Final Report Summary - NCAN (Neurophysiological correlates of the anatomically discrete organization of the auditory nerve)
Although the auditory nerve (AN) is one of the best studied sensory nerves, a neurophysiological correlate of its discrete anatomical organisation remains to be elucidated. The host laboratory introduced a new correlation analysis to enable the quantification of temporally locked responses to stochastic stimuli. The analysis has been used extensively in the AN and cochlear nucleus, and allows study of temporal relationships between neurons. The advantage of the correlograms obtained with this analysis is that they reflect differences between two fibres in all three domains where differences are expected (magnitude spectrum, phase spectrum, overall delay), and thereby optimise chances of observing a discrete organisation.
Initial cross-correlation analyses of temporal patterns of the neural responses to broadband noise suggest that the magnitude of correlation jumps indicative of discrete organisation is smaller or comparable to the inherent stochastic temporal jittering in the neural responses of the AN. Thus, it might be impossible to demonstrate quantal organisation with the current methods. On the other side, the spatial profile and extent over which correlated responses to broadband noise are found in the AN and its dependency on stimulus intensity have not yet been determined. Namely, cochlear peripheral filtering can produce correlated responses in fibres originating from nearby cochlear positions even to broadband noise. It is a key issue to assess the spatial profiles of across-fibre temporal correlations.
Recordings from single AN fibres were performed in acute experiments on anesthetised cats. The AN was visualised through a posterior fossa craniotomy. High impedance (40 - 60 Mohm) micropipettes were inserted into the AN while stimulating the ear with a calibrated speaker directly coupled to the ear canal. Action potentials from the AN were time-stamped at 1 µs resolution. We collected spike trains from 224 AN fibres at 50 dB SPL in 10 cats and 314 AN fibres at 70 dB SPL in 8 cats.
The observed spatial profiles show broad regions of correlation between 2 and 3 mm centred on the reference fibre across the entire cochlea. The correlation area, which provides a more global measure of level of correlation (correlation energy), clearly decreases towards the base, reflecting the gradual decline of temporal coding with increasing cochlear distance from the apex. The spatial profiles are surprisingly independent of stimulus intensity. Our analysis points towards the importance of monaural neural delays in determining the width of spatial correlation: if monaural delays are not available in the peripheral auditory system, the width of correlation is remarkably invariant with centre frequency. If, on the other hand, such delays are available, the spatial profiles are wider and less invariant.
Marie Curie IEF fellow Damir Kovacic was involved in a number of full-scale auditory neurophysiological experiments. To comply with the legal requirements for independent animal experimentation, he successfully obtained a license for independent auditory neurophysiology experiments. The results were presented at 33rd Annual Midwinter Research Meeting - Association For Research In Otolaryngology (6-10 February 2010) in Anaheim (CA), USA and at the Gordon Research Seminar (12-13 June 2010) at Colby-Saywer College, New London (NH), USA. This research will have potential implications for the better understanding of electric-neural interface between cochlear implant and the auditory nerve and benefits for optimal sound coding strategies in cochlear implants.
Initial cross-correlation analyses of temporal patterns of the neural responses to broadband noise suggest that the magnitude of correlation jumps indicative of discrete organisation is smaller or comparable to the inherent stochastic temporal jittering in the neural responses of the AN. Thus, it might be impossible to demonstrate quantal organisation with the current methods. On the other side, the spatial profile and extent over which correlated responses to broadband noise are found in the AN and its dependency on stimulus intensity have not yet been determined. Namely, cochlear peripheral filtering can produce correlated responses in fibres originating from nearby cochlear positions even to broadband noise. It is a key issue to assess the spatial profiles of across-fibre temporal correlations.
Recordings from single AN fibres were performed in acute experiments on anesthetised cats. The AN was visualised through a posterior fossa craniotomy. High impedance (40 - 60 Mohm) micropipettes were inserted into the AN while stimulating the ear with a calibrated speaker directly coupled to the ear canal. Action potentials from the AN were time-stamped at 1 µs resolution. We collected spike trains from 224 AN fibres at 50 dB SPL in 10 cats and 314 AN fibres at 70 dB SPL in 8 cats.
The observed spatial profiles show broad regions of correlation between 2 and 3 mm centred on the reference fibre across the entire cochlea. The correlation area, which provides a more global measure of level of correlation (correlation energy), clearly decreases towards the base, reflecting the gradual decline of temporal coding with increasing cochlear distance from the apex. The spatial profiles are surprisingly independent of stimulus intensity. Our analysis points towards the importance of monaural neural delays in determining the width of spatial correlation: if monaural delays are not available in the peripheral auditory system, the width of correlation is remarkably invariant with centre frequency. If, on the other hand, such delays are available, the spatial profiles are wider and less invariant.
Marie Curie IEF fellow Damir Kovacic was involved in a number of full-scale auditory neurophysiological experiments. To comply with the legal requirements for independent animal experimentation, he successfully obtained a license for independent auditory neurophysiology experiments. The results were presented at 33rd Annual Midwinter Research Meeting - Association For Research In Otolaryngology (6-10 February 2010) in Anaheim (CA), USA and at the Gordon Research Seminar (12-13 June 2010) at Colby-Saywer College, New London (NH), USA. This research will have potential implications for the better understanding of electric-neural interface between cochlear implant and the auditory nerve and benefits for optimal sound coding strategies in cochlear implants.
