Inferior-Colliculus Responses to Amplitude-Modulated and Unmodulated Acoustic Tones and Cochlear-Implant Pulse Trains
- Author(s): Schoenecker, Matthew Charles
- Advisor(s): Leake, Patricia A
- et al.
Cochlear implants (CIs) are neural prostheses that currently provide acoustic sensation to more than 120,000 profoundly hearing-impaired people throughout the world. The majority of these CI users are able to understand speech without lip reading and to converse over the telephone. The most fortunate among them can even perform and appreciate music. Unfortunately, however, many CI recipients receive much less benefit from their devices. In order to examine the neuronal bases for these disparate outcomes, a recording system was developed to overcome specific technical limitations in previous studies that recorded neuronal responses to cochlear implant stimulation in animal models. This recording system was then used in two studies comparing responses to normal acoustic stimulation and electrical cochlear implant stimulation in guinea pigs and cats.
The design of the recording system included a 32-channel recording amplifier and a software technique for removing stimulus artifacts from recordings of neural responses to high-rate electrical stimulation. Contemporary cochlear implants often deliver current pulse trains with carrier rates of 1000 pulses/s or higher. In neurophysiology studies, this electrical stimulation produces artifacts that are typically much larger than neuronal responses. Therefore, the recording system was specifically designed to record neuronal responses and cancel these electrical stimulus artifacts. When biphasic full-scale-input pulses (1.5-V) are applied directly to the amplifier inputs, each recording channel settles to 20 micro-volts in less than 80 microseconds. This fast recovery makes it likely that the recording electrode-electrolyte interface, not the recording electronics, will limit artifact settling times. Artifacts are blanked in software, allowing flexibility in the choice of blanking period and the possibility of recovering neural data occurring simultaneously with non-saturating artifacts. The system has been used in-vivo to record central neuronal responses to intracochlear electrical stimulation at rates up to 2000 pulses/s.
Using this recording system, systematic and quantitative comparisons of inferior-colliculus responses to acoustic stimulation and electrical stimulation in two configurations (monopolar and bipolar) were carried in guinea pigs and cats. Previous cochlear implant studies using isolated electrical stimulus pulses in animal models have reported that monopolar stimulus configurations elicit broad extents of neuronal activation within the central auditory system--much broader than the activation patterns produced by bipolar electrode pairs or acoustic tones. However, psychophysical and speech reception studies that use sustained pulse trains do not show clear performance differences between monopolar and bipolar configurations. To evaluate whether monopolar intracochlear stimulation can produce selective excitation of the inferior colliculus, activation widths were determined along the tonotopic axis of the inferior colliculus for acoustic tones and 1000-pulse/s electrical pulse trains in guinea pigs and cats. Electrical pulse trains were presented using an array of 6-12 stimulating electrodes distributed longitudinally along a space-filling silicone carrier positioned in the scala tympani of the cochlea. The data indicated that for monopolar, bipolar, and acoustic stimuli, activation widths were significantly narrower for sustained responses than for the transient response to the stimulus onset. Furthermore, monopolar and bipolar stimuli elicited similar activation widths when compared at stimulus levels that produced similar peak spike rates. Surprisingly, monopolar and bipolar stimuli produced narrower sustained activation than 60 dB SPL acoustic tones when compared at stimulus levels that produced similar peak spike rates. Therefore, the conclusion from these experiments was that intracochlear electrical stimulation using monopolar pulse trains can produce activation patterns that are at least as selective as bipolar or acoustic stimulation, if stimulus intensities are appropriately matched.
The second study compared responses to acoustic and monopolar electrical stimuli that were sinusoidally amplitude modulated (SAM), in order to better model the complex signals delivered by CI processors. For both normal hearing listeners and cochlear implant users, SAM signals produce psychophysical interactions that can extend across large differences in carrier frequency or large intracochlear electrode separations. However, the neural correlates of these phenomena are not well understood. This study was designed to determine whether SAM stimuli elicit activation across a broader extent of the frequency axis of the inferior colliculus than unmodulated steady-state stimuli, and whether this activation is strongly phase locked to the SAM stimulus envelope. To address these questions neuronal activity in the inferior colliculus of guinea pigs, normal cats and chronically deafened cats was recorded in response to acoustic and electrical stimulation. Quantitative analysis of recordings indicated that the extent of inferior colliculus activation was up to 70% broader for SAM stimuli than for unmodulated steady-state stimuli in normal cats and guinea pigs and 160% broader in chronically deafened cats. This activity was also phase-locked to the SAM envelope across a broad extent of the frequency axis of the inferior colliculus. These results suggest that a number of cross-carrier frequency interactions for SAM stimuli could occur at the level of the inferior colliculus. They also show that direct comparisons of responses to acoustic and electrical SAM stimuli can reveal attributes of neural processing that underlie specific psychophysical findings in CI recipients--and thereby can provide a powerful basis for guiding development of new processing strategies for future cochlear implants.