What can singing mice say about human speech?

Sadie Harley

scientific editor

Meet our editorial team
Behind our editorial process

Robert Egan

associate editor

Meet our editorial team
Behind our editorial process
Editors' notes

This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

peer-reviewed publication

trusted source

proofread

The GIST
Add as preferred source

Speech is a crowning achievement of human evolution, the skill that separates us from every other animal. So, it would stand to reason that evolving this capability required some enormous leap in brain complexity. A study published in Nature suggests otherwise.

Alston's singing mouse (Scotinomys teguina), a small rodent from the cloud forests of Central America, produces loud, elaborate songs humans can hear across a room. These mice can sing solo but often perform rapid-fire duets with split-second timing. Among all mammals, it's one of the closest parallels to the turn-taking of human conversations.

A team at Cold Spring Harbor Laboratory (CSHL) wanted to know what changed inside this animal's brain to make singing possible. The answer was surprisingly simple. The singing mouse didn't evolve a bigger brain, new brain regions, or new categories of neural connections.

Instead, evolution roughly tripled the number of neurons that connect the brain's mouth-movement control center with just two target regions. One is the cortex that controls hearing. The other is a midbrain structure that controls vocalizations for a variety of species, including humans. The rest of the brain wiring is essentially identical to that of an ordinary lab mouse!

Emily Isko, a grad student in the Banerjee lab, traced the differences using a molecular barcoding technique developed at CSHL by Professor Anthony Zador. This allowed the team to map thousands of individual cells across the whole brain.

"When you look at singing mice and lab mice side by side, their brains are almost indistinguishable," Isko said. "The differences only show up when you trace where individual neurons send signals."

"You might expect that evolving a whole new means of vocal communication would require a significant reorganization of brain circuitry," said Associate Professor Arkarup Banerjee.

"Instead, we found a couple of targeted changes to existing wiring patterns. Our approach gives the field a playbook. To understand how new behaviors evolve, find closely related species with big behavioral differences and start by mapping the wiring at high-resolution."

The findings have implications far beyond mice. At some point since humans split from chimps millions of years ago, our higher brain regions gained enough control over vocalization to produce speech. The singing mouse's two amplified brain regions are central to human vocal circuits.

Additionally, brain-imaging research has identified stronger connections between similar motor and auditory areas in humans than in other primates. The singing mouse may be replaying a version of the evolutionary trick that put our ancestors on the road to language.

"The fact that these changes are relatively simple and targeted raises an exciting possibility," said Zador. "If only a few specific wiring changes separate singing mice from lab mice, we might be able to engineer those changes ourselves. Could we make a lab mouse sing?"

Could tomorrow's pop stars see competition in unexpected places? Maybe not, but the discovery could someday provide new tools for speech therapy while addressing a question central to the human experience. How did language emerge?

Publication details

Arkarup Banerjee, Specific expansion of motor cortical projections in a singing mouse, Nature (2026). DOI: 10.1038/s41586-026-10458-y. www.nature.com/articles/s41586-026-10458-y On bioRxiv DOI: 10.1101/2024.09.13.612752

Journal information: Nature , bioRxiv

Key concepts

Biological neural networksOrganismal, population, evolutionary & ecological systems

Provided by Cold Spring Harbor Laboratory