Abstract

A Finite-Difference Time Domain (FDTD) physical model of the basilar membrane (BM) is implemented including the transition of mechanical energy on the BM into spike excitation. The spike train caused by this transition shows energy at all Bark bands of a harmonic input sound. Still higher partials are often not able to produce a regular interspike interval (IS) firing pattern all through one periodicity of the fundamental frequency. Instead drop-outs occur, where one or several spikes are missing. This produces higher ISI periodicities and therefore lower frequencies as the critical frequency of the respective Bark bands. These lower frequencies are integer subharmonics of the critical frequency. The lowest of these frequencies is the fundamental frequency of the input sound. Therefore, the model shows these fundamental frequencies in all Bark banks of active sound input frequency. The fundamental frequency of a harmonic sound is therefore present at many cochlea fibers and may be the reason for pitch perception. This mechanism also works for residual pitches. Compared to other pitch models which need further calculation in following neural nuclei, like autocorrelations, the pitch model proposed here has pitch intrinsically present right at the cochlea output with no need of further computations. From this view, a pitchness can be defined as the strength of a frequency present at many Bark bands, in accordance with the strength of a pitch perception.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call