Stimulus level effects on neural excitation and eCAP amplitude.

Abstract The common assumption that the electrically evoked compound action potential (eCAP) has a linear relationship with the number of excited nerve fibres is derived from the acoustical unitary response concept. This study tests the validity of this hypothesis for electrical stimulation. Five guinea pigs were implanted with the tip of a human HiFocus electrode. eCAPs were measured with the forward masking paradigm, using anodic- and cathodic-leading biphasic current pulses and the inter-pulse interval was varied. Masker and probe amplitudes were varied either individually or simultaneously. Surprisingly, at high levels decreasing eCAP amplitudes were measured with increasing stimulus current. In search for an explanation, the experimental conditions were implemented in our 3D computational model of the implanted guinea pig cochlea to perform a functional comparison. In the final experiment, with fixed inter-pulse interval (IPI) and anodic-leading pulses, increasing stimulus currents showed growing numbers of excited nerve fibres and decreasing eCAP amplitudes at high levels, again. While simulating the relative contribution of single fibres to the overall eCAP, an explanation for this could be found in a waveform change in the modelled single fibre action potentials at high levels. We conclude that highly stimulated nerve fibres have another contribution to the eCAP response than lower stimulated fibres, which leads to a reduction of the eCAP amplitude at high levels.