Comodulation masking release in a songbird

Comodulation masking release (CMR) describes the reduced masking of a pure tone when the masking is a noise that is coherently amplitude modulated (comodulated) over the total range of the spectrum compared to masking by an unmodulated noise of the same bandwidth and overall energy. The masking release results from cues available within a critical band and from cues generated by comparisons across critical bands ('true' CMR). Here we report data on masking release and 'true' CMR in a songbird, the European starling (Strunus vulgaris), that was demonstrated in a psychoacoustic experiment using a GO/NOGO paradigm. Masked thresholds for 2-kHz tones centered in digitally generated continuous masking noise of different bandwidths were determined, and the amount of masking release was calculated as the threshold difference between the unmodulated and the comodulated condition. In the first experiment the modulator was a 50-Hz lowpass noise. A masking release of 11.8 dB was found for the noise masker with the largest bandwidth (1600 Hz). With the masker bandwidth decreasing to 50 Hz, the birds' release from masking was reduced to 1.6 dB. The starling's 'true' CMR was 4 dB or 8 dB, depending on the definition that was applied. In a second experiment the masker bandwidth was constant (1600 Hz) and the cut-off frequency of the modulator was varied. A release from masking of 17.8 dB was found for a modulator cut-off frequency of 12.5 Hz. It decreased to 6.1 dB with an increase in the modulator cut-off frequency to 400 Hz. The duration of the test signal (100-750 ms) had little effect on the release from masking. Given the similarities in the release from masking and in CMR of starlings and humans, the starling may provide an excellent model for studying the mechanisms that underlie the generation of CMR.