ABSTRACT
We propose a model of amplitude modulation processing and we compare its performances with psychoacoustical data. Sensitivity to modulation depth, masking between two modulations and sensitivity to their dephasing are investigated. Mechanisms of amplitude modulation (AM) processing are discussed in light of these experiments. We conclude that there are other plausible neural mechanisms than the autocorrelogram for explaining AM sensitivity and harmonic grouping in the intermediate levels of the auditory system.
