A Re-examination of the Effect of Masker Phase Curvature on Non-simultaneous Masking

Robert P Carlyon,Sheila Flanagan,John M Deeks

doi:10.1007/s10162-017-0637-5

Abstract

Forward masking of a sinusoidal signal is determined not only by the masker’s power spectrum but also by its phase spectrum. Specifically, when the phase spectrum is such that the output of an auditory filter centred on the signal has a highly modulated (“peaked”) envelope, there is less masking than when that envelope is flat. This finding has been attributed to non-linearities, such as compression, reducing the average neural response to maskers that produce more peaked auditory filter outputs (Carlyon and Datta, J Acoust Soc Am 101:3636–3647, 1997). Here we evaluate an alternative explanation proposed by Wotcjzak and Oxenham (Wojtczak and Oxenham, J Assoc Res Otolaryngol 10:595–607, 2009). They reported a masker phase effect for 6-kHz signals when the masker components were at least an octave below the signal frequency. Wotcjzak and Oxenham argued that this effect was inconsistent with cochlear compression, and, because it did not occur at lower signal frequencies, was also inconsistent with more central compression. It was instead attributed to activation of the efferent system reducing the response to the subsequent probe. Here, experiment 1 replicated their main findings. Experiment 2 showed that the phase effect on off-frequency forward masking is similar at signal frequencies of 2 and 6 kHz, provided that one equates the number of components likely to interact within an auditory filter centred on the signal, thereby roughly equating the effect of masker phase on the peakiness of that filter output. Experiment 3 showed that for some subjects, masker phase also had a strong influence on off-frequency backward masking of the signal, and that the size of this effect correlated across subjects with that observed in forward masking. We conclude that the masker phase effect is mediated mainly by cochlear non-linearities, with a possible additional effect of more central compression. The data are not consistent with a role for the efferent system.

Highlights

For many decades, models and accounts of masking were based on the power spectrum of the masker and signal, as processed by the amplitude characteristics of a bank of putative auditory filters, and on the signalto-noise ratio at the outputs of those filters (Fletcher, 1940; Patterson, 1976; Glasberg and Moore, 1990)
Experiment 2 showed that the phase effect on off-frequency forward masking is similar at signal frequencies of 2 and 6 kHz, provided that one equates the number of components likely to interact within an auditory filter centred on the signal, thereby roughly equating the effect of masker phase on the peakiness of that filter output
The present study provides a further investigation of the effects of phase curvature on non-simultaneous masking and, in particular, investigates the potential role of the efferent system

Summary

Introduction

Models and accounts of masking were based on the power spectrum of the masker and signal, as processed by the amplitude characteristics of a bank of putative auditory filters, and on the signalto-noise ratio at the outputs of those filters (Fletcher, 1940; Patterson, 1976; Glasberg and Moore, 1990). Such models account for a wide range of data, more recent accounts have focussed on a number of important failures of the power spectrum model. The components of the two maskers were summed in so-called positive and negative Schroeder phase (Schroeder, 1970), corresponding to values of +1 and −1 for the parameter C in Eq 1, where N is the total number of components and θn is the phase of the nth component: θn 1⁄4 Cπnðn−1Þ=N ð1Þ

Methods

Results

Conclusion