The ability of cats to discriminate accurately among different synthetic, steady‐state vowels was examined across a range of stimulus levels and in background noise. Cats were trained to press and hold down a lever to produce a pulsed train of a standard vowel stimulus, and to release the lever only when a different vowel sound occurred. Five synthetic vowels were tested (/ε/, /æ/, /a/, /■/, and /■/) at levels of 30, 50, 70, and 90 dB SPL. In separate experiments, each of these vowels served in turn as the standard vowel. All cats discriminated among the vowels accurately, and in general performed at least as well at high stimulus levels as at low levels. Where differences in vowel discriminability occurred, they were correlated with the relative changes in first and second formant peaks. Cats appear to predominantly utilize upward frequency changes in either the first or second formants of the vowels to make the discriminations; downward formant changes produced considerably lower discrimination performances. In background noise, high vowel discriminability was still maintained at an average signal/noise ratio of −12.3 dB. Thus cats can discriminate among vowels at high signal levels and in background noise, despite the fact that the neural representations of vowels based on rate responses in the auditory nerve can be severely degraded under these conditions.

This content is only available via PDF.
You do not currently have access to this content.