If mice could talk and sing… (Well, maybe they can)

Maybe they can: A neurobiologist at Duke explores their ability for vocal learning

CorrespondentDecember 23, 2012 

  • Know your scientist Name: Erich Jarvis Education: B.A., biology and mathematics, Hunter College; Ph.D., neurobiology and animal behavior, Rockefeller University. At Rockefeller, he did graduate and postdoctoral work in the lab of Fernando Nottebohm, “who pioneered research on the neurobiology of song-learning in birds as a model for understanding neural plasticity in the adult brain,” according to the Howard Hughes Medical Institute. Background: Graduated from the High School of Performing Arts in New York City (on which the movie “Fame” was based) and was invited to audition for the Alvin Ailey American Dance Theater.

It’s a pretty common scene in a bar or club on any given Friday or Saturday night. One, maybe two, guys are eagerly chatting-up the hot girl with the cocktail in her hand. They’re trying their best to be witty, to be charming, to do anything to potentially win her affections – even using each other’s best lines.

Until now, however, no one considered that mice might be doing the same thing. But they could be. In fact, they’re probably more aggressive about it. Male mice don’t stop at pleasant conversation. They’ll chase a female while singing to her and trying to smell her.

Strange as it all may sound – and it does defy conventionally held beliefs about the ability of mice to “talk” – Erich Jarvis, Ph.D., neurobiology associate professor at Duke University, and his team have discovered that mice may have some of the same brain features that humans and songbirds use for vocalizations and pitch changes.

“It’s accepted dogma that humans and songbirds are the only beings that have the four brain areas needed to produce vocalizations,” said Jarvis, who is also a Howard Hughes Medical Institute investigator. “Based on our research, we believe mice have more limited versions of these behavior and brain traits.”

The HHMI, National Science Foundation, and the National Institutes of Health funded Jarvis’ research.

The findings indicate that male mice may be able to learn how to change their vocalizations to match another male mouse. If correct, scientists may be forced to reconsider a belief they’ve held for 60 years – that vocal learning is unique to humans and a small cadre of songbirds.

The experiment

With his former grad student Gustavo Arriaga, Jarvis used gene expression markers, which lit up neurons in the motor cortex – the part of the brain involved in planning, control and voluntary movement – of each mouse’s brain while they sang. When these song-specific neurons were damaged, the mice couldn’t keep their songs on-pitch or consistently repeat them, verifying their connection to vocalization.

In addition to the markers, the team injected a tracer to map the signals that controlled the songs as they migrated from neurons in the motor cortex to the brainstem and on to the larynx muscles. According to Jarvis, this direct channeling from motor cortex to larynx was the biggest surprise and puts into question whether these projections in mice work the same way as in humans and birds: Can mice learn vocalizations?

To make this determination, the team first had to record the sounds mice make. They placed a pair of male mice in a cage with a single female to prompt communication and used a 4-inch high-sensitivity microphone to capture the sounds. The powerful microphone was necessary because mice “speak” at a frequency between 30 and 40 KHz – too high-pitched for humans to detect. Humans hear sounds between 14 and 15 KHz.

Jarvis’s team monitored 12 pairs of male mice over an 8-week period to see whether they began to imitate each other or the pitch of their songs converged. They conducted the experiment twice, and by the eighth week, he said, the less-dominant mouse had modified its song to emulate the dominant male.

By digitally modulating the recordings to a frequency audible to humans, the investigators demonstrated that, by the end of the experiment, the male mice had virtually identical songs.

“The mice were changing their pitch so the smaller animal matched the song of the larger male,” Jarvis said. “This is a simple form of imitation – it’s pitch. Until now, it was thought that they didn’t have this ability for vocal learning.”

The controversy

Not everyone agrees with these findings, however. Kurt Hammerschmidt, a vocalization expert at the German Primate Center, is reticent to fully accept that mice can be true vocal learners. The problem, he said, is that Jarvis’ team simply didn’t analyze a large enough number of mice.

“Fewer animals is OK in neurobiological studies because we know anatomical structures found in one animal are also present in other animals,” Hammerschmidt said. “But with behavioral studies, we need more animals to look at motivation, arousal and experience.”

Hammerschmidt and other scientists worldwide have conducted experiments similar to Jarvis’s and have not replicated his findings. Hammerschmidt also disagrees that pitch convergence alone indicates that mice are vocal learners.

“All other studies focused on male courtship songs failed to find any evidence that learning is involved in the development of these vocalizations,” he wrote. “None of all other terrestrial mammals, except humans, are able to produce new sounds.”

Future implications

Additional research is needed, Hammerschmidt said, to verify whether Jarvis’s findings are correct.

If Jarvis and his team are correct, though, these findings could impact both science and health care.

Although mice don’t have the same speaking ability as humans have, understanding their potential capacity for vocal learning could shed light on how speech works in people, as well. It could open the door for further research into the brain’s circuitry and the basic principles of speech, Jarvis said.

The greatest impact of this research, however, could be its effect on neurological disease, he said. In particular, autism is the brain disorder that has the biggest impact on speech, and the NIH and Congress have invested millions to study the causes and biological makeup of this condition. In these studies, investigators have been able to take the gene variant from a child with autism and put it into the mouse genome, but they’ve been unable to pinpoint which area of the brain is affected. This research eliminates that limitation.

“Now, we have the brain pathway for them to look and play around with,” Jarvis said. “It could open the door for some gene drug therapy on this part of the brain or help determine how we can affect the whole system.”


News & Observer is pleased to provide this opportunity to share information, experiences and observations about what's in the news. Some of the comments may be reprinted elsewhere in the site or in the newspaper. We encourage lively, open debate on the issues of the day, and ask that you refrain from profanity, hate speech, personal comments and remarks that are off point. Thank you for taking the time to offer your thoughts.

Commenting FAQs | Terms of Service