After an echolocating bat locks on to an insect with its sonar beam, it can keep track of its prey despite receiving a slew of echoes from other objects—leaves, vines and so on. How does it separate echoes bouncing off its target from echoes bouncing off the surrounding clutter, especially when the echoes reach the bat at the same time?

The key, according to a new study of echolocation in the big brown bat (Eptesicus fuscus), is that objects in a bat's sonar beam produce echoes of a different character depending on where they fall within the beam. The bat can focus on the echoes from the center of the beam, where their target lies, and discount those from clutter on the periphery. The study appears in the July 29 issue of Science.

The distinction is enabled by the fact that the bats' sonar pulses have two distinct components, or harmonics, at different frequency levels. The higher-frequency harmonic forms a narrower beam than the widespread low harmonic, so central targets receive and reflect both harmonics in roughly equal measure. Off-target objects, on the other hand, fall outside the narrower beam of the high harmonic and thus reflect proportionally more of the low-frequency sounds. The harmonic structure also assists in isolating insect targets from background reflections—higher-frequency sounds diminish more quickly in air, so the high harmonic returns to the bat more weakly when reflected off of distant objects.

The reflected mixture of the two harmonics allows the bat to focus on what it is targeting—the echoes that come back with both harmonics intact. "It's not something that they actively do, it's really something more to do with how their auditory system perceives different kinds of echoes," says lead study author Mary Bates, who recently received her PhD from Brown University and is now pursuing a career as a science writer. (Bates has contributed to Scientific American in the past.) "Echoes from farther away or off to the side are not going to be perceived as well and are not going to interfere as much with the target that's being tracked," she says.

Bates and her co-authors, James Simmons of Brown and Tengiz Zorikov of Georgian Technical University's Institute of Cybernetics in Tbilisi, used an experimental setup with a live bat at the foot of a Y-shaped platform. The bats were trained to respond to a target echo from one arm of the platform and to avoid a "clutter" echo from the other arm. Microphones near the bat and loudspeakers at the ends of the Y could be used to manipulate the echo, either by filtering out one or the other harmonic or by artificially delaying the echo of one harmonic relative to another.

By so doing, the researchers could isolate the importance of the harmonic structure in the bats' separation of target from clutter. "It's really looking at the shape of the sound that it emitted and the shape of the echo that it receives," Bates says.

The researchers also tested how harmonics play into a facet of the bats' signal-processing machinery known as amplitude-latency trading. Auditory neurons in the big brown bat that process echolocation signals respond more slowly to diminished echoes. "When an echo comes back at a lower amplitude, the delay of the neuron firing is increased," Bates says. In the case of a two-harmonic chirp, amplitude-latency trading comes into play when a single echo contains two harmonics with different amplitudes—for instance, when a clutter object returns an echo with a diminished high harmonic. The experiments indicate that amplitude-latency trading for such sounds hampers the bat's depth perception, effectively defocusing the bat's sonar image of the reflecting object. In other words, harmonics allow a bat to produce a focused sonar image of the target object and a defocused image of the surrounding clutter.

The effect is not unlike the difference between central and peripheral vision in humans. "This research suggests that the sonar image that the bat perceives is sharpest along the central axis and that objects on the periphery may be blurred," says Cynthia Moss, a neuroscientist at the University of Maryland, College Park, who studies echolocation in bats. She points to a recent paper in Nature Neuroscience that reports a similar effect for sound localization in owls, which helps the birds track their prey. (Scientific American is part of Nature Publishing Group.) Owls can accurately track the source of sounds along their line of sight, but only at the expense of perception of sounds coming from the periphery, according to that study. "It may be a common perceptual strategy that operates across species and across modalities," Moss says.

3 Comments

1. 1. NatureTM 09:29 PM 7/28/11

   Very interesting! I bet the method the bats use would work well in robotics. Cancelling out noise can require some heavy DSP.

   Reply | Report Abuse | Link to this

2. 2. Bill Crofut 01:20 PM 7/30/11

   It would be interesting to know the evolutionary explanation of bat sonar. Even more interesting would be the evolutionary history of the bat. One report, purporting to do that, contains much descriptive information regarding the features of bats, but nothing that seems to fulfill the promise inferred from the title:
   
   "Despite many new discoveries about the rise of bats, mysteries remain. Bat ancestors must have existed prior to the Eocene, but we have no fossil record of them. Likewise, the identity of the closest relatives of bats is still unknown."
   [Nancy B. Simmons. Taking Wing: Uncovering the Evolutionary Origins of Bats.
   http://www.sciam.com/article.cfm?id=uncovering-bat-evolutionary-origins]
   

   Reply | Report Abuse | Link to this

3. 3. peteschramm 08:02 PM 8/1/11

   I would like to know if this strategy might also apply to dolphins. They would experience the same need for distinguishing targets from the background as well as frequencies from other dolphins in a pod.

   Reply | Report Abuse | Link to this