ECE professor studies frog chorusing behavior in the wild

10/23/2007 Lauren Eichmann

ECE Professor Doug Jones and Rama Ratnam, assistant professor in neurobiology at the University of Texas at San Antonio, have developed technology that further enables biologists to study frog chorusing behavior.

Written by Lauren Eichmann

ECE Professor Douglas L Jones and Rama Ratnam, assistant professor in neurobiology at the University of Texas at San Antonio, have developed technology that further enables biologists to study frog chorusing behavior.

Jones said frogs call for two reasons: so females can find males, and for males to define their territory and threaten other males. “While biologists have been able to study the interaction between small numbers of frogs, they have not been able to reliably study the interaction between large numbers of frogs spread out over many hundred square meters,” said Jones. “Ultimately the biologists want to understand frog chorusing behaviors. It sounds simple, but it’s usually complex. Do they alternate; do they take turns [chirping]? How do they adjust their calling in the middle of a group when many frogs are calling at once?”

Rama Ratnam, assistant professor in neurobiology at the University of Texas at San Antonio and Illinois alumnus, and ECE Professor Doug Jones spent a week at the Cibolo Nature Center in Bourne, Texas, testing equipment they developed to enable biologists to study frog chorusing behavior.
Rama Ratnam, assistant professor in neurobiology at the University of Texas at San Antonio and Illinois alumnus, and ECE Professor Doug Jones spent a week at the Cibolo Nature Center in Bourne, Texas, testing equipment they developed to enable biologists to study frog chorusing behavior.

Ratnam said most researchers find it difficult to locate and extract the sound from individual frogs since it is impossible to do visually or manually. “Recent advances in technology can solve this problem, but most of the biologists studying frog chorusing are not engineers,” he said. “It is not something most biologists can solve with conventional methods.”

According to Ratnam, the new technology blends the two disciplines — engineering and biology — to help form a cohesive answer to some pressing questions regarding frog chorusing behavior. Although Jones is an engineer, he has an interest in the biological aspects of hearing in noise, and Ratnam’s interests also lie in the signal processing aspects of hearing.

“How we are able to attend to one person talking when there are so many competing sounds is not known,” said Ratnam. “In fact, how the brain is able to unscramble the sounds and extract the voice of one person from all the background noise is one of the major research areas in auditory neurobiology. This phenomenon is called the cocktail-party effect. Female frogs trying to select mates from a chorus face the same situation. It turns out that many of the basic pathways for processing sounds in the frog's brain are similar to those of humans — and indeed to those of other vertebrates. Thus they are a useful model system for studying the human cocktail-party problem.”

Ratnam said he and other neurobiologists are interested in knowing why people with hearing loss perform poorly in cocktail-party situations. Understanding hearing in noisy environments by probing the neural mechanisms, particularly when there are many competing sounds — or talkers — in the background, may help them answer some of these questions, he said.

To test the equipment they developed, Jones and Ratnam spent a week in March at the Cibolo Nature Center in Bourne, Texas, where they set up a small camp. The location of the study, near SanAntonio, was essentially a seven-meter square area that straddled a wetland. Four large posts, each equipped with a microphone, were placed around the square study site while one post was strategically placed in the center and contained a microphone array of four directional devices at the same location.

During their field study, signal processing techniques allowed the data from the microphones to identify where and when any frogs were chirping. Each sound clip could be transferred to a computer screen via a ‘sound movie’ in which small boxes showed where frogs were visually spotted. These boxes helped to confirm the sound algorithm was correct in pinpointing location. Actual sound recordings from the microphones simultaneously visually recognized where and to what extent the amphibians were calling.

Small stakes along the perimeter of the site helped Jones and Ratnam visually confirm the location of each frog with what the signal processing techniques indicated. “[In the sound movie] you see where the frogs are without actually having a camera, and you can connect the calling to that,” said Jones. “So we’re basically seeing them with their sound.”

People have used microphone arrays for a number of years to study frog chorusing behavior, but it has not been completely reliable, said Jones. “There were so many more frogs than the practical number of microphones, that the previous signal processing methods have not been able to figure out where the frogs are,” Jones said. “So if we have two or more microphones we can start to get some idea by looking at the differences of that signal as it reaches the different microphones. We can start to figure out precisely where the sounds are coming from.”

Jones said it can still be difficult to deal with more than two sound sources at once. “But we developed techniques based on the way that the brain works for being able to find a lot of sources with only a few microphones.”

The Gulf Coast Toads and cricket frogs, the primary subjects of the study, are active for several hours after dark until as late as dawn, said Jones. “You can’t just observe them, since they come out at night and it’s dark. They don’t like any lights.” Jones and Ratnam used only small flashlights to confirm frog positions with their data.

From the field study, Jones said the cricket frogs — which are very tiny — tended to call simultaneously with short chirps, as it was observed via the sound movie. The Gulf Coast Toad, in comparison, had a rather long, isolated call.

Jones became interested in what he considers some fundamental scientific questions about frog behavior ecology after he met Ratnam. They were both working on an advanced hearing aid project at the Beckman Institute for Advanced Science and Technology at Illinois and had been talking about the possibility of the frog chorusing study for two or three years.

“We’ve developed techniques [at Beckman] that were originally aimed at hearing aids and similar applications, but based on biological inspiration from the animals,” said Jones. “So in some sense we’re taking these techniques back to the domain of the animals again, and in another sense trying to figure out what they’re trying to figure out in the environment.”

Ratnam said engineers have not succeeded in using conventional signal processing technology to improve the design of hearing aids in cocktail-party situations. “A different kind of solution is needed,” he said. “Here is where people like Doug have made a real contribution. Using neurobiological strategies underlying normal hearing, Doug and other people at Illinois were able to design a hearing aid that selectively picked out the voice of one person and suppressed the background noise to levels that were sufficient to improve hearing performance.”

According to Jones, their goal with the frog study is to build a permanent automated data analysis system and study site to record every night, all night long. They also plan to advance the technology so biologists and researchers can see the sound movies in real-time, and possibly even introduce their own frog sounds to experiments.

Jones and Ratnam hope to secure two permanent sites to study frogs near San Antonio; the Cibolo Nature Center, and a location in the protected wetlands of Mitchell Lake. A more controlled site, equipped with an observation room, is already established in Panama. Ratnam said they hope to model the two U.S. sites after the Panama location. According to Ratnam, Texas has a longer breeding period for frogs than other locations, which allows for an extended period of study. “For the first time we have the capabilities to really answer these questions with real frogs, in a real environment,” said Jones.

But funding is crucial and the equipment is fairly expensive, said Ratnam, who applied for a National Science Foundation Career Grant in July. It is a five-year grant that will help pay for equipment and travel expenses. He will likely hear if he received the grant this December.

Ratnam also plans to establish a workshop for frog chorusing, where researchers would study frogs in the wild once a year in the spring. The initiation of the workshop, however, is contingent upon receiving funding.

Ratnam received his bachelor’s degree in chemical engineering from the Indian Institute of Technology in Delhi, and PhD from Illinois in biophysics and computational biology. Jones earned his bachelor’s and master’s degrees, as well as his PhD, all in electrical engineering from Rice University. In addition to being a professor, Jones is currently a researcher with the Beckman Institute NeuroTech group. He works with faculty from various disciplines to develop computer-based systems that aid in the study of brain organization and function.


Share this story

This story was published October 23, 2007.