Author:

Abstract:

Worldwide, Cochlear Implants (CIs) have already partially restored hearing to over 219000 profoundly hearing-impaired people. CIs attempt to stimulate the survival Auditory Nerve (AN) fibers directly by means of electrical pulses. Electrical stimulation is provided via an array of electrodes implanted inside the cochlea, which directly activates the AN fibers by means of biphasic Symmetric (SYM) pulses, i.e. an initial phase followed by a second phase with the same amplitude but with opposite polarity. Because electrical stimulation is delivered extracellularly, AN fibers could become depolarized and hyperpolarized at several neural sites and the excitation may be substantially different depending on AN status, electrode-neuron interface, and polarity. The necessity of charge balanced stimulation has precluded the use of monophasic stimuli (pulses consisting only in one phase) to study polarity effects in human CI users and most of the research concerning polarity effectiveness has been investigated in animal and computational models. However, recently, asymmetric charge-balanced pulse shapes and CI devices with recording capabilities have allowed researchers to study polarity effectiveness on CI users either by means of Electrically evoked Compound Action Potential (ECAP) or by means of behavioral measurements. Contrary to the majority of the results obtained with animals and computational models, those studies suggest that the positive (anodic) current stimulates the AN more effectively than the cathodic one in human CI users. The studies presented in this document aim to examine how neurons at different sites are affected by different stimulus shapes as well as the potential benefits of combining non-standard pulse shapes by means of objective ECAP and Electrically evoked Auditory Brainstem Response (EABR) as well as behavioral experiments.In chapter 2 we study whether the high sensitivity to the anodic current observed formerly for an electrode in the middle of the array, can be generalized to stimulation at the base and apex of the CI array. Threeexperiments studied the effect of stimulus polarity on the ECAP obtained with the masker-probe paradigm on different sites along the cochlea in CI users. The results of all the experiments extend those of a previous study, showing that the AN in humans is preferentially activated by anodic stimulation, to different sites along the cochlea by means of Monopolar (MP) stimulation.In chapter 3 we present a quantitative analysis of ECAP responses in order to determine an objective detection criterion based on the statistical properties of the noise. The study quantitatively investigates the properties of the Background Noise (BN) and the remaining post average Residual Noise (RN) in ECAP responses recorded from human CI users with the Across-sweeps Noise Estimator (ASNE) method. The analysis of the statistical properties of the noise indicated that a Signal-to-Noise Ratio (SNR) of 1.7 dB as a detection criterion corresponds to a false positive detection rate of 1% with the used measurement setup. In chapter 4 we study whether the high sensitivity to the anodic current is also observed in Bipolar (BP) stimulation mode. Spreads of Excitation (SOEs) produced by different masking pulse shapes presented in MP and BP mode are examined by means of the ECAP, obtained using the forward-masked paradigm in human CI users. The SOEs obtained with the MP masker showed a main excitation peak close to the masker electrode, whereas SYM maskers produced bimodal excitation patterns showing two peaks close to the electrodes of the masker channel. The SOEs obtained using Symmetric with a long Inter-Phase Gap (SYM-IPG) (for which the masker’s second phase is responsible for most of the masking) and Pseudomonophasic (PS) maskers were more pronounced close to the masker electrode for which the effective phase was anodic. These results showed that the anodic polarity is the most effective one in BP mode and that the bimodal patterns produced by SYM maskers could be partially reduced by using asymmetric pulses.Chapter 5 investigates polarity effects at the level of the brainstem. Previous behavioral studies showing that the human AN of CI users is mainly excited by the anodic polarity have only been obtained with asymmetric PS, and it was assumed that only the short high-amplitude phase was responsible for the excitation. Similarly, in our previous studies it has been shown that ECAPs could only be obtained in response to the anodic phases of asymmetric pulses. However, it is possible that ECAP responses to cathodic responses were not detected due to the characteristics of the effective place of excitation across and along the AN fibers. In this chapter we measured EABRs to SYM and asymmetric pulse shapes presented for both polarities. It is shown that responses were time-locked to the short high-amplitude phase of asymmetric pulses, and were smaller, but still measurable, when that phase was cathodicthan when it was anodic. A behavioral experiment studied the polarity sensitivity at different intensities by means of a loudness balancing task between asymmetric stimuli of opposite polarity. The results showed that cathodic stimuli required higher amplitudes than anodic stimuli to reach the same loudness at different stimulus levels. This study provides the first evidence that cathodic stimulation can excite the auditory system of human CI listeners, and confirms that this stimulation is nevertheless less effective than for the anodic polarity.