Neuronal responses in cat primary auditory cortex to electrical cochlear stimulation. II. Repetition rate coding.

Abstract

1. Responses of neurons in primary auditory cortex (AI) of the barbiturate-anesthetized adult cat were studied using cochlear stimulation with electrical and acoustic stimuli. Neuronal responses to acoustic stimulation with brief biphasic clicks of the ear ipsilateral to the studied cortical hemisphere were compared with those evoked by electrical stimulation of the contralateral cochlea with brief biphasic electrical pulses delivered via a feline cochlear prosthesis. The contralateral ear was deafened immediately before implantation of the cochlear prosthesis. The feline cochlear prosthesis consisted of four bipolar electrode pairs and was placed in the scala tympani. Two bipolar electrode conditions were used for stimulation: one near radial pair with electrode spacing of 0.25-0.5 mm, and one longitudinal pair with electrode spacing of approximately 6 mm. 2. The firing rates obtained from single- and multiple-neuron recordings were measured as a function of stimulus repetition rate of electrical and acoustic pulses. From period histograms over a recording interval of 1,000 ms, the driven firing rate to repetition rates from 2 to 38 Hz was obtained and repetition rate transfer functions (RRTFs) were constructed. The RRTFs were characterized as low-pass or band-pass filters and several descriptors were obtained, such as the repetition rate producing the highest driven activity, high and low cutoff frequencies 6 dB below maximum firing rate, and maximum firing rate. 3. For a given neuron, the main characteristics of cortical RRTFs obtained with electrical and acoustic cochlear stimulation were quite similar. However, some small but statistically significant differences in the best repetition rate, cutoff frequencies, and maximum firing rate could be observed between the different stimulation modes. The proportion of band-pass RRTFs was larger for electrical stimulation (57%) than for acoustic stimulation (41%). The high cutoff frequencies for electrical stimulation were slightly but consistently higher than for acoustic RRTFs of the same neuron and the maximum firing rate for electrical stimulation was significantly higher than that evoked by ipsilateral acoustic stimulation. 4. The entrainment of cortical neurons to electrical and acoustic pulses was determined and entrainment profiles were constructed. For a given neuron, electrical entrainment profiles showed higher cutoff frequencies than with acoustic stimulation when judged with a fixed entrainment criterion of 0.25 spikes per event. The maximum entrainment seen for electrical stimulation was approximately 20% higher than seen for the same neuron with acoustic stimulation. 5. Correlation analysis of repetition coding and latency parameters revealed several relationships between these response aspects. Most prominent among them was a significant correlation between measures of the response latency and estimates of the ability to follow temporal repetitions for acoustic as well as electrical conditions. 6. Parametric and comparative evaluations of cortical responses to acoustic and electrical cochlear stimulation support the conclusion that the temporal resolution seen in cortical neurons is largely a consequence of central processing mechanisms based on cell and circuit properties and to a lesser degree a consequence of particular spatial and temporal peripheral excitation patterns. The slightly higher temporal resolution found for the electrical stimulation modes suggests that the temporally highly coherent electrical stimulation appears to engage, in a more effective manner, the excitatory/inhibitory mechanisms contributing to the response in AI than acoustic click stimulation with less temporal coherence. (ABSTRACT TRUNCATED)