Biosonar Emissions Research Articles

Biosonar sampling strategies in frequency-modulated echolocating bats

Echolocating bats dynamically control multiple aspects of their biosonar emissions (emission rate, frequency, bandwidth, duration, and direction) in parallel as a function of their surroundings and current goals. As environments become more difficult to navigate and tasks become more complex, bats converge on sampling strategies that have less variability in the timing and spatial changes of their echolocation. We investigated these biosonar dynamics within a virtual echo target presentation paradigm where we challenged big brown bats sitting on a platform to localize a single echo target, which was either stationary or moved unpredictably in azimuth. Acoustic clutter was also added at varying distances to investigate its effect on their sampling strategies during the task. Bats modified their biosonar strategies differently depending on the distance of added acoustic clutter and the distance that a target shifted, and individuals tended to converge on similar strategies. We also show that certain perceptual contexts can lead to unexpected biosonar strategies in individual bats, such as the use of longer-duration emissions with less emission ‘clustering’ when localizing a virtual target against nearby clutter.

Machine learning methods for reconstructing the acoustic fields of bat biosonar

Understanding the relationships between biosonar and flight in bats requires synchronized recordings of the echo inputs and the flight kinematics of the animals. Here, integrated arrays of 50 high-speed camera and 32 ultrasonic microphones have been set up to collect the data that is necessary for achieving this goal. The arrays are set up to record bats as they fly through an an obstacle course inside a cylindrical tunnel. Due to the complexity of the received signals in these scenarios, we have been developing custom strategies that rely on a combination of compressed sensing and deep learning to reconstruct the acoustic field inside the tunnel from the microphone-array measurements. By using a combination of the high-speed video and acoustic data, it can be attempted to determine the bat's location in the array from the image data and then reconstruct the properties of the sound fields that were emitted and received at this position using the acoustic array data. The goal of this research is to eventually determine the dynamic characteristics of the bat's biosonar emissions such as beampatterns and potentially time-variant characteristics as well as the biosonar inputs that the bats rely on for controlling their flight behaviors.

Echolocating bats modify biosonar emissions when avoiding obstacles during difficult navigation tasks

Big brown bats modulate their biosonar emissions in both time and frequency when flying through acoustically cluttered environments. We challenged single bats or bats in pairs to fly along a narrow U-shaped corridor 7 m in path length formed by rows of vertical plastic chains, while also avoiding additional interspersed horizontal chains. Bats emitted calls in complex sonar sound groups in all flight conditions. When no added horizontal obstacles were present, each bat’s flight success rates averaged 80%; introducing horizontal obstacles decreased success rates to 45-65%, with little improvement over several days. When two bats flew simultaneously from opposite ends of the U-shaped path, they rarely collided, instead swerving closely past or keeping farther apart in elevation. They increased call rate and call amplitude, while decreasing terminal FM broadcast frequency, as they approached near the middle of the U-shaped path. Increases in call amplitude persisted throughout the remainder of the flight while call rate and terminal frequency returned to baseline levels. These data show the limits of call modifications during difficult maneuvers in dynamic conditions. They demonstrate the bat’s effective awareness not only of the continuously distributed clutter but also of another, moving bat. [Work supported by ONR.]

Effects of dolphin hearing bandwidth on biosonar click emissions.

Differences in odontocete biosonar emissions have been reported for animals with hearing loss compared to those with normal hearing. For example, some animals with high-frequency hearing loss have been observed to lower the dominant frequencies of biosonar signals to better match a reduced audible frequency range. However, these observations have been limited to only a few individuals and there has been no systematic effort to examine how animals with varying degrees of hearing loss might alter biosonar click properties. In the present study, relationships between age, biosonar click emissions, auditory evoked potentials (AEPs), and hearing bandwidth were studied in 16 bottlenose dolphins (Tursiops truncatus) of various ages and hearing capabilities. Underwater hearing thresholds were estimated by measuring steady-state AEPs to sinusoidal amplitude modulated tones at frequencies from 23 to 152 kHz. Input-output functions were generated at each tested frequency and used to calculate frequency-specific thresholds and the upper-frequency limit of hearing for each subject. Click emissions were measured during a biosonar aspect change detection task using a physical target. Relationships between hearing capabilities and the acoustic parameters of biosonar signals are described here and compared to previous experiments with fewer subjects.

Relationship between biosonar click emissions, age, and hearing bandwidth in bottlenose dolphins, Tursiops truncatus

Previous studies have reported changes in biosonar emissions in a few odontocete subjects as audible frequency range is reduced with increasing age (i.e., presbycusis). For example, some animals have been observed to lower the dominant frequencies of their biosonar clicks to better match their reduced audible frequency range. In the present study, relationships between age, biosonar click emissions, auditory evoked potentials, and hearing bandwidth were studied in 10 bottlenose dolphins (Tursiops truncatus) of various ages and hearing capabilities. Underwater hearing thresholds were estimated by measuring steady-state auditory evoked potentials to sinusoidal amplitude modulated tones at frequencies from 20 to 160 kHz. Input-output functions were generated at each tested frequency and used to calculate frequency-specific thresholds and the upper cut-off frequency of hearing for each subject. Click emissions were measured during a physical target aspect change detection task at a fixed range of 10 m. Relationships between hearing thresholds and the acoustic parameters of biosonar signals will be presented and compared to previous experiments with fewer subjects. [Work supported by the Office of Naval Research.]

Big brown bats (Eptesicus fuscus) successfully navigate through clutter after exposure to intense band-limited sound

Echolocating big brown bats fly, orient, forage, and roost in cluttered acoustic environments in which aggregate sound pressure levels can be as intense as 100 to 140 dB SPL, levels that would impair auditory perception in other terrestrial mammals. We showed previously that bats exposed to intense wide-band sound (116 dB SPL) can navigate successfully through dense acoustic clutter. Here, we extend these results by quantifying performance of bats navigating through a cluttered scene after exposure to intense band-limited sounds (bandwidths 5–25 kHz, 123 dB SPL). Behavioral performance was not significantly affected by prior sound exposure, with the exception of one bat after exposure to one sound. Even in this outlying case, performance recovered rapidly, by 10 min post-exposure. Temporal patterning of biosonar emissions during successful flights showed that bats maintained their individual strategies for navigating through the cluttered scene before and after exposures. In unsuccessful flights, interpulse intervals were skewed towards shorter values, suggesting a shift in strategy for solving the task rather than a hearing impairment. Results confirm previous findings that big brown bats are not as susceptible to noise-induced perceptual impairments as are other terrestrial mammals exposed to sounds of similar intensity and bandwidth.

Reconstruction of acoustic scenes in biosonar

Biosonar mechanisms are highlighted by comparing acoustically reconstructed scenes derived from generalized methods with the performance of echolocating animals. Bioinspired computations replacing standard methods then offer a path towards understanding the animal’s solutions. Seemingly simple outdoor spaces present flying bats with complex scenes when vegetation and the ground are factored in, and even sound-treated research spaces such as flight rooms evolve into complex scenes as transmitted sounds propagate through the space. Underwater scenes, particularly for shallow water, are much more complicated because longer propagation distances create multipath reverberation that often overlaps with echoes. Interpulse intervals of biosonar emissions must be short to support rapid updating of tracking and perception, but reverberation means that pulse-echo ambiguity occurs. Two reconstructive methods are useful—the HARPEX method for visualizing the progression of echoes and reverberation following a sound, and a HARPEX-like method reconfigured as a forward-looking sonar. Acoustic datasets visualize airborne and underwater sonar scenes by reconstructing sound images for successive time frames following a transmitted sound using a tetrahedral soundfield microphone or a forward-looking sonar head. The complex acoustics of vegetation are examined with a simulation model built to understand the bats’ perception of vegetation. (Work supported by ONR.)

Ultra high definition 3D tracking of biosonar of Amazon River dolphins and others mammals into the wild

We designed and built a high capacity digital analog converter : JASON Qualilife DAQ [1] which is capable of sampling at 1 MHz on 5 channels simultaneously to assess cetacean biosonar and the source position and orientation relative to the array using time delay of arrivals of the biosonar acoustical signals. We used JASON with underwater portable arrays housing between 3 and 7 hydrophones (for more than 4 hydrophones we connected two cards sharing one common hydrophone). This has been used to monitor Physeter m. biosonar in near field, and Inia. g. in the Peruvian Amazon from 2014 to 2018. Results demonstrate the efficiency of this advanced low cost scientific instrumentation : it allows with only 4 hydrophones to track in 3D the precise movements of the I. g. dolphin, and to determine key behavioral features, like the velocity of its rostrum rotation, while following its highly defined biosonar emissions. This is the first, at our knowledge, that wild amazon dolphin biosonar is described inside its ecosystem. Perspectives on automatic low power trigger [2] for counting of the dolphins and extraction of individual acoustic invariant are given, as well as research avenues on biosonar of other species. [1] Gies et al, JASON Qualilife Daq, in DCLDE2018, smiot.univ-tln.fr,sabiod.org/bib [2] Fourniol et al., Low-Power Frequency Trigger for Environmental Internet of Things, in IEEE/ASME Mechatronic & Embedded Systems http://sabiod.org/bib, 2018 Samples and supp. material: http://sabiod.org/JASON.

Dynamic Echo Information Guides Flight in the Big Brown Bat.

Animals rely on sensory feedback from their environment to guide locomotion. For instance, visually guided animals use patterns of optic flow to control their velocity and to estimate their distance to objects (e.g., Srinivasan et al., 1991, 1996). In this study, we investigated how acoustic information guides locomotion of animals that use hearing as a primary sensory modality to orient and navigate in the dark, where visual information is unavailable. We studied flight and echolocation behaviors of big brown bats as they flew under infrared illumination through a corridor with walls constructed from a series of individual vertical wooden poles. The spacing between poles on opposite walls of the corridor was experimentally manipulated to create dense/sparse and balanced/imbalanced spatial structure. The bats’ flight trajectories and echolocation signals were recorded with high-speed infrared motion-capture cameras and ultrasound microphones, respectively. As bats flew through the corridor, successive biosonar emissions returned cascades of echoes from the walls of the corridor. The bats flew through the center of the corridor when the pole spacing on opposite walls was balanced and closer to the side with wider pole spacing when opposite walls had an imbalanced density. Moreover, bats produced shorter duration echolocation calls when they flew through corridors with smaller spacing between poles, suggesting that clutter density influences features of the bat’s sonar signals. Flight speed and echolocation call rate did not, however, vary with dense and sparse spacing between the poles forming the corridor walls. Overall, these data demonstrate that bats adapt their flight and echolocation behavior dynamically when flying through acoustically complex environments.

Interplay of lancet furrows and shape change in the horseshoe bat noseleaf.

Horseshoe bats emit biosonar pulses through the nostrils and diffract the outgoing ultrasonic pulses with baffles, so-called "noseleaves," that surround the nostrils. The noseleaves have complex static geometries and can furthermore undergo dynamic shape changes during emission of the biosonar pulses. The posterior noseleaf part, the lancet, has been shown to carry out anterior-posterior flicking motions during biosonar emissions with average lancet tip displacements of about 1 mm. Here, the acoustic effects of the interplay between the lancet furrows and shape change (lancet rotation) on the emission beam were investigated using the animated digital models obtained from the noseleaves of greater horseshoe bats (Rhinolophus ferrumequinum). It was found that forward lancet rotations increase the amount of sound energy allocated to secondary amplitude maxima (sidelobes) in the beampattern, but only in the presence of the furrows. The interaction between static and dynamic features can be readily quantified by roughness (standard deviation about local mean) of the amplitude distribution of the beampatterns. This effect goes beyond the static impact of the furrows on the width of the mainlobe. It could allow the bats to send out their pulses through a sequence of qualitatively different beampatterns.

Differences in foraging activity of deep sea diving odontocetes in the Ligurian Sea as determined by passive acoustic recorders

Characterizing the trophic roles of deep-diving odontocete species and how they vary in space and time is challenged by our ability to observe foraging behavior. Though sampling methods are limited, foraging activity of deep-diving odontocetes can be monitored by recording their biosonar emissions. Daily occurrence of echolocation clicks was monitored acoustically for five months (July–December 2011) in the Ligurian Sea (Mediterranean Sea) using five passive acoustic recorders. Detected odontocetes included Cuvier's beaked whales (Zipuhius cavirostris), sperm whales (Physeter macrocephalus), Risso's dolphins (Grampus griseus), and long-finned pilot whales (Globicephala melas). The results indicated that the foraging strategies varied significantly over time, with sperm whales switching to nocturnal foraging in late September whereas Risso's dolphins and pilot whales foraged mainly at night throughout the sampling period. In the study area, winter nights are about five hours longer than summer nights and an analysis showed that pilot whales and Risso's dolphins adjusted their foraging activity with the length of the night, foraging longer during the longer winter nights. This is the first study to show that marine mammals exhibit diurnal foraging patterns closely correlated to sunrise and sunset.

Probabilistic strategies for dynamic coordination of biosonar emissions in social groups

Echolocating bats are social animals and must be able to use their biosonar capabilities in a wide range of social contexts. Bats roosting or flying in groups routinely adapt their pulse emissions to accommodate sharing the acoustic space with potentially many bats. In this regard, echolocation may be bound by the same rules and constraints governing social communication. Bats can alter pulse acoustics, temporal patterns, projections patterns, and locomotor trajectories to minimize acoustic interference across individuals. Here, we present evidence that the highly gregarious free-tailed bat, Tadarida brasiliensis, primarily relies upon probabilistic strategies for mitigating acoustic interference rather than trying to predict or extract information from the pulses of other bats. When free-tailed bats hear the pulses of another bat they shift the timing of subsequent pulse emissions following a randomized back-off algorithm similar to those that have evolved in artificial wireless communications systems. I...

Detection and tracking of fluttering moths by echolocating bats

Aerial-hawking echolocating bats present an interesting model for studying how a moving listener interacts with moving sound sources. During foraging, relative movements between the bat and its prey introduce echo variation, which is further influenced by the timing of biosonar emissions with respect to insect wingbeats. Through systematic characterization of echoes from fluttering moths, this study aims at understanding how intermittent and highly variable echo information may be perceived and integrated by foraging bats for prey detection, tracking, and discrimination. Calibrated broadband echoes from frequency-modulated linear sonar sweeps were measured from live, fluttering moths of different morphological features with concurrent high-speed video recordings. The measurements are used to estimate the probability distribution of echo amplitudes as a function of moth morphology, wingbeat phase, and body orientation, as well as to predict changes of echo characteristics as the bat approaches the prey. Signal detection theory is then applied to model the bat’s perception of prey presence as influenced by changes in biosonar emission rate in different phases of the foraging process. These results can be further integrated with neurophysiological findings on the bat’s internal representation of space for a better understanding of this agile and efficient biosonar system.

Active Control of Acoustic Field-of-View in a Biosonar System

Author SummaryMost sensory systems have an active component, i.e. driven by an animal's behavior, which contributes directly to its perception. For example, eye movements are important for visual perception, sniffs are crucial for olfactory percepts, and finger movements for touch percepts. A classic example of an active-sensing system is bat echolocation, or biosonar. Echolocating bats actively emit the energy with which they probe their surroundings, and they can control many aspects of sensory acquisition, such as the temporal or spectral resolution of their signals. A key open question in bat echolocation concerns bats' ability to actively change the area scanned by their emitted beam. Here, we used a large microphone array to study the echolocation behavior of Egyptian fruit bats. We found that these bats apply a new strategy to alter the area scanned by their beam; specifically, bats changed their acoustic field-of-view by changing the direction of consecutively emitted beams. Importantly, they did so in an environment-dependent manner, increasing the scanned area more when there were more objects in their surroundings. They also increased their field-of-view when approaching a target. These findings provide the first example for active changes in sensing volume, which occur in response to changes in environmental complexity and target-distance, and they suggest that active sensing of space is more flexible than previously thought.

Dynamics of dolphin biosonar search and detection

The biosonar emissions of a free-swimming dolphin were measured with respect to direction and timing while the animal searched for and reported objects at various ranges and bearings from a test enclosure. The technology allows a quantitative description of the pulse-by-pulse scan pattern of the freely moving animal and subsequent inferences about search strategies, echo-integration time, functional beam coverage, and learning effects on biosonar performance. Presently, the methods are being extended to test an animal from a boat in the open ocean with measurement of the animal’s attitude and movement in three dimensions as well as the spectral content of emitted pulses during search, detection, and discrimination. The technique combines standardized test paradigms, and instrumentation pack carried by the animal, and a positioning system. The animal’s pack is a self-contained, data-acquisition system with sensors for magnetic azimuth, acoustic emissions, attitude, and motion. It is based on a high-speed, real-time microprocessor with open software architecture for maximum flexibility. The positioning system employs a high-accuracy real-time differential GPS for prepositioning targets and monitoring the position of the boat and animal during a search.

Animal Sonar: Processes and Performance

<i>Animal Sonar: Processes and Performance</i>

Topographic representation of vocal frequency demonstrated by microstimulation of anterior cingulate cortex in the echolocating bat, Pteronotus parnelli parnelli.

1. A midline region of brain dorsal and anterior to the corpus callosum, presumably anterior cingulate cortex, has been explored for its role in the production of vocalization in the mustached bat, Pteronotus p. parnelli. 2. Vocalizations elicited by microstimulation were virtually indistinguishable from natural biosonar sounds. The spectral content, relative intensity of harmonic components, and durations of emitted pulses are comparable to spontaneous emissions. 3. The frequencies of elicited vocalizations were within the range typically used by the mustached bat during Doppler-shift compensation. The frequency of the second-harmonic constant-frequency component (CF2) covered the range from 57-62 kHz, but was most commonly emitted at frequencies of 59-61 kHz. 4. An increase in the frequency of vocalizations over a number of consecutive pulses towards a steady-state plateau is evident in both spontaneous vocalizations and emissions elicited by microstimulation just above threshold. Increasing the stimulus intensity caused the frequency of emissions to approach the steady state more rapidly. 5. The anterior cingulate cortex appears to be organized topographically for increasing frequency of elicited biosonar sounds along a rostrocaudal axis. The area from which biosonar emissions were elicited was overrepresented for a 2 kHz band of frequencies just below the bats' CF2 resting frequency. Audible vocalizations with a complex spectrum resembling social cries can also be elicited by microstimulation, but only in an area that is adjacent and posterior to the biosonar region. 6. Some examples of both elicited and spontaneous vocalizations contained a relative intensity pattern of the harmonic components which deviated from the typical pattern. This suggests that mustached bats are capable of actively altering the spectrum of their pulses to subserve different tasks in echolocation.