Frequency Analysis of Complex Waveforms in Sound Stimuli Discriminated by Human Auditory Cortex

Abstract

Recognizing objects in the environment from the sounds they produce is one of the primary functions of the auditory system. Recognition is possible, in part, because acoustic features of sounds often represent physical properties of their sources.This present study is the first to investigate whether a single omitted frequency component in the onset period produces significant differentiation.Neural activity of 10 human subjects is recorded coincident with the sound stimulus by whole-head Magneto-encephalography (MEG). In the first set of stimuli constructed from a group of carefully chosen anharmoninc frequency, repetitive (‘standard’) tones with five selected onset frequencies were randomly embedded by rare (‘deviant’) tones, all tones having randomly varying inter stimulus intervals. In the deviant tones one of the frequency components was omitted during the onset period relative to the standard tones. The frequency of the test partial of this complex tone was intentionally selected to preclude its reinsertion by generation of harmonics or combination tones due to either the nonlinearity of the ear, the electronic equipment or the brain processing.In the second set of stimuli, time structured as above, repetitive (‘standard’) tones with five selected sustained partial frequency components were embedded with rare (‘deviant’) tones for which one of these selected partials was omitted this time in the sustained tone. As before, the carefully frequency selection again precluded their reinsertion by generation of harmonics or combination tones due to the nonlinearity of the ear, the electronic equipment and brain processing.By comparing the magnetic field responses (mismatch negativity or MMNm) of the two data sets, the presence of significant mismatch negativity is determined to be due to elimination of a single frequency in the onset period of sound stimuli rather than the sustain period.