Benjamin Falk

Benjamin FalkJohns Hopkins University | JHU · Center for Imaging Science

· Doctor of Philosophy
  • About
    Research items
    Research Experience
    Aug 2007 - May 2015
    PhD Student
    University of Maryland, College Park · Program in Neuroscience and Cognitive Science
    College Park, United States
    Followers (60)
    View All
    Sharon Miriam Swartz
    Holger R Goerlitz
    Yossi Yovel
    Jenni A. Stanley
    Danish Shaikh
    Denis Medinas
    Lea R. Dougherty
    Vadim Kashtelyan
    Joxerra Aihartza
    Danuta Maria Wisniewska
    Following (65)
    View All
    Sharon Miriam Swartz
    Holger R Goerlitz
    Antton Alberdi
    Ostaizka Aizpurua
    Kenneth S Breuer
    Anders Hedenström
    Florian Titus Muijres
    Yossi Yovel
    Jenni A. Stanley
    William E Conner
    Research Items
    Echolocating bats often operate in the presence of conspecifics and in cluttered environments, which can be characterized as a "cocktail party nightmare." Each bat's sonar vocalization can result in an echo cascade from objects distributed in direction and range. Adding to the acoustic clutter are the signals from neighboring bats. Past studies demonstrate that bats adapt their echolocation to avoid signal jamming from conspecifics by adjusting the frequencies of their vocalizations, as well as going silent. When bats fly alone in densely cluttered environments, they adjust the frequencies of call pairs to disambiguate overlapping echo streams. How do echolocating bats adapt to both conspecific signals and clutter? We sought to answer this question by flying big brown bats in a large room equipped with high-speed video and audio recording equipment. In baseline trials, bats flew alone in an empty room and were later introduced to an artificial forest, first individually and later in pairs. The echolocation behavior and flight paths are analyzed to evaluate the spectro-temporal adjustments of bat calls and silent behavior as animals progressed from open room, to forest, to forest with conspecifics. The results shed light on how echolocating bats adapt to a "cocktail party nightmare."
    Bats are the only mammals capable of powered flight, and they perform impressive aerial maneuvers like tight turns, hovering, and perching upside down. The bat wing contains five digits, and its specialized membrane is covered with stiff, microscopically small, domed hairs. We provide here unique empirical evidence that the tactile receptors associated with these hairs are involved in sensorimotor flight control by providing aerodynamic feedback. We found that neurons in bat primary somatosensory cortex respond with directional sensitivity to stimulation of the wing hairs with low-speed airflow. Wing hairs mostly preferred reversed airflow, which occurs under flight conditions when the airflow separates and vortices form. This finding suggests that the hairs act as an array of sensors to monitor flight speed and/or airflow conditions that indicate stall. Depilation of different functional regions of the bats' wing membrane altered the flight behavior in obstacle avoidance tasks by reducing aerial maneuverability, as indicated by decreased turning angles and increased flight speed.
    INTRODUCTION Echolocating bats use biological sonar to determine the spatial location of objects in the environment. Central to an echolocating bat's spatial perception is directional control of its sonar signals with respect to objects in the environment. The extent to which bats of different species employ different sonar beam-directing behaviors to localize objects in the environment is the focus of this study. E. fuscus has been shown to aim the maximum intensity of its sonar vocalizations at insect prey as it prepares to intercept (Ghose and Moss, 2003). In a landing task investigated with the same microphone array, R. aegyptiacus placed the maximum slope of intensity of each click towards the landing target (Yovel et al, 2010). Do these differences in beam-directing behaviors of E. fuscus and R. aegyptiacus arise from the task (landing vs. insect capture) or their sonar production mechanisms (laryngeal vs. lingual) and structure (frequency modulated sweeps vs. clicks) of their echolocation calls? To begin to address these questions, we studied the beam directing behavior of E. fuscus as it performed in a landing task comparable to the one studied previously in R. aegyptiacus (Yovel et al., 2010). SUMMARY Within the set of frequencies measured by our microphone array, the bat's beam direction lies off-center of the target. Further analysis is needed to determine whether the bat is aiming the maximum slope of its sonar beam on the target or if the skew is due to some other cause. E. fuscus directs its sonar beam at the target from as far as 4 meters. This is in contrast to measurements of the sonar beam direction of E. fuscus when tracking insect prey, in which the tracking angle did not converge to the target until as close as 0.5 meters (Ghose and Moss, 2003). The landing target is much larger in comparison to an insect, returning a much stronger echo which is easier to localize. The accuracy with which E. fuscus directs its sonar beam towards the landing target is less than reported for insect capture. The flight maneuvers required for landing on the target are likely not as complex and might not require as accurate a sonar aim by the bat. The width of the sonar beam of E. fuscus has been measured to be 70° (Hartley and Suthers, 1989), so less accurate sonar tracking may still be sufficient for landing. FUTURE DIRECTIONS Use a wide-bandwidth microphone array to measure the sonar behavior of Eptesicus fuscus during landing and prey capture. Using these recordings, analyze the sonar directing behavior outside of the 28 to 42 kHz band previously reported. E. fuscus may use a different strategy at higher frequencies, including changing beam width or beam shape. REFERENCES Ghose, K., & Moss, C. F. (2003). The sonar beam pattern of a flying bat as it tracks tethered insects.
    Active-sensing systems abound in nature, but little is known about systematic strategies that are used by these systems to scan the environment. Here, we addressed this question by studying echolocating bats, animals that have the ability to point their biosonar beam to a confined region of space. We trained Egyptian fruit bats to land on a target, under conditions of varying levels of environmental complexity, and measured their echolocation and flight behavior. The bats modulated the intensity of their biosonar emissions, and the spatial region they sampled, in a task-dependant manner. We report here that Egyptian fruit bats selectively change the emission intensity and the angle between the beam axes of sequentially emitted clicks, according to the distance to the target, and depending on the level of environmental complexity. In so doing, they effectively adjusted the spatial sector sampled by a pair of clicks-the "field-of-view." We suggest that the exact point within the beam that is directed towards an object (e.g., the beam's peak, maximal slope, etc.) is influenced by three competing task demands: detection, localization, and angular scanning-where the third factor is modulated by field-of-view. Our results suggest that lingual echolocation (based on tongue clicks) is in fact much more sophisticated than previously believed. They also reveal a new parameter under active control in animal sonar-the angle between consecutive beams. Our findings suggest that acoustic scanning of space by mammals is highly flexible and modulated much more selectively than previously recognized.
    The distribution of pulse intervals is bi-modal and is similar across the different experimental settings. The two peaks in the histogram represent the inter-pair intervals (right peak) and intra-pair intervals (left peak). In the no-object experiments (light gray), the inter-pair intervals were slightly higher than in the other setups, and increased from ∼90-ms to ∼100-ms. (EPS)
    The inter-click angle increases when locking onto the target, in all individual bats. “B,” before locking; “A,” after locking. Error bars, mean ± s.e.m. Note also that the basal angle differs systematically between individual bats. “*” significant difference; individual significance values: p<0.005; p<0.02; p<0.001; p<0.001; p<0.05, respectively, for the five bats. (EPS)
    The inter-click angle increases with the increase in environmental complexity, in all individual bats. Notice that the basal angle differs between individual bats. “0,” no object; “1,” one object; “5,” multiple-object (five-objects) experiments. Error bars, mean ± s.e.m. (EPS)
    When animals move through the environment, they receive dynamic sensory information from surrounding objects. Past research has demonstrated that visually guided animals rely on optic flow to estimate their relative velocity and distance to objects. More recently, the flight and echolocation behavior of big brown bats was studied in a corridor constructed of poles of variable spacing. When the pole spacing on opposite walls was symmetric, animals centered themselves in the corridor, and when pole spacing was asymmetric, bats steered toward the less echoic wall. This finding raised the question whether bats adjusted their flight paths based on the flow of echoes returning from the corridor walls, or whether the animals steered away from the wall reflecting more intense echoes. To address this question, bats flew through the original corridor with the additional experimental condition of felt coverage on one corridor wall, reducing the reflectivity of pole echoes on that side, while retaining the flow pattern. If the bat’s flight path is influenced by the echo intensity profile of a corridor wall, it should deviate toward the felt-covered wall. By contrast, if the bat’s behavior is influenced by acoustic flow cues, flight paths should resemble those in the original (no-felt) conditions.
    The beam pattern of sonar signals emitted by echolocating animals, such as bats and toothed whales, directly influences the acoustic information available for guiding task-specific behaviors. The lingual echolocating bat, Rousettus aegyptiacus, emits broadband transient sonar clicks that resemble those of dolphins. The clicks are emitted in left-right pairs, with the maximum intensity slope of each signal pointing toward the target during navigation. However, detailed beam pattern characteristics of these lingual sonar clicks remain unknown. Using a loosely populated three-dimensional microphone array, we systematically characterize the multi-frequency beam structure of R. aegyptiacus in the entire azimuth-elevation domain. The bat’s head aim was recorded by an infrared high-speed motion-capture camera system. We show that the sonar beam of R. aegyptiacus tongue clicks is vertically elongated and exhibits an unusual multi-frequency structure that has not been described previously in the literature. Specifically, the high-intensity portion of the beam shifts medial in azimuth from the left and right click directions and downward in elevation with increasing frequency. Combined with close-up videos of mouth movement as the bats click, these results suggest a more sophisticated beam formation and steering mechanism than conventional simple aperture models.
    Echolocating bats face the challenge of coordinating flight kinematics with the production of echolocation signals used to guide navigation. Previous studies of bat flight have focused on kinematics of fruit and nectar-feeding bats, often in wind tunnels with limited maneuvering, and without analysis of echolocation behavior. In this study, we engaged insectivorous big brown bats in a task requiring simultaneous turning and climbing flight, and used synchronized high-speed motion-tracking cameras and audio recordings to quantify the animals' coordination of wing kinematics and echolocation. Bats varied flight speed, turn rate, climb rate and wingbeat rate as they navigated around obstacles, and they adapted their sonar signals in patterning, duration and frequency in relation to the timing of flight maneuvers. We found that bats timed the emission of sonar calls with the upstroke phase of the wingbeat cycle in straight flight, and that this relationship changed when bats turned to navigate obstacles. We also characterized the unsteadiness of climbing and turning flight, as well as the relationship between speed and kinematic parameters. Adaptations in the bats' echolocation call frequency suggest changes in beam width and sonar field of view in relation to obstacles and flight behavior. By characterizing flight and sonar behaviors in an insectivorous bat species, we find evidence of exquisitely tight coordination of sensory and motor systems for obstacle navigation and insect capture.
    Smooth spherical beads were used as S+. Differently textured beads were used as S-, each presented one at a time as an alternate to S+. Can bats integrate echoes over time to discriminate sonar targets? We used the same behavioral paradigm with newly designed stimuli. Smooth spherical Delrin balls were machined with a groove cut 1 mm in depth around their circumferences. Holes were drilled to hang them from the ceiling. Each bead was machined identically, but presented horizontally (S+) or vertically (S-). A random sampling of individual echoes with no regard to sequence or position makes the targets indistinguishable. Can the bats still discriminate the groove orientation when the beads are at different heights? S+ and S-were hung at different heights, randomized from trial to trial, to determine whether the bat could continue to discriminate the orientations. METHODS Experimental Paradigm We trained bats to discriminate between two tethered beads under low IR illumination, one designated the positive stimulus (S+) and the other the distracter (S-). The bats were trained to seek out and hit S+ suspended from the ceiling and avoid hitting S-, also suspended from the ceiling. The positions of the targets in the room changed with every trial, and the target positions were randomized. S+ and S-were suspended for the full duration of each trial. Training Using operant conditioning, the bats were trained in the following order, with some modifications per bat: 1. Train bat to catch mealworms from tether 2. Associate bridge stimulus with food reward while bat on platform 3. Allow bat to learn to land on platform 4. Train bat to connect physical contact with S+ (either on platform or hanging in the room) to the bridge stimulus and food reward on platform 5. Introduce S-to teach bat to prefer S+ and avoid S-Experiments were run in a large carpeted flight room (6.4 x 7.3 x 2.5m) lined with acoustic foam in low light, long wavelength conditions. Sonar vocalizations were recorded using ultrasonic microphones and analyzed off-line. The bat's 3-D flight path was reconstructed using stereo images taken from high-speed video recordings. An array of 16 microphones was used to record the directional aim of the sonar beam of the bat as it flew. Echo Recordings The stimuli were ensonified by playing computer generated FM sweeps and recording the returning echoes.
    Insectivorous bats rely on echolocation and other sensory modalities to perform complex flight behaviors as they track flying prey and maneuver around obstacles. Microscopic hairs on the bat’s wing contribute sensory information to adaptive flight behaviors. The wing hairs are hypothesized to provide the flying bat with sensory information about air currents across the wing membrane. Neurophysiological studies have shown that bat wing hairs are directionally sensitive to airflow, and behavioral studies have shown that wing hair removal results in a decrease in the bat’s flight maneuverability (Sterbing-D’Angelo et al., PNAS, 2011). Here we report data from a behavioral study in which bats encountered gusts of wind as they performed an obstacle avoidance and prey-capture task. Echolocation signals, flight trajectories and wing kinematic data in baseline trials were compared with data in experimental trials, in which hairs were removed from the trailing and leading edges of the bat wing. Wind gusts were introduced randomly from different directions as the bat maneuvered through a narrow opening in a net to gain access to a tethered insect one meter beyond the opening. The effects of air turbulence on the bat’s flight maneuverability and echolocation behavior were quantified with high-speed video and sound recordings. By studying the effects of air turbulence on wingbeat kinematics, flight path, and echolocation behavior in bats with and without wing hairs, we hope to learn more about the mechanisms by which mammals and other organisms process sensory information to adapt their behaviors.
    Echolocating bats employ active sensing as they emit sounds and listen to the returning echoes to probe their environment for navigation, obstacle avoidance, and pursuit of prey. The sensing behavior of bats includes the planning of 3D spatial trajectory paths, which are guided by echo information. In this study, we examined the relationship between active sonar sampling and flight motor output as bats changed environments from open space to an artificial forest in a laboratory flight room. Using high-speed video and audio recordings, we reconstructed and analyzed 3D flight trajectories, sonar beam aim and acoustic sonar emission patterns as the bats captured prey. We found that big brown bats adjusted their sonar call structure, temporal patterning, and flight speed in response to environmental change. The sonar beam aim of the bats predicted the flight turn rate in both the open room and the forest. However, the relationship between sonar beam aim and turn rate changed in the forest during the final stage of prey pursuit, during which the bat made shallower turns. We found flight stereotypy developed over multiple days in the forest, but did not find evidence for a reduction in active sonar sampling with experience. The temporal patterning of sonar sound groups was related to path planning around obstacles in the forest. Together, these results contribute to our understanding of how bats coordinate echolocation and flight behavior to represent and navigate their environment.
    Bats use sonar to identify and localize objects as they fly and navigate in the dark. They actively adjust the timing, intensity, and frequency content of their sonar signals in response to task demands. They also control the directional characteristics of their sonar vocalizations with respect to objects in the environment. Bats demonstrate highly maneuverable and agile flight, producing high turn rates and abrupt changes in speed, as they travel through the air to capture insects and avoid obstacles. Bats face the challenge of coordinating flight kinematics with sonar behavior, as they adapt to meet the varied demands of their environment. This thesis includes three studies, one on the comparison of flight and echolocation behavior between an open space and a complex environment, one on the coordination of flight and echolocation behavior during climbing and turning, and one on the flight kinematic changes that occur under wind gust conditions. In the first study, we found that bats adapt the structure of the sonar signals, temporal patterning, and flight speed in response to a change in their environment. We also found that flight stereotypy developed over time in the more complex environment, but not to the extent expected from previous studies of non-foraging bats. We found that the sonar beam aim of the bats predicted flight turn rate, and that the relationship changed as the bats reacted to the obstacles. In the second study, we characterized the coordination of flight and sonar behavior as bats made a steep climb and sharp turns while they navigated a net obstacle. We found the coordinated production of sonar pulses with the wingbeat phase became altered during navigation of tight turns. In the third study, we found that bats adapt wing kinematics to perform under wind gust conditions. By characterizing flight and sonar behaviors in an insectivorous bat species, we find evidence for tight coordination of sensory and motor systems for obstacle navigation and insect capture. Through these studies, we learn about the mechanisms by which mammals and other organisms process sensory information to adapt their behaviors.
    Top co-authors