Binaural Hearing and the Cochlear Phase Response

Early psychoacoustic research suggested that the human auditory system is insensitive to differences in the relative phases of spectral components of a multicomponent sound. However, research from the last two decennia provides evidence that listeners can detect phase differences between the stimulus components that interact within a single auditory filter. The most impressive demonstration of phase sensitivity is given by the masker-phase effect, i.e. the more than 20-dB variation in masking effect caused by a harmonic complex when varying the phase relations between its components. This masking paradigm is widely used to obtain a psychoacoustical measure of the phase response of the cochlea. Interaural time differences (ITD) are important information for the auditory system to determine the lateral position of a sound source. Studies on ITD perception using multicomponent stimuli set the relative phases of the components either to a constant or to random values. Because ITD perception relies on the temporal properties of the cochlear response to the signal, it can be expected that the phase relation between the components of a harmonic complex affects ITD perception, particularly at higher frequency regions where multiple components interact within an auditory filter. Therefore, the first main goal of BiPhase is to systematically study the effect of phase variations in harmonic complexes on ITD sensitivity in normal-hearing (NH) listeners. Cochlear hearing impairment is known to often cause degraded ITD sensitivity compared to NH listeners. A critical factor in case of multicomponent stimuli appears to be the physiologically- demonstrated alteration of the phase response in the impaired cochlea. Furthermore, psychoacoustic studies showed that the masker-phase effect is markedly reduced in hearing-impaired (HI) listeners. Although the origins for this reduction are still subject of research, it appears likely that phase relations that are optimal for ITD perception in normal hearing can be suboptimal in impaired hearing. Thus, the second main goal of BiPhase is to systematically study phase effects on ITD sensitivity in HI listeners and compare them to the effects found in NH listeners. The reduction or even absence of compression in impaired cochleae has been identified as an important reason for the reduction of the masker-phase effect in HI listeners. Assuming that compression is not required to observe phase effects in ITD perception (compression may actually reduce phase effects), ITD-based phase effects could be stronger than masking-based phase effects in HI listeners. Therefore, the third main goal of BiPhase is to compare phase effects measured with the ITD-based and masking-based paradigms in NH and HI listeners. BiPhase is expected to provide more insight into cochlear phase effects for multicomponent stimuli in normal and impaired hearing. The ITD-based paradigm could be a useful tool to determine the cochlear phase response in hearing impairment. The better understanding of phase effects on ITD perception may provide the basis for developing better processing schemes for hearing devices in order to restore the access to ITD cues in HI listeners. This in turn may improve their ability to localize sound sources and understand speech in noise.

It is well established that the auditory system uses temporal cues to extract information from everyday sounds such as speech. Temporal information appears to be especially important for people suffering from cochlear hearing impairment (HI), because the reduction of amplitude compression as a result of outer hair cell loss in the cochlea results in worsened place coding. Despite the importance of the cochlear phase response (PR) for encoding temporal information, relatively little is known about the PR in the impaired cochlea. While the loss of compression likely considerably alters the PR and, thus, temporal coding, the common technique to estimate the PR, the masking paradigm, shows no elusive results.The BiPhase project studied the role of compression in the measurement of the PR in normal hearing and the feasibility of a new measurement paradigm which does not rely on cochlear compression and is, thus, may be better applicable in HI listener. To avoid listener-specific factors in HI, we focused on testing normal-hearing listeners. First, we studied the role of compression by combining the masking paradigm with the precursor approach, intended to reduce cochlear compression by means of an efferent feedback mechanism. The experimental results, together with a modeling approach, confirmed the hypothesis that the loss of compression largely reduces phase effects in masking and therefore hinders meaningful application of the masking approach in HI listeners. A follow-up study addressed open questions regarding the influence of various stimulus parameters and the underlying mechanisms. In the second part of BiPhase, we tested the proposed method which does not rely on masking (and thus, on compression), instead relying on envelope ITD cues that are thought to be preserved in HI listeners. The results were roughly comparable to the masking paradigm and could be largely predicted by a model of the binaural auditory periphery, except for an unexpected peak in the pattern of threshold ITDs across stimulus phase conditions. A modified testing paradigm, relying on the extent of laterality, showed did not show the peak and and was completely consistent with the model prediction. Simulation of cochlear hearing impairment predicted the general applicability of the ITD-based method to HI listeners. Finally, we explored the feasibility of an iterative version of the ITD-paradigm that allows for the measurement of non-uniform PRs potentially occuring in HI listeners.Despite rough consistence with the expected overall pattern, results with the current method appear to be too variable for practical application with HI listeners. Overall, the BiPhase project showed the feasibility of a new paradigm for measuring the cochlear PR relying on interaural temporal envelope cues. We also identified a number of questions for future studies, e.g.: Can the method be simplified to monaural envelope rate pitch sensitivity instead of envelope ITD sensitivity? How successful is the method in actual HI listeners? After successfully addressing these questions, the application of individually measured PRs of HI listeners in future hearing devices may enhance envelope pitch and ITD cues which in turn may improve auditory communication in challenging listening environments involving multiple sound sources.