Zebra finches produce soft laryngeal whistles during thermal panting that are not adaptive vocal signals.

Birds and mammals converged upon the same physical mechanism of vocal fold vibration^1,2 to produce the wide variety of communicative vocal signals critical for their reproduction, social interactions, survival, and speciation.^3,4,5 Recent work reported high-frequency (7-11 kHz) vocalizations in zebra finches,⁶ termed "heat" or "incubation" calls, that are suggested to have close-range communicative relevance in the context of global warming.^6,7 However, their acoustics are poorly described, and by what biophysical mechanism they are produced remains unknown. We recorded heat-associated vocalizations in adult zebra finches in vivo and showed that they are extremely soft, frequency-modulated vocalizations with source levels of 13.9 ± 3.3 dB sound pressure level (SPL) at 1 m and dominant frequencies of 6.8 ± 0.6 kHz. Through in vitro experiments, we establish that these vocalizations are aerodynamic whistles produced inside the avian larynx, not syrinx, during inspiration. Respiratory air flow during whistle production is higher than during regular song and consistent with thermal panting for evaporative cooling.^7,8 Laryngeal geometry and dimensional flow analysis suggest that these whistles are laminar-flow whistles that occur when a flow boundary layer is in a transition phase from laminar to turbulent flows.^9,10,11 Our data imply that in earlier experiments,^6,7 playbacks were conducted at 30 dB above the physiologically relevant playback level (i.e., several magnitudes). Furthermore, heat whistles are at least 35 dB below the behavioral perception limit even of adults¹² with more sensitive hearing than juveniles.^13,14 Therefore, we show-contra previous works^6,7-that heat whistles cannot function as adaptive signals or cues in parent-embryo communication.