tinnitus, commonly referred to as "ringing in the ears," is a common problem affecting many people

around the world. According to studies performed in various countries, tinnitus affects 10 to 20
percent of the general population (McFadden, 1982; Coles, 1987; Drukier, 1989); in the United
States, this translates into 25 to 52 million Americans. This affliction is even more prevalent in the
elderly over the age of 65 years, with approximately 30 percent reporting tinnitus (Sataloff et al,
1987; Salomon, 1989). For about 5 percent of the general population (about 13 million Americans),
prolonged tinnitus is moderately or significantly annoying, causing them to seek help (McFadden,
1982). Consequently, this population is labeled as having clinically significant tinnitus. Finally, 1 out
of 100 adults reports tinnitus as a debilitating problem (Coles, 1996) (about 2.6 million Americans).
Typically, tinnitus is associated with hearing loss, otosclerosis, ear infections, acoustic neuroma,
Meniere's syndrome, and aging. Also, more than 200 prescription and nonprescription drugs list
tinnitus as a potential side effect. Much less is known about the prevalence of increased sensitivity
to sound, hyperacusis. Currently, in the literature there are limited data published by Vernon
(Vernon, 1987; Vernon and Press, 1998), Coles (Coles and Sood, 1988), Hazell and Sheldrake (Hazell
and Sheldrake, 1992), and Jastreboff' (Jastreboff et al, 1996b, 1998). Data presented by Vernon are
incongruent with the remaining reports, stating that only 0.3 percent of tinnitus patients have
hyperacusis, whereas other data indicate that about 40 percent of tinnitus patients have some
degree of hyperacusis. Furthermore, our data indicate that about 25 percent of tinnitus patients are
bothered more by their hyperacusis than their tinnitus, and thus require specific treatment for
hyperacusis. We are not aware of epidemiologic data related to the prevalence of hyperacusis in the
general population. This is an unfortunate situation, since these data could help in estimating the
need for health services and for planning cost-effective, yet quality, services. Assuming, however,
that in our practice we are working with patients with clinically significant tinnitus (5% of the general
population), and 25 percent of those have significant hyperacusis, then about 1.25 percent of the
general population (3.25 million Americans) has significant hyperacusis. This is a rather conservative
estimate, as there are cases of hyperacusis without tinnitus. In spite of a long recorded history of
tinnitus, reaching as far back as the ancient Babylonian and Egyptian civilizations (Feldmann, 1988),
and its high prevalence today, there is no cure for tinnitus. It appears that all approaches used in the
past failed to provide systematic relief to tinnitus patients. Practically all of the treatments
previously used were effective only on a subpopulation of patients, had to be continued through the
patient's life, and were frequently accompanied by significant side effects. Furthermore, the very
existence of a long list of treatments that may potentially provide help, and the fact that the single
most common approach is telling patients "to learn to live with it," argues strongly against their
effectiveness. In this paper, we propose that Tinnitus Retraining Therapy (TRT), when implemented
properly, (1) is highly effective, (2) does not have side effects, (3) needs to be implemented over a
finite amount of time, and (4) can be used on all patients.

Definitions As there are various definitions of tinnitus, hyperacusis, and phonophobia, we are
presenting the definitions, as proposed by us, to ensure a clearer understanding. nnnitus is
commonly defined as a noise in the ears or head, frequently described as ringing, buzzing, humming,
hissing, the sound of escaping steam, etc. In 1982, the Committee on Hearing, Bioacoustics and
Biomechanics proposed a definition of tinnitus as "the conscious experience of a sound that
originates in the head of its owner" (McFadden, 1982). The definition of tinnitus we are promoting is
"the perception of a sound which results exclusively from the activity within the nervous system
without any corresponding mechanical, vibratory activity within the cochlea" (Jastreboff, 1995), that
is, tinnitus as an auditory phantom perception (Jastreboff, 1990, 1995). Somatosounds are sounds
generated by structures in and adjacent to the ear, including spontaneous otoacoustic emission. The
term "objective tinnitus" has been used to describe somatosounds. This classification is inaccurate
because it is dependent on the equipment used and the skill of the observer, not the
pathophysiology of the sound. Hyperacusis is defined as abnormally strong reactions occurring
within the auditory pathways resulting from exposure to moderate sound; as a consequence,
patients express reduced tolerance to suprathreshold sounds. This phenomenon may be, but
typically is not, related to recruitment (Moore, 1995; Jastreboff, 1998; Jastreboff et al, 1998).
Phonophobia is defined as abnormally strong reactions of the autonomic and limbic systems
(without abnormally high activation of the auditory system by sound), resulting from enhanced
connections between the auditory and limbic systems. This can be described at the behavioral level
as "patients being afraid of sound." Increased sound sensitivity is abnormally high sensitivity to a
sound resulting from the sum effects of hyperacusis and phonophobia. The seemingly simple
definition of tinnitus has profound implications on proposing the mechanisms of tinnitus and
consequently on its treatment. This definition stresses the involvement of the nervous system as a
key component responsible for the emergence of tinnitus and problems arising from its presence,
thus moving its mechanisms away from the cochlea to the central nervous system. The definition
further indicates the existence of a link between the mechanisms of tinnitus and that of the
phantom limb and phantom pain phenomena, which indeed appear to exist. Certain common
aspects of tinnitus and phantom pain are used for the classification of tinnitus and hyperacusis
patients and their treatment.

OUTLINE OF THE NEUROPHYSIOLOGIC MODEL Proposed in the 1980s (Jastreboff, 1990),

Ha 7-8

related neuronal activity inappropriately activates the limbic and autonomic nervous systems,
resulting in the development of annoyance, anxiety, sleep disturbances, and a number of somatic
problems. Connections between the auditory and other systems are based upon the principle of
conditioned reflexes and as such cannot be easily and directly changed. Furthermore, the limbic and
autonomic nervous systems may be completely normal, and the problem arises from their activation
by inappropriate stimulus (i.e., tinnitus-related neuronal activity). Hyperacusis and phonophobia
activate the limbic and autonomic nervous systems as well, but with different mechanisms than
tinnitus. Once established, reactions of these systems are controlled by the conditioned reflex
principle but triggered by external sounds rather than tinnitus-related neuronal activity. How to
Treat Tinnitus and Hyperacusis The neurophysiologic model offers an approach to treat both tinnitus
and hyperacusis. For tinnitus, the approach is based on observations that, although there is no
reliable method for attenuating the tinnitus source (cure), the brain exhibits a high level of plasticity.
It is possible to habituate to any sensory signal, as long as the signal is not associated with any
negative implications. Thus, the conclusion is to induce habituation of tinnitus (by interfering with
tinnitus-related neuronal activity above the tinnitus source). Habituation is a normal, common, and
necessary function of the brain, as pointed out by Konorski in 1967 (Konorski, 1967) following the
original postulate from Pavlov in 1928. Its necessity results from the fact that although the brain can
detect very weak sound patterns if significant (our name called out in a noisy room, cry of our baby,
our language), it cannot handle more than one conscious task at any given time (inability to read a
book and write a letter; to understand someone talking while reading; to listen to two people talking
at the same time). The question is how do we manage the huge amount of sensory stimulations that
bombard us all of the time? For example, how are we able to drive a car? The solution to this
problem performed by the brain is to (1) select and block all unimportant stimuli from reaching our
awareness at the subconscious level and block reactions that these stimuli would otherwise evoke
(habituate); (2) automize as many tasks as possible into the subconscious, nonverbal reflexes
(driving a car, eye movements when reading, walking, running, etc.); (3) prioritize all remaining
tasks; and (4) perform one task at a time, starting from the most important. Note that the selection
and blockage of unimportant signals have to occur at a subconscious level on the basis of past
experiences. The selection process cannot be done on a conscious level, as it would consume our
attentional abilities and nullify the purpose of selection. On the basis of the neurophysiologic model,
it has been proposed to treat tinnitus by inducing its habituation. To achieve habituation, a specific
approach based on the general principles of the brain function (e.g., the physiologic mechanisms of
perception, the role of subcortical auditory pathways, the functional properties of the limbic and
autonomic nervous systems and their interaction with the auditory system, mechanisms of
conditioned reflexes, and the physiologic mechanisms of the brain plasticity) was suggested
(Jastreboff, 1990, 1995; Jastreboff et al, 1996a). Use of habituation to help tinnitus patients was first
proposed by Hallam et al (1984), but their model was psychological rather than physiologic.
Consequently, their approach to induce and sustain habituation was different, focused on
reassurance, relaxation, and attention distraction (Jakes et al, 1986), and did not appear to produce
significant, sustained improvement in tinnitus patients. Habituation of tinnitus has two main goals.
The primary clinical goal is to habituate reactions of the limbic and autonomic nervous systems (Fig.
3A). Tinnitus-related neuronal activity is blocked from reaching both systems. Consequently, patients
who achieve full habituation of their reactions do not experience annoyance, anxiety, or any other
negative reactions of their brain and the body. Note that the patients still perceive their tinnitus and,
in case of exclusive habituation of reactions, the proportion of the time when they are aware of
tinnitus is still the same. Nevertheless, since these patients no longer are bothered by their tinnitus,
even when perceived, tinnitus ceases to be an issue. The secondary goal is to achieve a habituation
of the perception (Fig. 3B). In this case, tinnitus-related neuronal activity is blocked before it reaches
the level of awareness, and patients are unaware of the presence of tinnitus. Note that even when a
very high level of habituation of the reaction and perception is achieved, it is still not a cure for
tinnitus, as patients can hear their tinnitus anytime when their attention is focused on it; also, the
tinnitus pitch and loudness are the same as at the beginning of the treatment. As predicted by the
model, tinnitus maskability may decrease a few decibels (Jastreboff et al, 1994), but the change is
too small to have a practical significance for patients. Although some patients achieve a high level of
tinnitus habituation, to be of clinical significance, habituation does not have to be complete, and
patients can experience significant improvement even with partial habituation of reactions and
perception. The final goal of the treatment is that tinnitus ceases to have an impact on the patient's
life. In conclusion, Tinnitus Habituation Therapy (THT), aimed at inducing tinnitus habituation, has
been proposed as a treatment for tinnitus on the basis of the neurophysiologic model of tinnitus
(Jastreboff, 1990). Since the habituation-related modifications of the nervous system occur above
the source of tinnitus, the etiology of tinnitus is irrelevant, and any type of tinnitus, as well as
somatosounds, can be treated by inducing habituation.

TINNITUS RETRAINING THERAPY Outline of the Treatment Habituation can be achieved or facilitated
by a number of approaches, including counseling, combined, for example, with medications,
biofeedback, hypnosis, etc. However, the easiest implementation of THT involves a parallel use of
retraining counseling with sound therapy . This therapy has become popular under the name of
Tinnitus Retraining Therapy (TRT). These two components perform different functions and both have
to be used for a method to be called TRT. Retraining counseling acts to decrease the level of
stimulation from the cortical areas of the brain to the limbic and autonomic nervous systems and to
decrease the general level of activity within these two systems. During counseling, a number of
points are presented to the patient: (1) the perception of tinnitus results from a compensation
occurring within the auditory system, (2) tinnitus is a problem because of the activation of emotional
(limbic) and autonomic nervous systems, and (3) by using the plasticity of the nervous system, it is
possible to retrain the brain to achieve habituation of tinnitus-induced reactions and tinnitus
perception. Once the patient accepts these ideas as realistic and "making sense" to them, this puts
tinnitus into the category of neutral stimuli, to which they may gradually habituate. The crucial point
to recognize, for both the therapist and the patient, is that tinnitusinduced reactions are governed
by the conditioned reflex principle. Consequently, the conscious realization of the benign nature of
tinnitus is not sufficient to remove these reactions, and a significant amount of time is needed for
their gradual extinction. Sound therapy provides significant help in the process of habituation by
decreasing the strength of tinnitus-related neuronal activity within the auditory system and from the
auditory system to the limbic and autonomic nervous systems. Therefore, the strength of
tinnitusrelated activation within all loops presented in Figure 1 decreases, making habituation of
tinnitus easier. Specifically, sound therapy acts by providing the auditory system with constant,
lowlevel, neutral auditory signals to (1) decrease the contrast between tinnitus-related neuronal
activity and background neuronal activity, (2) interfere with the detection of the tinnitus signal, and
(3) decrease enhanced gain within the auditory pathways.