Page 40 - Winter 2020

PSYCHOACOUSTICS OF TINNITUS
Loss of Tuning
One of the earliest proposed theories of tinnitus initia- tion was the discordant damage theory. According to this theory (an extension of theories proposed by Tonndorf 1981a,b), the outer hair cells (OHCs) of the mammalian cochlea are more prone to damage than the inner hair cells (IHCs), resulting in imbalanced activity via type I and type II afferent fibers that, respectively, carry signals from the ear to the dorsal cochlear nucleus (DCN), the first audi- tory center in the brain. The alteration of input to the DCN results in loss of inhibition and compensatory mechanisms at more central sites, including bursting neural activity, mapping reorganization, decreased inhibition, and central gain mentioned in Tinnitus: What Is It?.
Kaltenbach and Afman (2000) showed that significant IHC damage can prevent the onset of hyperactivity in the DCN. Tonndorf ’s (1981) original model suggested a decoupling of stereocilia (the hair-like projections from the cell) between the OHCs and the tectorial membrane (a membrane floating above the hair cells) that leads to loss of energy and increased noise at the level of the hair cell underlying tinnitus generation. Tonndorf ’s follow-up theory (1987) suggested that tinnitus was equivalent to chronic pain in the somatosensory system and a result of preferential damage to the OHCs and established an analogy of tinnitus to chronic pain.
In contrast to the discordant damage theory, cochlear insults that commonly lead to chronic tinnitus in humans have been found to produce a long-term decrease in the auditory neuronal spontaneous activity (Liberman and Dodds, 1984). Tinnitus is strongly correlated with HL and cochlear damage as a result of ototoxicity or noise exposure. Specifically, IHC/synaptic loss has been specu- lated to produce tinnitus.
To explore this relationship, a behavioral gap detection task was used to determine the presence of tinnitus in a chinchilla model with selective IHC loss following adminis- tration of carboplatin. Carboplatin is an ototoxic anticancer drug known to cause significant IHC loss (>80% loss) while leaving OHCs largely intact (<5% loss) in the chinchilla, an effect unique to the chinchilla model (Lobarinas et al., 2013b). Preliminary data showed overall poorer gap detec- tion performance when tested at lower presentation levels, but the findings were not frequency specific. The absence of frequency-specific deficits suggested that these animals did
not perceive tinnitus even with severe IHC loss. Thus, IHC damage alone does not seem sufficient to generate tinnitus and support the discordant dysfunction theory of tinnitus or a combination of OHC and IHC/synapse injury at play.
Changes to psychophysical tuning curves may offer insight into differentiating OHC vs. IHC/synaptic contri- butions to the onset of tinnitus but are currently limited to humans in regard to tinnitus effects. A psychophysical tuning curve is a method that can be used to generate comparable data to the physiological frequency threshold curve for a single auditory nerve fiber. A narrowband noise of variable center frequency is used as a masker, and a fixed frequency and fixed-level pure tone at about 20 dB HL is commonly the target. The level of masker is found that just masks the tone for different masker fre- quencies. With OHC damage, the tuning curve becomes flattened and less sharp due to loss of sensitivity.
For example, Tan et al. (2013) examined psychophysical tuning curves in persons with HL and tinnitus and in per- sons with HL and no tinnitus. Both groups were compared with a reference group of persons with normal hearing.
The normal-hearing group showed expected patterns of low thresholds and sharp tuning curves; these patterns are thought to reflect the nonlinearity of the OHCs. Interest- ingly, the HL group with tinnitus showed better thresholds, greater residual compression, and better tuning than the no-tinnitus group in the midfrequency range. This was likely reflective of the greater high-frequency HL of the tinnitus group relative to the no-tinnitus group that had a wider array of patterns. Thus, the finding could simply reflect differences in hearing thresholds; however, after matching participants based on HL, the pattern persisted. Tan et al. suggested that the findings may be explained by the tinnitus group having residual OHC function and a preferential loss of IHCs or afferents.
The difference in the animal model of widespread loss of IHCs and lack of tinnitus evidence compared with psy- choacoustic tuning curves in humans implicating IHCs/ synapse may also be explained by the discordant damage theory. The carboplatin model creates a pure loss of IHCs/synapses without damage to OHCs. Still, humans may still have some level of damage to their OHCs not reflected in their tuning curves. In other words, it would be parsimonious to suggest that there is likely a ratio of damage to both hair cell types involved and necessary
40 Acoustics Today • Spring 2021