period of 60–74 days using surgical ear canal ligation. Click-evoked ABR were measured in the ligated ear, open ear, and both ears of control animals immediately after the opening of the ligated ear. The quantitative analysis of wave Ia, I, and II, generated by the inner hair cells, the spiral ganglion, and cochlear nucleus globular cells respectively, evoked by 80-dB clicks showed normal amplitude for wave Ia and I after removal of the occlusion. This suggests an immediate restoration to normal function in these areas

Table 3. Activity in the central pathway of the plugged ear (monaural deprivation) and the contralateral ear, respectively, during the deprivation period (2-DG) and immediately after the removal (Evoked response)

CN MSO IC MG AI AAF

Plugged ear pathway (2-DG)

Contralateral ear pathway (2-DG)

Plugged ear pathway (Evoked response)

;1 =1 ;1;2 ;2 ;2 ;2

=1 "1 =1;2 ;2 ;2 =2

;3

Contralateral ear pathway (Evoked response)

;3

=3

;3

=3

;, Significant decrease; ", Significant increase; =, No difference from control ear, [;"], No significant decrease or increase. 1. Hutson et al. (2009), 2. Hutson et al. (2008), 3. Popescu and Polley (2010).

P. Fournier et al. / Neuroscience 283 (2014) 64–77

in all animals. However, wave II amplitude remained attenuated despite hearing restoration. This suggests no immediate reversal of neural activity within the cochlear nucleus. In the AI and central nucleus of the IC of all animals, evoked responses in the ligated ear immediately after opening were significantly suppressed compared to sham-control animals (see Table 3). The suppression was equally distributed across the tonotopic map (not restricted to any frequency), although the ligation created more deprivation for high frequencies. In adult animals, no increase in evoked activity was observed in either the ICc or AI when stimulation was presented in the open ear (contralateral to the ligation ear). Taken together, these findings suggest a decrease in neural activity within the brainstem, midbrain, and auditory cortex during and immediately after auditory deprivation, either binaurally or monaurally. This decrease in activity is accompanied by an up-regulation of excitatory receptors and a decrease in inhibition receptors, at least in the pathway of the deprived ear when monaural deprivation is applied. The evoked neural responses are also significantly suppressed following deprivation. The recovery seems to be expressed differently in different structures. Normal-hearing animals: stimulation The effect of non-traumatic sound on the auditory system has been investigated since the end of the 1980s, mainly to determine its protective properties on subsequent traumatic sound exposure (for a review, see Niu et al., 2007). Exposure to low-level non-damaging sounds before and even after sound trauma has been shown to reduce subsequent trauma-related damage. A proposed underlying mechanism implies dopamine release from the lateral efferent pathways, which then inhibits auditory nerve activity, as well as a protective effect due to activation of the hypothalamo–pituitary–adrenal (HPA) axis. Although most studies have focused on the protective effect of non-traumatic noise rather than the actual physiological impact of non-traumatic noise on the auditory system, other mechanisms have been explored. Kujawa and Liberman (1999) compared compound action potentials (CAPs) and distortion products of otoacoustic emissions (DPOAEs) in guinea pigs exposed to nontraumatic noise (85-dB SPL, octave-band noise, 6 h daily, 10 consecutive days) and age-matched control guinea pigs. Animals exposed to sound showed greater response amplitude for both CAPs and DPOAEs compared to controls. The authors concluded that the non-traumatic sounds used in their study changed the physiology of the outer hair cells. The effect of exposure to moderate sound levels on the auditory central nervous system has been extensively reviewed (Pienkowski and Eggermont, 2011, 2012; Eggermont, 2013). Passive exposure to moderate (680-dB SPL) continuous tone (Canlon and Fransson, 1995), tone pips (Pienkowski and Eggermont, 2009, 2010a,b) and noise (Pienkowski et al., 2011, 2013), have been reported to have no effect on ABR thresholds, suggesting no modifications in the brainstem. Despite no significant change in ABR thresholds, passive exposure to

69

sound had a dramatic suppression effect on neural activity within the primary and secondary auditory cortices (Pienkowski and Eggermont, 2009, 2010a,b; Pienkowski et al., 2011, 2013). For instance, after exposing cats to 4–20-kHz tones at 68-dB SPL continuously 24 h per day for 6 weeks, Pienkowski and Eggermont (2009) found decreased responsiveness of the AI to sound frequencies of the exposure band but increased responsiveness for frequencies at the outer edges of the band using multi-unit recordings and local field potentials. They also found tonotopic reorganization, including underrepresentation of units with characteristic frequencies compromised in the tone assemblies (4–20 kHz) within AI and AII and overrepresentation of units with frequencies below (20 kHz) stimulus frequencies. Cortical suppression appears to increase with exposure time up to 4 weeks (Pienkowski et al., 2011). Some recovery was seen after a few weeks by placing animals in a quiet room, but full recovery was not obtained after 8–12 weeks (Pienkowski and Eggermont, 2009). Mechanisms of the exposure-induced plasticity remain unknown. Together these results suggest that moderate levels of sound stimulation lead to modification in outer hair cells physiology, have no effect on brainstem responses close to threshold, but are associated with diminished responses from neurons with characteristic frequencies in the stimulated frequency range and increased responses from neurons with characteristic frequencies outside that range. Hearing-impaired animals: deprivation and stimulation Studies in animals with trauma-induced high-frequency hearing loss have shown that putting animals into an enriched acoustic environment reduces hearing loss and prevents the cortical reorganization of the tonotopic map typically observed when animals are put into a silent environment after trauma (Norena and Eggermont, 2005, 2006). The acoustic stimulation, however, has to match the frequencies impaired by the hearing loss, that is, high-frequency complex stimulus for a high-frequency hearing loss. It is unknown whether low-frequency stimulation would have similar effects on low-frequency hearing loss. Other neural modifications such as protein expression, glucose activity and evoked activity have not been investigated in deprived or stimulated animals following noise trauma. Behavioral measures. Loudness is by definition a perception. It is difficult to relate perceptual changes to behavior in an animal model, because it usually requires training the animals to report their perceptual state. The acoustic startle reflex (ASR) has been proposed as a potential behavioral marker of loudness modification in animals that do not require training. The ASR is a simple, primitive reflex produced by a sudden and unexpected loud sound, and it involves a limited number of neurons in the brainstem (for a review, see Koch, 1999). The magnitude of the reflex evoked at high intensity is enhanced following three to four weeks of deprivation (Sun et al., 2011) (see Table 4). Moreover, when a

70

P. Fournier et al. / Neuroscience 283 (2014) 64–77

wide range of dynamic sound levels is presented, the response magnitude per sound level becomes steeper, similar to loudness growth functions in humans, with modifications seen mainly for the highest levels (>100-dB

SPL). Higher acoustic reflex reactivity has also been proposed as a potential animal model of hyperacusis (or hypersensitivity to loud sound), mainly because of the enhanced reactivity to high-level sounds (Ison et al.,

Table 4. Acoustic startle response in hearing-impaired animals Studies

Animals

Hearing loss inducer

Monaural/ Recovery binaural time

Type of measures

Chen (1978) BALB/C mice of ±P21 40 Acoustic trauma 40 Sham

125–127-dB bell sound for 25 s

Binaural

1, 3, or 7 days Behavioral

Sun et al. (2012)

Narrow band noise, 12-kHz, 120-dB SPL, 1 h

Binaural

1h

Decreased startle reflex thresholds in the acoustic trauma group but higher startle reactivity for high-intensity stimuli (seen at 0, 1, 3, and 7 days after the trauma) Audiogenic seizure in the acoustic trauma group after 3 and 7 days only Behavioral Decreased sound-evoked potentials in the IC and Increased sound-evoked potentials in the AC physiological (recovered 2 days post-exposure) Startle response increased for high-intensity stimulus despite hearing loss

Acoustic trauma Octave-band noise (8–16 kHz) for 2 h at 100- or 94-dB SPL

Binaural

10–14 days Behavioral (Subset tested and at 3 days, physiological 6 weeks or 10 weeks)

25 Sprague–Dawley rats (P 3–6 months) Electrophysiological test (n = 17) Behavioral test (n = 8) Hickox and CBA/CaJ male mice Liberman (P16–18 weeks) (2014) Mice noise exposed to ‘‘neuropathic noise’’ Mice exposed to ‘‘non-neurophatic noise’’ Control mice

Chen et al. (2013)

18 Hamsters (LVG strain) (P67-P68) 9 Noise exposed 9 Controls

Acoustic trauma 10 kHz tone at 115-dB SPL for 4h

Binaural

2 weeks post- Behavioral exposure to and 28 weeks physiological

Ison et al. (2007)

16 C57BL/6J mice, 8 female and 8 male (P10 weeks) Tested from 10 to 53 weeks of age Adult male Sprague– Dawley rats (P3– 5 months)

Normal aging

Binaural

None

Behavioral and physiological

Salicylate injection Binaural (systemic and local ‘‘cochlea’’)

1 h, 4 h, 1 day Behavioral and 2 days and after injection physiological

Lu et al. (2011)

Adult male Sprague– Salicylate injection Binaural Dawley rats (P3– (systemic and 5 months) local ‘‘cochlea’’)

1 h, 1 day and Behavioral 2 days post and treatment physiological

Sun et al. (2011)

30 Sprague-Dawley Tympanic rat pups (P16) membrane 14 deprived ruptured bilaterally eight controls

1–16 weeks

Sun et al. (2009)

Binaural

Behavioral and physiological

Reported effect

Complete threshold recovery after 1–2 weeks ABR: reduced wave I, later peaks unchanged or enhanced, suggesting compensatory neural hyperactivity in the auditory brainstem in both groups acoustically traumatized (more in the group of 100 dB then 94 dB) Higher startle reflex amplitude and higher inhibition of the reflex (ASR) by a prepulse (PPI) at high-intensity sound in mice with cochlear neuropathy (mice with reduced wave I/traumatized at 100 dB) Clear elevation of the startle response for stimulus intensity over 105-dB SPL that persisted over 28 weeks in exposed animals Reduction in startle response with the presence of background noise only in the acoustic trauma group Linear increase for 3 kHz and 6 kHz thresholds with age After 6 months of age increased acoustic startle reflex reactivity for high-intensity stimulus for 3 kHz and 6 kHz Increased amplitude of sound-evoked field potentials of the auditory cortex (AC) of conscious rats, but not the inferior colliculus (IC) Increased amplitude of the startle response for high-intensity stimulus after salicylate Increased amplitude of sound-evoked activity in AC and reduced spontaneous firing rates Increased ASR amplitude after systemic salicylate injection Suppression of sound-evoked activity increase in AC and ASR amplitude by GABAB Agonist Audiogenic seizures 2 weeks after the surgery and maintained over16 weeks in the binaural deprived group Lower startle reflex thresholds and higher magnitude at high-intensity stimulus for the deprived group Strong C-fos staining in the IC of deprived rats compared to controls C-fos staining in the CN or AC light or absent in both groups

P. Fournier et al. / Neuroscience 283 (2014) 64–77

2007; Sun et al., 2009). This hyper-responsiveness of the ASR to loud sounds has also been seen following permanent hearing loss of different etiologies, such as acoustic trauma (Chen, 1978; Sun et al., 2012; Chen et al., 2013; Hickox and Liberman, 2014), salicylate injection (Sun et al., 2009; Lu et al., 2011), conductive hearing loss (Sun et al., 2011) and presbyacusis (Ison et al., 2007) (see Table 4), all conditions also known to produce tinnitus. However, enhanced ASR reactivity in humans following deprivation needs to be tested and correlated with loudness growth functions before the startle reflex can be considered as a behavioral correlate of loudness. ASR reactivity modulation after brief stimulation has not been experimentally tested.

BI-DIRECTIONAL MODIFICATION OF LOUDNESS BY TRANSIENT (REVERSIBLE) DEPRIVATION AND STIMULATION Normal-hearing listeners

Behavioral measures. Loudness can be reversibly changed in normal hearing listeners by temporary auditory deprivation or stimulation, either monaurally or binaurally (Formby et al., 2003, 2007; Munro and Merrett, 2013; Schaette et al., 2014). Temporary auditory deprivation by ear plugging increases the slope of loudness growth functions, particularly from comfortable to high loudness levels (Formby et al., 2003, 2007; Schaette et al., 2014) (shift of loudness growth functions in Fig. 1). In contrast, auditory stimulation (e.g., using a noise generator) has the reverse effect, producing shallower loudness growth functions (Formby et al., 2003, 2007; Munro and Merrett, 2013) (loudness growth function is shifted to the right in Fig. 1). In addition, the modulation is not frequency-specific: loudness can be modified at frequencies that are unaffected or barely affected by deprivation or stimulation (Formby et al., 2003, 2007; Munro and Merrett, 2013; Schaette et al., 2014).

71

Monaural stimulation and deprivation do not significantly change the loudness growth function in the contralateral ear (Munro and Merrett, 2013; Schaette et al., 2014). While changes in the stapedial reflexes have been reported (see below), loudness growth functions may be insufficiently sensitive to capture these small effects. However, both binaural deprivation and stimulation change the loudness growth functions similarly in both ears (Formby et al., 2003, 2007). To date, binaural and monaural manipulations have not been directly compared. Consequently, it is unclear whether binaural and contralateral monaural manipulations have the same effect within an ear, or whether the added ipsilateral stimulation/deprivation in the binaural condition contributes to the effect. The minimum duration required to produce an observable shift in loudness functions is unknown. Most studies have reported deprivation and stimulation durations between five consecutive days (Munro and Merrett, 2013) and 4 weeks (Formby et al., 2007). The loudness modulation effect is most often reported as a difference score (in dB) between intensity measured at baseline and after the deprivation period for each loudness category. Maximal shifts in loudness following deprivation appear to plateau around 6 dB after 5 days (Formby et al., 2003; Munro and Merrett, 2013; Schaette et al., 2014) although one study reported more unambiguous effects over a four-week period (Formby et al., 2007). Furthermore, because all these studies used attenuation and stimulation intensity of 20–25 dB, it is unclear whether there is a dose–response relationship between deprivation/stimulation and loudness shift, that is, whether increased attenuation/stimulation induces stronger modulation. Recovery to pre-experimental levels can take from 24 h (Schaette et al., 2014) to 1 week after removal of the devices (Formby et al., 2003). Although the stimulation/deprivation duration probably correlates with the required recovery time, no empirical data are available. Physiological measures. In addition to loudness growth functions, sound deprivation and stimulation also

Fig. 1. Theoretical loudness growth functions in a normal-hearing listener (center curve), a normal-hearing listener following auditory deprivation (steeper function, mainly at high intensities), and a normal-hearing listener following stimulation (shallower function, mainly at high intensities).

72

P. Fournier et al. / Neuroscience 283 (2014) 64–77

induce physiological modifications in the auditory periphery and central auditory systems. The stapedial reflex, which is mediated by a peripheral reflex arc, is triggered at higher sound levels after monaural stimulation (Munro and Merrett, 2013) and at lower sound intensities after deprivation in normal listeners (Decker and Howe, 1981; Munro and Blount, 2009; Maslin et al., 2013; Schaette et al., 2014). The opposite effects have been reported in control ears (Munro and Blount, 2009; Maslin et al., 2013; Schaette et al., 2014), but the effects are generally minor, and might be artefacts (as suggested by Decker and Howe, 1981). In the deprived ear, the decrease in stapedial reflex threshold occurs after short-term deprivation periods as short as 10 h (Decker and Howe, 1981). Changes at early stages of the central auditory system, measured by recording brainstem potentials, were also observed after all deprivation periods in that study (10, 20, and 30 h). These changes include shorter wave I latency post-deprivation for all deprivation periods. No significant differences were noted for the control ear. However, it remains unclear whether stapedial reflex thresholds (a brainstem reflex) are good indicators of loudness (a higher-level, cognitive judgment), because some studies have reported no correlation between stapedial reflex thresholds and uncomfortable loudness levels (ULL) (Olsen, 1999) or categorical loudness scales (Olsen et al., 1999a,b). Two further studies reported no significant correlation between loudness modification and stapedial reflex thresholds after 5 days of monaural amplification or earplugging, despite significant changes in both measures (Munro and Merrett, 2013; Schaette et al., 2014). The evidence thus appears to favor two different mechanisms. However, one study showed that loudness functions and stapedial reflex thresholds changed with the same time course and regained their baseline values over 24 h after earplug removal (Schaette et al., 2014). Thus a relationship might still exist between loudness and stapedial reflexes.

Deprivation effects have also been examined in the auditory cortex (Maslin et al., 2013). One study assessed changes in the acoustic reflex threshold and hemispheric laterality following 7 days of monaural deprivation. A significant decrease was found in the ipsilateral acoustic reflex threshold in the deprived ear and a significant (but small: 2 dB) increase in the ipsilateral acoustic reflex threshold in the control ear. No difference in interaural asymmetry was found in the AI or non-AI using fMRI. Thus, contralateral activation was greater than ipsilateral activation in all areas both before and after earplugging. However, in that study of Maslin and collaborators, cortical activity was measured with the earplug on before and after deprivation. The results therefore diverge from those of other studies that measured pre- and post-deprivation activities immediately before and after earplug removal. More data are needed before drawing conclusions on the effects of earplugging on the auditory cortex. Hearing-impaired listeners Behavioral measures. Loudness modification using auditory stimulation has also been reported in individuals with permanently damaged hearing. Monaural fitting of hearing aids in first-time users increased ULL in the aided ear after 12–60 months (mean of 32 months) of use (Munro and Trotter, 2006; Munro et al., 2007). Comparing experienced users of monaural and binaural aids to unaided listeners (mean of five and six years of hearing use, respectively) with similar thresholds shifts, Hamilton and Munro (2010) found higher ULL in fitted compared to non-fitted ears in experienced users, and higher levels in bilateral users than in unaided listeners. Olsen et al. (1999c) also compared loudness functions between 18 full-time hearing aid users (P15 h per day and 0.5– 27 years of experience) and 18 non-users with similar hearing thresholds. They used a categorical loudness scale to compare between-group mean levels for each

Fig. 2. Theoretical loudness growth functions in a normal-hearing listener, a listener with mild hearing loss showing recruitment (steeper loudness function at low intensities), and a listener with mild hearing loss following amplification (steeper loudness function at low intensities and shallower loudness function at high intensities).

P. Fournier et al. / Neuroscience 283 (2014) 64–77

category. They found significant differences for the loud category (users: mean of 106.6-dB hearing level (HL), non-users: mean of 102.5-dB HL) and very loud category (users: mean of 112.4-dB HL, non-users: mean of 110.7dB HL), with no significant differences for lower categories. These results suggest a similar pattern of loudness modulation for hearing-impaired individuals and normal listeners, with shallower loudness function slopes after stimulation, and affecting mainly the loudness of highintensity sound (see Fig. 2). Physiological measures. Auditory stimulation in hearing-impaired listeners is also associated with physiological modifications. Differences in stapedial reflex thresholds have been reported between fitted and non-fitted ears of experienced monaural hearing aid users with similar hearing loss in both ears, with lower stapedial reflex thresholds for non-fitted ears (Munro et al., 2007). In addition to a decrease in the slope of loudness growth functions following 6 months of binaural amplification, particularly for the loudest categories, Philibert et al. (2005) found a concurrent reduction in wave V latency in the right ear only. Munro et al. (2007) compared ABR between fitted and non-fitted ears of experienced and non-experienced hearing aid users and found no difference in wave III and wave V latencies, but higher wave V to SN10 amplitude at some stimulus levels (70- and 80-dB nHL). All the participants had symmetrical high-frequency hearing loss. These results suggest that stimulation is associated with modifications of brainstem-evoked activity, as seen in normal listeners. Conclusions. Taken together, these results provide evidence of a bi-directional modification of loudness, more specifically, of loudness growth functions, following deprivation and stimulation in normal-hearing persons. The reviewed studies indicate that loudness can be modified by acoustic stimulation even when the auditory periphery is permanently damaged, and that this is accompanied by brainstem-evoked activity changes, as was observed in normal-hearing listeners. However, because most hearing-impaired participants had high-frequency sensorineural hearing loss, presumably linked to aging, comparisons with young normal-hearing individuals with reversible hearing loss and stimulation can only be tentative. The generalization of results to more severe hearing loss and different etiologies is also speculative. Interpretation of the brainstem data is also hindered by the lack of physiological data from normal-hearing and hearingimpaired listeners.

IMPLICATIONS FOR HEARING PATHOLOGIES Tinnitus Tinnitus, defined as the perception of a sound in the absence of an external sound source, is of particular interest in terms of short deprivation and stimulation. Phantom sound can be perceived by normal hearers placed in sound-deprived environments such as a soundproof booth (Heller and Bergman, 1953; Tucker

73

et al., 2005; Del Bo et al., 2008; Knobel and Sanchez, 2008). In a recent study, 11 of 18 participants wearing earplugs continuously for 7 days reported a tinnitus-like perception at the end of the earplug period (Schaette et al., 2012). Ear plugging induced high-frequency hearing loss and high-frequency tinnitus, assessed by likeness rating methods. Indeed, the tinnitus spectrum comprises the frequencies most affected by earplug attenuation. The phantom sound gradually disappeared after earplug removal, and all participants had returned to normal by the seventh day. Unfortunately, loudness growth functions within and outside the deprived frequency region were not measured. This could have revealed a relationship between loudness modification and tinnitus. Hyperacusis There is also evidence that loudness is pathological in tinnitus sufferers. Behaviorally, tinnitus has been associated with a high prevalence of self-reported hypersensitivity (80%) (Dauman and Bouscau-Faure, 2005), lower ULL, steeper loudness growth functions (He´bert et al., 2013), and higher ASR responsiveness (Fournier and He´bert, 2013). Tinnitus is also associated with a wave I amplitude reduction of the ABR response and normal (Schaette and McAlpine, 2011) or greater (Gu et al., 2012) wave V amplitude. One study attempted to distinguish the neural correlates of tinnitus from those of hypersensitivity to sound (Gu et al., 2010). Participants included individuals without tinnitus but with moderate hypersensitivity to sounds, individuals with tinnitus but without sound hypersensitivity, and matched controls. Tinnitus sufferers without hypersensitivity were difficult to recruit. Compared to matched controls, and for the same stimulus level using fMRI, higher activity was found in the midbrain, thalamus, and AI in hypersensitive individuals. In tinnitus individuals, higher activity was found in the AI only. Sound therapies In terms of stimulation, sound therapy using hearing aid(s), sound generator(s), or other stimulation devices (CDs, music) with or without counseling or other forms of information, is commonly used for tinnitus management. The effectiveness of different sound therapies, combined or separately, has been repeatedly investigated, with findings published in Cochrane Reviews (Tinnitus retraining therapy: Phillips and McFerran 2010; Sound therapy: Chisholm and El Refaie, 2012; Hearing aids: Hoare et al., 2014). The overall conclusion is that there is little evidence to support the effectiveness of sound therapy for tinnitus management, mainly due to the lack of high quality double-blind randomized controlled trials. The disparity in outcome measures between studies has also posed a barrier to metaanalysis and comparisons between studies. There are two key concepts in tinnitus management: the tinnitus percept and the psychological reaction to it, including distress. Most studies in the above-mentioned Cochrane Reviews reported that sound therapies produced some improvement in scores on validated questionnaires

74

P. Fournier et al. / Neuroscience 283 (2014) 64–77

addressing tinnitus distress, severity, or handicap. However, the effect of sound therapy on the tinnitus percept has not been extensively studied. Moffat et al. (2009) reported no change in tinnitus loudness or pitch after 1 month of high-bandwidth amplification, and a weak effect from conventional amplification (reduced contribution of low frequencies in the tinnitus spectrum after 1 month). These results suggest that sound therapy at least partly reduces tinnitus distress, but does not affect the percept. Studies with better experimental designs could determine whether this is a real or placebo effect. Improvement in hypersensitivity to sounds after sound therapy has been reported in many studies as a secondary treatment benefit for tinnitus (Formby and Gold, 2002; Davis et al., 2007; Formby et al., 2007; Formby and Keaser, 2007). Indeed, improvement in ULL in tinnitus sufferers was reported following hearing aid use (Formby and Keaser, 2007), noise generators (Formby and Gold, 2002; Formby et al., 2007; Formby and Keaser, 2007), and Neuromonics devices, or sound therapy devices using a spectrally modified complex sound (Davis et al., 2007). There is some evidence that noise generators provide more improvement in ULLs than hearing aids in tinnitus sufferers (Formby and Keaser, 2007). Moreover, improvement in loudness growth functions was also reported following sound therapy (enriched acoustic environment) in individuals suffering from hypersensitivity to loud sounds (or hyperacusis), with shallower loudness functions post-treatment (Norena and CheryCroze, 2007). To summarize, tinnitus can be produced in normal listeners by short sound deprivation. Furthermore, hypersensitivity to loud sounds is one of the most common comorbidities in tinnitus sufferers, such that common mechanisms for both may be postulated (Noren˜a, 2011; but see Zeng, 2013, and Knipper et al., 2013 for different positions). Sound therapy is widely used as a tinnitus treatment, despite little evidence for its effectiveness. The well-documented loudness modulation by sound therapy in tinnitus sufferers could contribute to reduce psychological distress in these patients, and this direction should be further investigated.

CONCLUSIONS There is sufficient evidence to conclude that loudness can be modulated in normal listeners by sound deprivation and stimulation. However, many questions remain unanswered, including (a) whether or not there is a dose–response relationship between increasing deprivation/stimulation and increasing loudness modulation, (b) what is the minimum duration required to produce an observable loudness shift, and (c) how persistent is loudness modulation over time. Both deprivation and stimulation produce physiological modifications throughout the central auditory system. In humans, the available data suggest that these two manipulations modify neural activity at the brainstem level. However, the different methodologies used across studies make it difficult to localize and define the nature of those modifications. The effects of deprivation and

stimulation on cortical activity in humans remain largely unknown. The animal literature points toward a decrease in neuronal activity from the cochlear nucleus up to the cortical areas during and immediately after a deprivation period. Stimulation in animals does not appear to be associated with modifications of brainstem responses, but instead with diminished cortical-evoked activity, which affects only the neurons with the frequencies that are compromised by stimulation. Moreover, the effects are not restricted to the manipulated frequencies. The evidence points toward an essential role by the auditory cortex in loudness coding in humans. Future studies could attempt to identify the physiological correlates of loudness modulation at this level. In light of the current review, there is no evidence of a neurophysiological compensation or gain during or immediately after a short reversible deprivation or stimulation period. The gain theory implies that neurons in the higher structures of the central auditory nervous system will increase their activity to compensate for the deprivation and vice versa for stimulation. This should be observable by higher evoked responses in higher centers for the same stimulus level immediately after deprivation compared to baseline conditions. Yet the current literature suggests a decline in neural-evoked activity following deprivation, rather than an increase. Since human studies in this review had protocols similar to most animal studies (i.e., earplug for a few days and then testing immediately after removal), it can be argued that the decrease in loudness in humans might be linked to a decrease in neural activity rather than an increase, at least in normal listeners. The few available physiological studies investigating neural-evoked responses in normal human listeners following deprivation or stimulation do not provide convergent evidence of neural modifications despite changes in stapedial reflex thresholds. There is still a possibility that a compensation or gain process occurs after longer deprivation or recovery periods, as shown in the Keilmann and Herdegen study (1997) after 1 week following the end of a deprivation period. Further research is needed to clarify this question. Loudness modulation can be achieved by stimulation in individuals suffering from hearing impairment, tinnitus, and hyperacusis. According to the current review, there is enough evidence to conclude that (at least some) modulation of loudness by stimulation in hearingimpaired individuals is achievable. Yet, the underlying neurophysiological mechanisms are even less understood than for normal listeners. To our knowledge, the impact of deprivation on supra-thresholds sensitivity of hearing-impaired individuals has not been studied. If transient tinnitus can be produced by temporary deprivation in normal listeners, could permanent tinnitus be produced in hearing impaired using the same methodology? Because tinnitus appears to be intrinsically related to hypersensitivity to loud sounds, the impact of sound therapy on loudness should be considered in randomized controlled trials investigating the effectiveness of sound therapy. Improved loudness measures could in turn explain improvements in distress and handicap scores without modification of the

P. Fournier et al. / Neuroscience 283 (2014) 64–77

tinnitus percept. This hypothesis is further supported by a population study relating ULL to both tinnitus prevalence and severity (He´bert et al., 2012), in which ULL was a better predictor of both tinnitus prevalence and severity compared to hearing loss, suggesting the potential role of pathological loudness in the generation of tinnitus and the associated psychological distress. Since there is some evidence for increased brain responses associated with hyperacusis as well as altered loudness percepts following deprivation/stimulation, neural correlates should be found when loudness percepts are modulated by sound manipulation. Yet, evidence for such neural correlates in normal hearing is lacking and should be the target of future studies.

Acknowledgments—This research was supported by the Institut Robert-Sauve´ en Sante´ et Se´curite´ au Travail (IRSST) and the Re´seau de Neurobioimagerie du Que´bec (RBIQ). PF was supported by the Fonds de Recherche du Que´bec – Sante´ (FRQS) and Institut Robert-Sauve´ en Sante´ et Se´curite´ au Travail (IRSST). We thank Dr. Larry E. Roberts for insightful comments on an earlier version of this paper.

REFERENCES Batkin S, Groth H, et al (1970) Effects of auditory deprivation on the development of auditory sensitivity in albino rats. Electroencephalogr Clin Neurophysiol 28:351–359. Blatchley BJ, Wiliams JE, et al (1983) Age-dependent effects of acoustic deprivation on spherical cells of the rat anteroventral cochlear nucleus. Exp Neurol 80:81–93. Canlon B, Fransson A (1995) Morphological and functional preservation of the outer hair cells from noise trauma by sound conditioning. Hear Res 84:112–124. Chen C-S (1978) Acoustic trauma-induced developmental change in the acoustic startle response and audiogenic seizures in mice. Exp Neurol 60:400–403. Chen G, Lee C, et al (2013) Behavioral evidence for possible simultaneous induction of hyperacusis and tinnitus following intense sound exposure. J Assoc Res Otolaryngol 14(3):413–424. Chisholm HJ, El Refaie A (2012) Sound therapy (masking) in the management of tinnitus in adults (Review). Cochrane Database Syst Rev 12. Coleman JR, O’Connor P (1979) Monaural and binaural sound deprivation on cell development in the anteroventral cochlear nucleus of rats. Exp Neurol 64:553–566. Dauman R, Bouscau-Faure F (2005) Assessment and amelioration of hyperacusis in tinnitus patients. Acta Otolaryngol 125(5):503–509. Davis PB, Paki B, et al (2007) Neuromonics tinnitus treatment: third clinical trial. Ear Hear 28(2):242–259. Dean I, Harper NS, et al (2005) Neural population coding of sound level adapts to stimulus statistics. Nat Neurosci 8(12):1684–1689. Dean I, Robinson BL, et al (2008) Rapid neural adaptation to sound level statistics. J Neurosci 28(25):6430–6438. Decker NT, Howe SW (1981) Short-term auditory deprivation effect on brainstem electrical response. Hear Res 4:251–263. Del Bo L, Forti S, et al (2008) Tinnitus aurium in persons with normal hearing: 55 years later. JAMA Otolaryngol Head Neck Surg 139(3):391–394. Eggermont JJ (2013) On the similarities and differences of nontraumatic sound exposure during the critical period and in adulthood. Front Syst Neurosci 7. Feng AS, Rogowski BA (1980) Effects of monaural and binaural occlusion on the morphology of neurons in the medial superior olivary nucleus of the rat. Brain Res 189:530–534.

75

Formby C, Gold SL (2002) Modification of loudness discomfort level: evidence for adaptative chronic auditory gain and its clinical relevance. Semin Hear 23(1):21–33. Formby C, Keaser M (2007) Secondary treatment benefits achieved by hearing-impaired tinnitus patients using aided environmental sound therapy for tinnitus retraining therapy: comparisons with matched groups of tinnitus patients using noise generators for sound therapy. Semin Hear 28(4):276–294. Formby C, Sherlock LP, et al (2003) Adaptive plasticity of loudness induced by chronic attenuation and enhancement of the acoustic background (L). J Acoust Soc Am 114(1):55. Formby C, Gold S, et al (2007) Secondary benefits from tinnitus retraining therapy: clinically significant increases in loudness discomfort level and expansion of the auditory dynamic range. Semin Hear 28(4):227–260. Formby C, Sherlock L, et al (2007) Adaptive recalibration of chronic auditory gain. Semin Hear 28(4):295–302. Fournier P, He´bert S (2013) Gap detection deficits in humans with tinnitus as assessed with the acoustic startle paradigm: does tinnitus fill in the gap? Hear Res 295:16–23. Glasberg BR, Moore BCJ (2002) A model of loudness applicable to time-varying sound. J Audio Eng Soc 50(5):331–342. Gu JW, Halpin CF, et al (2010) Tinnitus, diminished sound-level tolerance, and elevated auditory activity in humans with clinically normal hearing sensitivity. J Neurol Neurophysiol 104(6): 3361–3370. Gu JW, Herrmann BS, et al (2012) Brainstem auditory evoked potentials suggest a role for the ventral cochlear nucleus in tinnitus. J Assoc Res Otolaryngol 13(6):819–833. Hamilton A-M, Munro KJ (2010) Uncomfortable loudness levels in experienced unilateral and bilateral hearing aid users: evidence of adaptive plasticity following asymmetrical sensory input? Int J Audiol 49(9):667–671. Harrison RV, Negandhi J (2012) Resting neural activity patterns in auditory brainstem and midbrain in conductive hearing loss. Acta Otolaryngol 132(4):409–414. Hart HC, Palmer AR, et al (2002) Heschl’s gyrus is more sensitive to tone level than non-primary auditory cortex. Hear Res 171:177–190. Hart HC, Hall DA, et al (2003) The sound-level-dependent growth in the extent of fMRI activation in Heschl’s gyrus is different for lowand high-frequency tones. Hear Res 179(1–2):104–112. He´bert S, Canlon B, et al (2012) Emotional exhaustion as a predictor of tinnitus. Psychother Psychosom 81(5):324–326. He´bert S, Fournier P, et al (2013) The auditory sensitivity is increased in tinnitus ears. J Neurosci 33(6):2356–2364. Heller MF, Bergman M (1953) Tinnitus aurium in normally hearing persons. Ann Otol 62:73–83. Hickox AE, Liberman MC (2014) Is noise-induced cochlear neuropathy key to the generation of hyperacusis or tinnitus? J Neurol Neurophysiol 111(3):552–564. Hoare D, Edmondson-Jones M, et al (2014) Amplification with hearing aids for patients with tinnitus and co-existing hearing loss (Review). Cochrane Database Syst Rev(1). Howe SW, Decker NT (1984) Monaural and binaural auditory brainstem responses in relation to the psychophysical loudness growth function. J Acoust Soc Am 76(3):787–792. Hutson KA, Durham D, et al (2008) Consequences of unilateral hearing loss: cortical adjustment to unilateral deprivation. Hear Res 237(1–2):19–31. Hutson KA, Durham D, et al (2009) Reversible conductive hearing loss: restored activity in the central auditory system. Audiol Neurootol 14(2):69–77. Ison JR, Allen PD, et al (2007) Age-related hearing loss in C57BL/6J mice has both frequency-specific and non-frequencyspecific components that produce a hyperacusis-like exaggeration of the acoustic startle reflex. J Assoc Res Otolaryngol 8(4):539–550. Keilmann A, Herdegen T (1997) The c-Fos transcription factor in the auditory pathway of the juvenile rat: effects of acoustic deprivation and repetitive stimulation. Brain Res 753:291–298.

76

P. Fournier et al. / Neuroscience 283 (2014) 64–77

Knipper M, Van Dijk P, Nunes I, Ru¨ttiger L, Zimmermann U (2013) Advances in the neurobiology of hearing disorders: recent developments regarding the basis of tinnitus and hyperacusis. Prog Neurobiol 111:17–33. Knobel K, Sanchez T (2008) Influence of silence and attention on tinnitus perception. JAMA Otolaryngol Head Neck Surg 138(1):18–22. Koch M (1999) The neurobiology of startle. Prog Neurobiol 59:107–128. Kujawa SG, Liberman MC (1999) Long-term sound conditioning enhances cochlear sensitivity. J Neurol Neurophysiol 82(2):863–873. Langers DRM, van Dijk P, et al (2007) FMRI activation in relation to sound intensity and loudness. Neuroimage 35(2):709–718. Liberman MC, Kiang NY (1978) Acoustic trauma in cats. Cochlear pathology and auditory-nerve activity. Acta Otolaryngol 358:1–63. Lu J, Lobarinas E, et al (2011) GABAergic neural activity involved in salicylate-induced auditory cortex gain enhancement. Neurosci 189:187–198. Maslin MRD, Munro KJ, et al (2013) Investigation of cortical and subcortical plasticity following short-term unilateral auditory deprivation in normal hearing adults. NeuroReport 24(6):287–291. McGinn MD, Henry KR (1974) Acute vs chronic acoustic deprivation: effects on auditory evoked potentials and seizure in mice. Dev Psychobiol 8(3):223–232. McGinn MD, Willott JF, et al (1973) Effects of conductive hearing loss on auditory evoked potentials and audiogenic seizures in mice. Nat New Biol 244:255–256. Me´nard M, Galle´go S, et al (2008) Relationship between loudness growth function and auditory steady-state response in normalhearing subjects. Hear Res 235(1–2):105–113. Min J-A, Lee S-H, et al (2012) Clinical characteristics associated with different strengths of loudness dependence of auditory evoked potentials (LDAEP) in major depressive disorder. Psychiatry Res 200(2–3):374–381. Moffat G, Adjout K, et al (2009) Effects of hearing aid fitting on the perceptual characteristics of tinnitus. Hear Res 254(1– 2):82–91. Moore BCJ (2003) The perception of loudness. An introduction to the psychology of hearing. San Diego, Elsevier Science: A. Press. p. 127–161. Munro KJ, Walker AJ, et al (2007) Evidence for adaptive plasticity in elderly monaural hearing aid users. NeuroReport 18(12):1237–1240. Munro KJ, Blount J (2009) Adaptive plasticity in brainstem of adult listeners following earplug-induced deprivation. J Acoust Soc Am 126(2):568. Munro KJ, Merrett JF (2013) Brainstem plasticity and modified loudness following short-term use of hearing aids. J Acoust Soc Am 133(1):343. Munro KJ, Trotter JH (2006) Preliminary evidence of asymmetry in uncomfortable loudness levels after unilateral hearing aid experience: evidence of functional plasticity in the adult auditory system. Int J Audiol 45(12):684–688. Munro KJ, Pisareva NY, et al (2007) Asymmetry in the auditory brainstem response following experience of monaural amplification. Neuroreport 18(17):1871–1974. Niu X, Tahera Y, et al (2007) Environmental enrichment to sound activates dopaminergic pathways in the auditory system. Physiol Behav 92(1–2):34–39. Noren˜a AJ (2011) An integrative model of tinnitus based on a central gain controlling neural sensitivity. Neurosci Biobehav Rev 35(5):1089–1109. Norena AJ, Chery-Croze S (2007) Enriched acoustic environment rescales auditory sensitivity. NeuroReport 18(12):1251–1255. Norena AJ, Eggermont JJ (2005) Enriched acoustic environment after noise trauma reduces hearing loss and prevents cortical map reorganization. J Neurosci 25(3):699–705. Norena AJ, Eggermont JJ (2006) Enriched acoustic environment after noise trauma abolishes neural signs of tinnitus. Neuroreport 17(6):559–563.

Olsen SO (1999) The relationship between the uncomfortable loudness level and the acoustic reflex threshold for pure tones in normally-hearing and impaired listeners – a meta-analysis. Audiology 38:61–68. Olsen SO, Rasmussen AN, et al (1999a) The acoustic reflex threshold: not predictive for loudness perception in normallyhearing listeners. Audiology 38:303–307. Olsen SO, Rasmussen AN, et al (1999b) Loudness perception is influenced by long-term hearing aid use. Audiology 38:202–205. Olsen SO, Rasmussen AN, et al (1999c) The relationship between the acoustic reflex threshold and levels of loudness categories in hearing-impaired listeners. Audiology 38:308–311. O’Neill BV, Croft RJ, et al (2008) The loudness dependence of the auditory evoked potential (LDAEP) as an in vivo biomarker of central serotonergic function in humans: rationale, evaluation and review of findings. Hum Psychopharmacol 23(5):355–370. Park Y-M, Lee S-H (2013) Clinical usefulness of loudness dependence of auditory evoked potentials (LDAEP) in patients with bipolar disorder. Psychiatry Invest 10(3):233. Pasic TR, Rubel EW (1989) Rapid changes in cochlear nucleus cell size following blockade of auditory nerve electrical activity in gerbils. J Comp Physiol 283:474–480. Philibert B, Collet L, et al (2005) The auditory acclimatization effect in sensorineural hearing-impaired listeners: evidence for functional plasticity. Hear Res 205(1–2):131–142. Philips DP (1987) Stimulus intensity and loudness recruitment: neural correlates. J Acoust Soc Am 82(1):1–12. Phillips J, McFerran D (2010) Tinnitus retraining therapy (TRT) for tinnitus (Review). Cochrane Database Syst Rev(3). Picton T, Stappells D, et al (1981) Auditory evoked potentials from the human cochlea and brainstem. J Otholaryngol Suppl. 9:1–41. Picton T, John SM, et al (2003) Human auditory steady-state responses. Int J Audiol 42:177–219. Pienkowski M, Eggermont JJ (2009) Long-term, partially-reversible reorganization of frequency tuning in mature cat primary auditory cortex can be induced by passive exposure to moderate-level sounds. Hear Res 257(1–2):24–40. Pienkowski M, Eggermont JJ (2010a) Intermittent exposure with moderate-level sound impairs central auditory function of mature animals without concomitant hearing loss. Hear Res 261(1–2): 30–35. Pienkowski M, Eggermont JJ (2010b) Passive exposure of adult cats to moderate-level tone pip ensembles differentially decreases AI and AII responsiveness in the exposure frequency range. Hear Res 268(1–2):151–162. Pienkowski M, Eggermont JJ (2011) Cortical tonotopic map plasticity and behavior. Neurosci Biobehav Rev 35(10):2117–2128. Pienkowski M, Eggermont JJ (2012) Reversible long-term changes in auditory processing in mature auditory cortex in the absence of hearing loss induced by passive, moderate-level sound exposure. Ear Hear 33(3):305–314. Pienkowski M, Munguia R, et al (2011) Passive exposure of adult cats to bandlimited tone pip ensembles or noise leads to longterm response suppression in auditory cortex. Hear Res 277(1– 2):117–126. Pienkowski M, Munguia R, et al (2013) Effects of passive, moderatelevel sound exposure on the mature auditory cortex: spectral edges, spectrotemporal density, and real-world noise. Hear Res 296:121–130. Popescu MV, Polley DB (2010) Monaural deprivation disrupts development of binaural selectivity in auditory midbrain and cortex. Neuron 65(5):718–731. Rabinowitz NC, Willmore BDB (2011) Contrast gain control in auditory cortex. Neuron 70(6):1178–1191. Ro¨hl M, Uppenkamp S (2012) Neural coding of sound intensity and loudness in the human auditory system. J Assoc Res Otolaryngol 13(3):369–379. Ro¨hl M, Kollmeier B, et al (2011) Spectral loudness summation takes place in the primary auditory cortex. Hum Brain Mapp 32(9):1483–1496.

P. Fournier et al. / Neuroscience 283 (2014) 64–77 Schadow J, Lenz D, et al (2007) Stimulus intensity affects early sensory processing: sound intensity modulates auditory evoked gamma-band activity in human EEG. Int J Psychophysiol 65(2):152–161. Schaette R, McAlpine D (2011) Tinnitus with a normal audiogram: physiological evidence for hidden hearing loss and computational model. J Neurosci 31(38):13452–13457. Schaette R, Turtle C, et al (2012) Reversible induction of phantom auditory sensations through simulated unilateral hearing loss. Plos One 7(6):e35238. Schaette R, Munro KJ, et al (2014) Plasticity and modified loudness following short-term unilateral deprivation: Evidence of multiple gain mechanisms within the auditory system. J Acoust Soc Am 135(1):315–322. Serpanos YC, O’Mailley H, et al (1997) The relationship between loudness intensity functions of the click-ABR wave V latency. Ear Hear 18(5):409–419. Sie KCY, Rubel EW (1992) Rapid change in protein synthesis and cell size in the cochlear nucleus following eight nerve activity blockade or cochlea ablation. J Comp Neurol 320:501–508. Sigalovsky IS, Melcher JR (2006) Effects of sound level on fMRI activation in human brainstem, thalamic and cortical centers. Hear Res 215(1–2):67–76. Silva I, Epstein M (2010) Estimating loudness growth from tone-burst evoked responses. J Acoust Soc Am 127(6):3629. Sun W, Lu J, et al (2009) Salicylate increases the gain of the central auditory system. Neuroscience 159(1):325–334. Sun W, Manohar S, et al (2011) Early age conductive hearing loss causes audiogenic seizure and hyperacusis behavior. Hear Res 282(1–2):178–183. Sun W, Deng A, et al (2012) Noise exposure enhances auditory cortex responses related to hyperacusis behavior. Brain Res 1485:108–116.

77

Trune DR, Morgan CR (1988) Influence of developmental auditory deprivation on neuronal ultrastructure in the mouse anteroventral cochlear nucleus. Dev Brain Res 42:304–308. Tucker DA, Phillips SL, et al (2005) The effect of silence on tinnitus perception. JAMA Otolaryngol Head Neck Surg 132(1):20–24. Wang H, Yin G, et al (2011) Monaural conductive hearing loss alters the expression of the GluA3 AMPA and glycine receptor a1 subunits in bushy and fusiform cells of the cochlear nucleus. Neuroscience 199:438–451. Watkins PV, Barbour DL (2008) Specialized neuronal adaptation for preserving input sensitivity. Nat Neurosci 11(11):1259–1261. Webster DB (1983) Late onset auditory deprivation does not affect brainstem auditory neuron soma size. Hear Res 12:145–147. Wen B, Wang GI, et al (2009) Dynamic range adaptation to sound level statistics in the auditory nerve. J Neurosci 29(44):13797–13808. Whiting B, Moiseff A, et al (2009) Cochlear nucleus neurons redistribute synaptic AMPA and glycine receptors in response to monaural conductive hearing loss. Neuroscience 163(4):1264–1276. Wutzler A, Winter C, et al (2008) Loudness dependence of auditory evoked potentials as indicator of central serotonergic neurotransmission: simultaneous electrophysiological recordings and in vivo microdialysis in the rat primary auditory cortex. Neuropsychopharmacology 33(13):3176–3181. Wyss C, Hitz K, et al (2013) The loudness dependence of auditory evoked potentials (LDAEP) as an Indicator of serotonergic dysfunction in patients with predominant schizophrenic negative symptoms. Plos One 8(7):e68650. Zeng F-G (2013) An active loudness model suggesting tinnitus as increased central noise and hyperacusis as increased nonlinear gain. Hear Res 295:172–179. Zenker Castro F, Juan Barajas de Prat J, et al (2008) Loudness and auditory steady-state responses in normal-hearing subjects. Int J Audiol 47(5):269–275.

(Accepted 7 August 2014) (Available online 15 August 2014)