Research On Vocal Regulations Of Five Species Of CF-FM Bats In Noisy Environment

Echolocating bats rely highly on sound for navigation,foraging,and many other activities,yet the efficiency of echolocation can be strongly hindered by background noise.Considering the evolutionary success of bats as dominant echolocating animals,they may have evolved behavioral and physiological mechanisms to cope with noise interference.Current studies on the effects of noise on vocal regulation behaviors of bats have mainly focused on FM(frequency modulation)bats,and the limited studies on CF-FM(constant frequency-frequency modulation)bats revealed contrasting Lombard effects(i.e.,a rise in call amplitude in response to increasing ambient noise level)between the Hipposiderid bats(Chiroptera: Hipposideridae),and the Rhinolophus bats(Chiroptera: Rhinolophidae),with the latter showing virtually no Lombard effect in Greater Horseshoe Bat(Rhinolophus ferrumequinum).Is this phenomenon a specificity of species R.ferrumequinum or a general feature of Rhinolophus bats? Furthermore,the question of whether bats have the same adaptive strategies in response to noise disturbance in different behavioral states and tasks remains unclear.In this study,we investigated the vocal regulation behaviors of four species of Rhinolophus bats and the pratt’s leaf-nosed bat(Hipposideros pratti)exposed to noise playbacks.Using a multi-channel microphone array system,we first investigated and compared the vocal regulation behaviors of these five species during a free hanging state.Next,combined with highspeed stereo video cameras,we compared the differences in echolocation signal characteristics of H.pratti between the resting and flying states under noise treatments.The main results are as follows:1.All five CF-FM bats showed dynamic regulations of multiple echolocation signal characteristics when exposed to noise playbacks during the resting state,including the Lombard effect.However,the magnitude of signal adjustments were significantly different between species.The increases in call peak amplitude(i.e.,the Lombard effect)were more pronounced in Rhinolophus osgoodi,Rhinolophus macrotis,and Rhinolophus siamensis,with increases of 7.09 d B,6.64 d B,and 5.79 d B,respectively,compared to the median of the control group(p < 0.001).The Lombard effects were relatively weak in Rhinolophus sinicus and H.pratti,with increases of 2.61 d B and 1.50 d B,respectively(p < 0.001).2.H.pratti exhibited more significant Lombard effects in flight than at rest,and other echolocation signal characteristics in flight differed from the resting experiments.The median call peak amplitude increased by 7.90 d B and 6.65 d B of the two phases under the flight state(p < 0.001)and only 1.63 d B under the resting state(p < 0.001)in the 80 d B SPL group compared to the 0 d B SPL group when exposed to 10 k Hz-90 k Hz white noise.3.H.pratti exhibited different vocal regulation strategies in response to noise interference at different time stages,which may be related to the operant training or motors that mimicked the fluttering insects.Compared to the animal training period in May,H.pratti had a 5.20 ± 0.83 d B(p < 0.001)reduction in call peak amplitude during motor rotation session in November 2022 when exposed to 10 k Hz-90 k Hz white noise.The call peak amplitude increased by 0.76 ± 0.39 d B and 1.60 ± 0.38 d B(p < 0.001)in the 80 d B SPL group compared to silence control groups when exposed to 10 k Hz-90 k Hz white noise during the training period in May and the post-training period in November,respectively.The call peak amplitude and peak frequency differed significantly(p < 0.001)between motor rotation and non-rotation sessions in the silence control group during the post-training period in November.In summary,this study showed that at rest the vocal regulations of different CF-FM bats in response to noise interferences are differential,and H.pratti can flexibly adjust vocal parameters in response to noise disturbance in different contexts and task states,and the task-related Lombard effect is dynamic.In addition,the echolocation signals of H.pratti are plastic in noise.These data bear implications for understanding the efficient echolocation system of bats in noise.