Temporal Pitch Sensitivity in an Animal Model: Psychophysics and Scalp Recordings

Abstract

Cochlear implant (CI) users show limited sensitivity to the temporal pitch conveyed by electric stimulation, contributing to impaired perception of music and of speech in noise. Neurophysiological studies in cats suggest that this limitation is due, in part, to poor transmission of the temporal fine structure (TFS) by the brainstem pathways that are activated by electrical cochlear stimulation. It remains unknown, however, how that neural limit might influence perception in the same animal model. For that reason, we developed non-invasive psychophysical and electrophysiological measures of temporal (i.e., non-spectral) pitch processing in the cat. Normal-hearing (NH) cats were presented with acoustic pulse trains consisting of band-limited harmonic complexes that simulated CI stimulation of the basal cochlea while removing cochlear place-of-excitation cues. In the psychophysical procedure, trained cats detected changes from a base pulse rate to a higher pulse rate. In the scalp-recording procedure, the cortical-evoked acoustic change complex (ACC) and brainstem-generated frequency following response (FFR) were recorded simultaneously in sedated cats for pulse trains that alternated between the base and higher rates. The range of perceptual sensitivity to temporal pitch broadly resembled that of humans but was shifted to somewhat higher rates. The ACC largely paralleled these perceptual patterns, validating its use as an objective measure of temporal pitch sensitivity. The phase-locked FFR, in contrast, showed strong brainstem encoding for all tested pulse rates. These measures demonstrate the cat’s perceptual sensitivity to pitch in the absence of cochlear-place cues and may be valuable for evaluating neural mechanisms of temporal pitch perception in the feline animal model of stimulation by a CI or novel auditory prostheses.

Electrophysiological and Psychophysical Measures of Temporal Pitch Sensitivity in Normal-hearing Listeners

“…No such asymmetry was observed for our pulse-rate changes in the main part of experiment 2, although a significant difference in this direction was obtained in condition 9, which used 94-pps high-region stimuli presented with a 0.5-Hz switch rate. It is also worth noting that the absence of a direction effect in our main data also differ from our findings in the cat using similar stimuli, and where we observed markedly larger ACCs for upward than for downward frequency shifts [27].…”

“…C Amplitude of the composite FFR peak as a function of pulse rate for the contralateral (black) and ipsilateral (red) montage. Data from the cat [ 27 ] are shown in blue …”

“…Amplitudes were constant between 3365–4755 Hz for the “HIGH” condition and 7800–10,800 Hz for the “VHIGH” condition, and decreased with a slope of 48 dB/octave beyond these cutoff frequencies. The frequencies for the HIGH region were chosen to be one octave below those used in our recent study with the cat [ 27 ], so as to roughly compensate for the different audibility ranges of the two species. The frequencies for the VHIGH region were the same as those used in the study by Macherey and Carlyon [ 28 ], which provides useful information on the resolvability of harmonics and on the limits of temporal pitch for these stimuli.…”

“…The black and red lines in Fig. 2C show the effect of montage on the variation in FFR with pulse rate; the blue lines show data from the cat [27] and will be considered in the "Discussion" section Figure 3 shows the composite FFR amplitude for conditions 1-6. A two-way (rate X region) repeated-measures ANOVA revealed a significant effect of rate (F(5, 55) = 30.7, p < 0.001) This was expected based on previous research [38][39][40], and the overall decrease with increasing rate likely reflects low-pass filtering by the head [12] and possibly a decrease in phase locking and/or synchrony with increasing pulse rate.…”

“…[3,[24][25][26]. By using very similar stimuli for the FFR, ACC and perception, we could compare temporal pitch coding at multiple stages of the human auditory system and relate the results to those obtained in an analogous study in the cat [27], from which species we plan to obtain single-unit recordings. Our results demonstrate that an ACC can be observed to a change in temporal pitch and, by comparing the pattern of results to those observed for the FFR and in the psychophysical experiments, provide a potentially important link between physiology and perception.…”