The Perception of Complex Acoustic Patterns in Noise by Blue Monkey (Cercopithecus mitis) and Human Listeners

Abstract

Bluemonkeys (Cercopithecus mitis) were trained to detect acoustic signalsembedded in noise. Masked thresholds were determined for four hu consonant-vowel speechsounds (6a, pa, ga, and ka), and four blue monkey (boom,pyow, chirp, and trill). The ability of monkey listeners to hear these was compared with humans. Results showed that monkey and human wasvery similar. The mean difference between species for these eight stimuli in broad-bandnoise environment was 2.3 dB. The signal-to-noise ratio for ranged from4.8 dB for the ka to -23.8 dB for the boom. The four monkey calls audibleat a signal-to-noise level that was 8.1 dB less than that required for the the speech sounds. However, most of this effect was due to the audibility of boom.With the boom excluded, the mean signal-to-noise ratio for detection of the re 7sounds was -0.5 dB, and the mean difference in the audibility of andmonkey sounds within this set was 2.6 dB. These results contrast findingswhich used simulated rain forest noise as the masking noise (Brown, 1986). rainforest noise, test signals were audible at signal-to-noise ratios approximately dBless than those reported here, and the observed difference in the human and monkey utterances was larger. These findings variations inthe amplitude and spectrum of the ambient noise may have influenceon the audibility of vocal signals in nature.a strongsuggest that rather smallrelative audibilityof 10Inwith previousthe speechmainingthedetectionof wereperceptionthehearingsignals innoise callsmancomplex