Discrimination of synthetic full-formant and sinewave/ra-la/continua by budgerigars (Melopsittacus undulatus) and zebra finches (Taeniopygia guttata).

Discrimination of three synthetic versions of a/ra-la/ speech continuum was studied in two species of birds. The stimuli used in these experiments were identical to those used in a previous study of speech perception by humans [Best et al., Percept. Psychophys. 45, 237-250 (1989)]. Budgerigars and zebra finches were trained using operant conditioning and tested on three different series of acoustic stimuli: three-formant synthetic speech, sinewave versions of those tokens, and isolated F3 tones from the sinewave speech. Both species showed enhanced discrimination performance near the /l/-/r/ boundary in the full-formant speech continuum, whereas for the F3 continuum, neither species showed a peak near this boundary. These results are similar to human discrimination of the same continua. Budgerigars also showed a peak in discrimination of the sinewave analog continuum paralleling that for full-formant syllables, similar to humans who are induced to perceive sinewave speech as speech. Zebra finches, by contrast, showed a relatively flat function mirroring their performance for F3 sinewaves, similar to humans who are induced to perceive sinewave speech as nonspeech. These data provide new evidence of species similarities and differences in the discrimination of speech and speechlike sounds. These data also strengthen and refine previous findings on the sensitivities of the vertebrate auditory system to the acoustic distinctions between speech sound categories.