Research paper Effects of hearing aid fitting on the perceptual characteristics of tinnitus G. Moffat a , K. Adjout b,d,e , S. Gallego c , H. Thai-Van b,c,d,e , L. Collet b,c,d,e , A.J. Noreña a, * a Laboratoire de neurobiologie intégrative et adaptative, UMR CNRS 6149, 3 Place Victor Hugo, Marseille, F-13331 Cedex 03, France b Université de Lyon, Lyon F-69003, France c Hospices Civils de Lyon, service d’audiologie et d’explorations orofaciales, hôpital Edouard Herriot, Lyon, F-69003, France d CNRS, UMR5020, Neurosciences sensorielles, comportement, cognition, Lyon, F-69007, France e Institut Fédératif des Neurosciences de Lyon, Lyon-Bron, F-69677, France article info Article history: Received 11 December 2008 Received in revised form 18 April 2009 Accepted 20 April 2009 Available online 3 May 2009 Keywords: Tinnitus Hearing aid Tinnitus pitch Neural plasticity Deafferentation abstract Restoration of auditory input through the use of hearing aids has been proposed as a potentially impor- tant means of altering tinnitus among those tinnitus sufferers who experience significant sensorineural hearing loss. In animal models of neural plasticity induced by noise trauma, high-frequency stimulation in deafferented regions of the auditory spectrum has been shown to modulate cortical reorganization after hearing loss, a result which suggests that the neural basis of tinnitus is subject to interference by acoustic stimulation. This study drew on deafferentation models to investigate the effect of hearing aids on the psychoacoustic properties of the tinnitus sensation, using both conventional amplification and high-bandwidth amplification regimes. The tinnitus percept was affected only weakly in the conventional amplification group, and was not at all affected in the high-bandwidth group. The changes observed under conventional, low-to-medium frequency amplification may indicate that the perceptual character- istics of tinnitus depend on the pattern of sensory inputs – notably a contrast in activity between adjacent central auditory regions of more and less afferent activity – while the absence of modifications in the high-bandwidth amplification group suggests limit on the tractability of the tinnitus percept. This limit to the malleability of the tinnitus percept may arise from either the extent of hearing deficits or the dura- tion and robustness of the neuroplastic changes that originally give rise to tinnitus. Ó 2009 Elsevier B.V. All rights reserved. 1. Introduction The phantom auditory perception of tinnitus is sufficiently pre- valent in the general population (Pilgramm et al., 1999; Fabijanska et al., 1999) and can so severely impair quality of life that the development of effective therapies is imperative. A prerequisite for the development of such therapies is an understanding of the neurophysiological mechanisms this disorder. These mechanisms have been partially addressed in animals using treatments known to induce tinnitus in humans, such as noise trauma (Kaltenbach et al., 2005; Kaltenbach, 2006; Eggermont, 2005; Noreña and Eggermont, 2003, 2006) and salicy- late injection (Ochi and Eggermont, 1996, 1997; Eggermont and Kenmochi, 1998; Guitton et al., 2003) among others. Although much of the neural substrate of tinnitus remains poorly under- stood, a general consensus has arisen around a ‘‘central tinnitus” model that could account for the most common cases of subjective tinnitus. This model postulates that tinnitus is related to central neuroplastic changes arising due to a sensory deafferentation (Rauschecker, 1999; Norena et al., 2002; Noreña and Eggermont, 2003, 2006; Eggermont and Roberts, 2004; Eggermont, 2005). The current model is supported by the fact that most tinnitus sufferers exhibit hearing loss (Sirimanna et al., 1996; Pilgramm et al., 1999) and it is well known that hearing loss induces dramatic central changes (Robertson and Irvine, 1989; Rajan et al., 1993; Harrison et al., 1991,1993; Eggermont and Komiya, 2000; Noreña and Eggermont, 2005; Milbrandt et al., 2000; Mossop et al., 2000; Kaltenbach et al., 2005; Kaltenbach, 2006). Sensory deaffer- entation in the form of cochlear lesions or noise-induced excitotox- icity in auditory nerve fibers (Pujol and Puel, 1999), for example, is generally associated with a decrease in spontaneous and evoked firing in the auditory nerve (Liberman and Dodds, 1984; Wang et al., 1997). Hearing loss, then, can be considered as a decrease in sensory inputs to central auditory structures. One important consequence of this decrease in afferent input is a concurrent decrease in central inhibition (Suneja et al 1998; Milbrandt et al., 2000; Argence et al., 2006). Release from central inhibition has been proposed as a mechanism by which new responses arise in central auditory structures after a noise-induced hearing loss; excitatory inputs that are initially sub-threshold under afferent-dependent inhibition become supra-threshold with its decrease, a phenomenon referred to as an ‘‘unmasking” of 0378-5955/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.heares.2009.04.016 * Corresponding author. E-mail address: arnaud.norena@univ-provence.fr (A.J. Noreña). Hearing Research 254 (2009) 82–91 Contents lists available at ScienceDirect Hearing Research journal homepage: www.elsevier.com/locate/heares