Challenges to inter-spike interval models of pitch: the responses of cochlear nucleus neurons to band-pass filtered pulse trains

Bleeck, S., Ingham, N., Carlyon, R.P. and Winter, I.M. (2006) Challenges to inter-spike interval models of pitch: the responses of cochlear nucleus neurons to band-pass filtered pulse trains. 29th Annual MidWinter Research Meeting: Association for Research in Otolaryngology, Baltimore, USA. 05 - 09 Feb 2006. pp.36 .

Abstract

Carlyon et al. (J Acoust Soc Am 112: 621-33, 2002) found that a pulse train with alternating inter-pulse intervals between 4 and 6 ms (“4-6 stimulus”) is perceived as having a periodicity of 5.7ms. Using a model that relies on neuronal first order intervals (FOIs) for pitch perception, they showed that the recovery of the auditory nerve from adaptation is not sufficient to explain this effect and suggested that a weighting function that favours longer FOI intervals was needed to account for the perceived pitch. Here we investigate whether there are neurons, or populations of neurons, at the next stage of processing, the cochlear nucleus (CN), that give greater representation to longer intervals. We measured the responses of single neurons in the ventral CN of anesthetised guinea pigs to isochronous and anisochronous pulse trains. Pulse trains were band-pass filtered between 3900 Hz and 5300 Hz. Isochronous pulse trains had interpulse intervals of 3, 4, 5, 6 and 7 ms. Anisochronous pulse trains had alternating interpulse intervals between 4 and 6 ms. The results are based on the responses of 58 units, classified by the shape of their post stimulus time histograms as Primary with notch (PN, 9), transient chopper (CT, 22), sustained chopper (CS, 10) and onsets (ON, 17). PN and CT units showed a recovery behaviour that is quantitatively similar to the auditory nerve. ON and CS units showed a faster recovery between 4 and 6 ms, but not enough to account for the size of the psychophysical effect. The average of FOI distributions for single units generally has too many long intervals to account for the observed pitch of the 4-6 stimulus. When sequentially interleaving the spikes from all units the average FOI distribution contains more shorter intervals, but, by interleaving spikes from just 2 to 4 units the average FOI becomes small enough to explain the observed pitch of the 4-6 stimulus. The location of units that may participate in such an analysis remains obscure.

This record has no associated files available for download.

Contributors

Author: S. Bleeck

Download statistics