The critical band and the critical ratio

Abstract

The critical band is arguably one of the most important psychophysical descriptors of auditory function. The critical ratio is often used in animal psychophysical research to estimate critical bandwidth because it is easier to obtain than estimates of critical bandwidth. However, in most cases the underlying assumptions required to use the critical ratio as an estimate of critical bandwidth are rarely met. The critical bandwidths for chinchilla and guinea pigs are similar to those obtained for human subjects, while the critical ratios are considerably higher, an observation that violates the use of the critical ratio as an estimate of critical bandwidth. This paper will describe the assumptions necessary for the critical ratio to estimate critical bandwidth. We will describe data on critical bands and critical ratios for chinchilla and guinea pigs, and we will provide evidence that the chinchilla appears to be a wideband processor of the signal in critical-ratio experiments. The inability of the chinchilla to process the signal in its narrow critical band in the critical-ratio psychophysical procedure leads to the large critical ratios. We recommend that the critical ratio not be used to estimate critical bandwidth. [Work supported by NIDCD grants to Dr. Yost and to Dr. Shofner.]