Neuromagnetic representation of temporal and spectral pitch in the auditory cortex
Temporal integration in the auditory system within 30 milliseconds plays a crucial role in pitch processing. Current models suppose that pitch perception is based on the processing of (i) the temporal regularity of a sound and (ii) its spectral envelope. Although, the representation and integration of both characteristics remains unclear, the latency of the late AEFs exhibit a high correlation to the inverse of the perceived pitch (Gutschalk et al., 2002; Ritter et al., 2005, 2007; Rupp et al., 2005). In an interdisciplinary project (Institut für Theoretische Physik Heidelberg, Prof. Dosch and the Centre for the Neural Basis of Hearing, Department of Physiology in Cambridge, Prof. Patterson) we are developing critical stimuli by manipulating the temporal regularity as well as the temporal and spectral envelope of sounds. Spatio-temporal source analysis will be applied to locate pitch-related centers in the audiory cortex and to investigate the temporal course of the AEF. These experiments will be carried out in subjects who perceive dominantly the fundamental pitch of an uncomplete harmonic complex sound and a complementary group of listeners, who dominantly perceive specific harmonics of such sounds. The first group exhibits larger AEF and larger grey matter volumes in the left hemisphere while the latter shows the opposite behaviour. The choice of these subjects will provide valuable information to investigate the dependency of pitch perception on temporal and spectral aspects.
Differences in auditory evoked fields and grey matter volume of the left and right Heschl’s gyrus between musicians and non-musicians
The extensive research on auditory fields and morphological characteristics (Schneider et al., 2002; 2005) showed that musicians exhibit enhanced P30 components (located in the primary auditory cortex, i.e. medial Heschl’s gyrus) and P50-components (located in lateral Heschl’s gyrus) compared to non-musicians. Furthermore, the magnitude of these signals are correlated with (i) the absolute grey matter volume of these areas and (ii) a psychometric test to assess musicality. Currently, these investigations will be continued in a cross-sectional and longitudinal design in children and adolescents to further investigate the influence of training on the plasticity of these components, because the P50 amplitude is strongly correlated to the long-term duration of professional training while the P30 amplitude is affected to a much lesser degree.
Early representation of the auditory nerve spike pattern in the primary auditory cortex
Alternating the phases of a harmonic tone produces puretone masked thresholds differing by more than 20 dB. This effect is due to the characteristics of the auditory filters on the basilar membrane which differently affect ringing within channels. Based on the analysis of early AEFs and basilar membrane simulations (Rupp et al., 2002), we analyze in a collaboration with Prof. Dau (DTU Oersted, Denmark) the representation of such peripheral effects in the auditory cortex. First results indicate that neuromagnetic responses are highly consistent with perceptual properties obtained with the same stimuli and with results from simulations of neural activity at the output of cochlear preprocessing. This suggests that the activity wave in the auditory nerve keeps its across-frequency timing structure travelling to the auditory cortex and that the AEF that is reflected in the MEG response is strongly related to the timing structures. Due to the high correlation of basilar membrane characteristics and the morphology of the auditory evoked fields, we are currently optimizing semi-realistic models of the cochlea to enhance the validity of spike simulations in the auditory nerve (Sieroka et al., 2006). Such investigations might support the development of hearing aid algorithms that simulate the compressive behaviour of the cohlea which is found in normal hearing subjects.
Fig. 1: Simulations of the neural activity in the auditory nerve of a harmonic complex sound f0=250 Hz (a) tonotopic axis with high frequencies at the top and low frequencies at the bottom), and the stabilized auditory image (b) that results from an alignment across channels due to a non-linear strobing mechanism. This computational stage results in a buffer with a decay of excitation within 32 ms. The distance between the ridges corresponds to the perceived pitch of a tone and the height of the ridges determines the salience of pitch.
|(c) The magnitude of the N100 evoked by tones with different temporal envelopes (ramped vs. damped with different half-life-times) as derived from sources in the lateral Heschl’s gyrus correspond to both, (d) the psycho-acoustically-determined salience of these tones when compared to each other, and (e) the height of the first ridge of the stabilized auditory image.|