Feature

During spring and summer in Florida's Apalachicola National Forest, a chorus of barking tree frogs fills the night air. At sunset, groups of up to 150 male frogs gather in ponds. Then the females emerge from the forest, listen to the males' barks and choose their mates.

James Madison University behavioral ecologist Christopher Murphy, PhD, has spent years listening to those sounds as part of his work to determine how female tree frogs pick the male they want out of the intermingled croaks. Behavioral ecologists and comparative psychologists have long been interested in similar questions of how animals in large groups communicate with one another. After all, many live and communicate in groups, posing special challenges for the animals, be they baby birds trying to make their hungry squawks heard above their siblings or singing crickets competing for mates.

But for years, the daunting technical challenges involved in recording and analyzing such complicated sounds made it difficult for scientists to study what was happening in animal groups. Instead, researchers tended to concentrate on simpler questions of how one animal would communicate with another.

Now, new technologies—including microphone arrays originally developed for military applications and computer software used by musicians—are helping scientists study acoustic communication in animal groups.

"As the technology has evolved, really sophisticated technology can be brought to bear to address questions people had years ago, but they didn't have a good way of collecting raw data," says Joshua Schwartz, PhD, a Pace University biologist who studies animal communication.

And in many cases, the data are revealing that the animals' communication strategies are even more sophisticated than anyone expected.

Tricky suitors

Female tree frogs need to pick out a male with a strong bark but also a nearby location, because the female is vulnerable to predators as she hops out of the forest to find her mate, says Murphy. But how can a female frog tell whether a weak bark is coming from a wimpy mate that's close by or a hunky frog that's far away?

In a study published in the Journal of Comparative Psychology (Vol. 122, No. 3), Murphy tested three hypotheses. The first—that the females simply judged the loudest croaks to come from the closest males—was a likely mechanism, but one that might produce errors if a more distant male had a louder croak than a closer male. The second hypothesis was that the females might use the degradation of the sound of the bark (the fact that sound gets "fuzzier" as its source gets farther away) to estimate location. The third was that the frogs might use a sound gradient—the fact that the rate at which a sound gets louder increases as you approach its source.

To test the sound-degradation hypothesis, Murphy set up a simulated tree frog chorus created through an array of eight speakers that croaked in synthesized male bullfrogs' tones. To play back the calls, he used software developed for recording and playing back multiple music tracks in concert settings.

His previous research had shown that female frogs prefer barks with a pitch around 450 hz, but they will choose a less attractive, higher-pitched male if the attractive one is too far away. So the researcher set six of the speakers to play an unattractively shrill bark and two to play an attractive, dulcet-toned bark.

Murphy placed a female frog 15 meters away from one of the attractive barking speakers and 18 meters from the other attractive bark. But to trick the female, the calls played from the closer speaker had originally been recorded at a distance of four meters from the source, and the call from the farther speaker had originally been recorded at a distance of one meter—so the overall degradation level of the sound by the time it reached the female was the same from both speakers.

Given that research trickery, if the females relied on sound degradation to estimate distances, they would randomly choose which speaker to move toward. Instead, they picked the closer speaker in 17 of 20 trials.

In a similar test in which he incrementally adjusted the volume of the speakers as the females hopped closer to them, Murphy found that the frogs didn't rely on the sound gradient, either. Even with his state-of-the-art speaker array, Murphy could not fool the female frogs, he found.

He now suspects that the females use a more complex method called triangulation to estimate distance, which would rely on the frog's being able to analyze how quickly the direction the sound is coming from changes as she moves—a closer frog's call would change more quickly than a faraway potential mate.

Murphy says the results suggest that the tree frogs possess greater cognitive abilities than scientists thought: No animal other than humans has been shown to use sound triangulation to measure distance.

Buddy-up, bullfrogs

About 1,500 miles up the East Coast, in the woods of central Massachusetts, researchers Andrea Megela Simmons, PhD, and James Simmons, PhD, are also using new technology to study frog choruses. The couple, both Brown University professors, are experts on animal acoustic communication. Several years ago, they were working on a project with the Defense Advanced Research Projects Agency, developing specialized, sensitive microphone arrays for battlefield applications. In the course of that research, James Simmons says, "we realized that bullfrog mating calls, when slowed down, sound like a vehicle moving across a battlefield."

With Mary Bates, a Brown psychology graduate student, they transferred some of their testing from the Aberdeen Proving Grounds in Maryland to a Massachusetts pond, employing microphone arrays to answer a long-running question in bullfrog research: why the male bullfrogs so often tended to call overtop of one another.

"Sometimes they're polite animals who take their turns," says Andrea Simmons. "But as the season progresses, they become more and more prone to calling at the same time. This would make it hard for a female bullfrog to distinguish one male from another."

To figure out whether the frogs were interrupting one another on purpose, the researchers needed to record all of the calls and link them to individual frogs—a task that would be nearly impossible to do by simply observing the frogs. So instead, they used the device they developed for DARPA—two cubes, placed about 10 meters apart, each composed of multiple small microphones. By feeding the data into a software program that analyzed the time at which the sound from each individual frog call arrived at each cube, the researchers could map each frog's location.

That map suggested that the frogs didn't space themselves randomly in the pond. Instead, they grouped together in clusters of four to six frogs. Andrea Simmons suspects that the clusters are the males' attempt to form a "superfrog" with a more powerful voice than one frog alone that could attract more females than one individual frog, benefiting each male. Her first paper on the subject only proved that the frogs do cluster into overlapping callers; she's currently doing further research to try to prove the "superfrog" theory.

Both frog studies illustrate a trend in animal communications research, says University of Tennessee, Knoxville, psychologist Todd Freeberg, PhD, who studies communication in Carolina chickadees.

"Twenty years ago, to bring multiple laptops into the field and multiple microphone arrays would have been prohibitively expensive and incredibly complex," he says. "But now it's possible."

And the technological advances are opening up more and more research possibilities as researchers are able to record and analyze ever-more-complicated patterns of chirps, croaks and barks.


Lea Winerman is a writer in Washington, D.C.