Peripheral and central components of comodulation masking release

Abstract

Comodulation masking release (CMR) is the improvement in threshold that occurs when the subaudio amplitude modulations of a masker envelope in the critical band around a target tone are replicated on one or more flanking bands outside the critical band; the phenomenon is of obvious significance for perceptual grouping and attention and may stimulate a literature on parametric variations as large as that for the BMLD. Basic general questions first need to be answered about the nature of the effect, specifically concerning the relative roles of (presumably peripheral) psychophysical suppression and of (presumably central) temporal pattern comparison. In the first experiment the effect was tested monotically, diotically, and dichotically with one monotic control condition implementing 40-dB intensity disparity (as might result from worst-case transcranial conduction in the dichotic condition). Although intensity effects were present, a significant CMR was observed dichotically, suggesting that peripheral explanations such as suppression must be incomplete. The second experiment varied the phase (0°, 90°, 180°, and 270°) of the critical band and flanking band modulations. This experiment also produced results that cannot be completely explained by either a central statistical correction model, or a peripheral suppression model alone; negative CMRs were seen in some subjects.