Critical bands and critical ratios in animal psychoacoustics: An example using chinchilla data

This paper suggests that critical ratios obtained in noise-masked tone studies are not good indicators of critical bandwidths obtained in both human and nonhuman animal subjects. A probe-tone detection study using chinchilla subjects suggests that they may be broadband processors in detection tasks as opposed to human subjects who use narrow-band, critical-band processing. If chinchilla and other nonhuman animal subjects are wideband processors, this can partially explain why their critical ratios are significantly greater than those measured in human subjects. Thus, large critical ratios obtained for nonhuman animals may indicate processing inefficiency rather than wide critical bands.