The normalized interaural correlation: Accounting for NoS pi thresholds obtained with Gaussian and "low-noise" masking noise

Recently, Eddins and Barber [J. Acoust. Sec. Am. 103, 2578-2589 (1998)] and Hall et al. [. Acoust. Sec, Am. 103, 2573-2577 (1998)] independently repor ted that greater masking of interaurally phase-reversed (S pi) tones was pr oduced by diotic low-noise noise than by diotic Gaussian noise. Based on qu antitative analyses, Eddins and Barber suggested that their results could n ot be accounted for by assuming that listeners' judgments were based on con stant-criterion changes in the normalized interaural correlation produced b y adding the S pi signal to the diotic masker. In particular, they showed t hat a model like the one previously employed by Bernstein and Trahiotis [J. Acoust. Sec. Am. 100, 3774-3784 (1996)] predicted an ordering of threshold s between the conditions of interest that was opposite to that observed. Be rnstein and Trahiotis computed the normalized interaural correlation subseq uent to half-wave, square-law rectification and low-pass filtering, the par ameters of which were chosen to mimic peripheral auditory processing. Tn th is report, it is demonstrated that augmenting the model by adding a physiol ogically valid stage of "envelope compression" prior to rectification and l ow-pass filtering provides a remedy. The new model not only accounts for th e data obtained by Eddins and Barber (and the similar data obtained by Hall er al.), but also does not diminish the highly successful account of the c omprehensive set of data that gave rise to the original form of the model. Therefore, models based on the computation of the normalized interaural cor relation appear to remain valid because they can account, both quantitative ly and qualitatively, for a wide variety of binaural detection and discrimi nation data. (C) 1999 Acoustical Society of America. [S0001-4966(99)06108-1 ].