Detection of changes in timbre and harmonicity in complex sounds by zebra finches (Taeniopygia guttata) and budgerigars (Melopsittacus undulatus).

Abstract

Thresholds for detecting alterations in the timbre and harmonicity of complex harmonic signals were measured in zebra finches, budgerigars, and humans. The stimuli used in this experiment were designed to have particular salience for zebra finches by modeling them after natural zebra finch calls. All 3 species showed similar abilities for detecting an amplitude decrement in a single component of a harmonic complex. However, zebra finches and budgerigars were extraordinarily sensitive to the mistunings of single harmonics and exhibited significantly lower thresholds compared with humans at 2 different fundamental frequencies, 570 Hz and 285 Hz. Randomizing relative phases of components in a harmonic stimulus resulted in a significant increase in threshold for detecting mistunings in zebra finches but not in humans. Decreasing the duration of mistuned harmonic stimuli resulted in higher thresholds for both birds and humans. The overall superiority of birds in discriminating inharmonicity suggests that birds and mammals may use different strategies in processing these complex harmonic sounds.

Bird song, ecology and speciation

“…Several other studies show that birds are remarkably sensitive to spectral and temporal detail in relatively complex song (e.g. Sinnott et al 1980;Cynx et al 1990;Hurly et al 1992;Lohr & Dooling 1998). Artificial circumstances in tests with operant conditioning may overestimate sensitivity when findings are extrapolated to the field, but field playbacks also confirm that minute acoustic changes, close to the perceptual resolution of a species, may be meaningful to birds under natural conditions (Slabbekoorn & ten Cate 1998).…”

Chickadee songs provide hidden clues to singers’ locations

“…Intriguingly, recent acoustic analyses show that dominant singing chickadees reliably produce songs with comparable spectral energy in their fee and bee notes (Hoeschele et al, 2010), providing a stable baseline for cross-note comparisons. Experimental studies of timbre discrimination in zebra finches and budgerigars show that they can detect changes of 1 -3 dB in harmonic levels, which is slightly better than performance by humans (Cynx et al, 1990;Lohr & Dooling, 1998). Human sensitivities for detecting such changes in two non-harmonic tones predict that deviations in the range of 0.2 -0.5 dB are potentially discriminable by birds and other vertebrates.…”

“…Profile analysis involves detecting an absolute increment in the amplitude of one tonal signal among a background of other tonal signals. Spectral shape discrimination similarly involves distinguishing patterns in spectral levels; recognition of such differences is thought to contribute to timbre perception (Cynx, Williams, & Nottebohm, 1990;Hoeschele, Cook, Guillette, Hahn, & Sturdy, 2014;Lohr & Dooling, 1998). In the case of coincident reverberation cues within received fee-bee songs, both directly received and reverberant spectral peaks decrease in amplitude with distance, but at different rates, so that the relative difference in amplitude between the two peaks gradually decreases with distance.…”

Species differences in the identification of acoustic stimuli by birds

“…However, there are some studies that highlight the differences in hearing between the two species of animals. Zebra finches are better than budgerigars at detecting a mistuned harmonic (Lohr and Dooling, 1998) and discriminating temporal fine structure (Dooling et al, 2000). Budgerigars have better spectral resolving power than zebra finches for stimuli within 2-3 kHz (Okanoya and Dooling, 1987), are better at absolute sound localization (Park and Dooling, 1991), and at discriminating amongst stimuli mimicking the precedence effect (Dent and Dooling, 2004).…”