Combining signal-detection theory and interaural cross-correlation approaches to account for the binaural abilities of normal-hearing listeners and listeners with “slight” hearing loss

Abstract

Outstanding among the manifold contributions David M. Green made to psychophysics in general and psychoacoustics in particular was his explication and application of Signal Detection Theory (SDT). Several experimental contexts will be discussed in which a SDT approach to modeling has yielded successful and intuitively appealing accounts of measures of binaural processing. The focus will be, primarily, on relatively recently published empirical data and quantitative modeling from our laboratory. Those reports demonstrate that data obtained in binaural detection experiments conducted across the last five decades can be accounted for by combining a signal-detection-based decision variable with a cross-correlation-based model of binaural processing. Key to the development of the unified account of those experimental results was (1) the inclusion of “internal noise” within stages of the model; and (2) the calculation and inclusion of the variability of the interaural correlations of the outputs of the model for both masker-alone and signal-plus-masker conditions. It will be shown how the SDT-inspired approach has also proven useful in determining and explaining why some listeners with slight, but clinically negligible, elevations in audiometric thresholds exhibit reliable and meaningful deficits in both binaural detection and binaural discrimination tasks. [Work supported by Office of Naval Research (N00014-15-1-2140; N00014-18-1-2473)]