Summary: Speech has long been seen as a huge jump in brain complexity. New research, however, shows that the neural changes needed to produce complex vocalizations can be surprisingly small. Comparing ordinary laboratory mice with Alston’s singing mice, a Central American species known for rapid duets, researchers found that the difference is not a larger brain or new brain regions but a precise increase in specific neural projections.
Evolution appears to have tripled the number of neurons linking the mouth-movement control center to just two target areas. This focused rewiring—rather than wholesale reorganization—may represent a simple evolutionary route toward the kinds of vocal control that eventually supported human language.
Key Facts
- Conversation-like timing: Alston’s singing mice (Scotinomys teguina) produce loud, structured songs and engage in split-second turn-taking during rapid duets, a behavior that parallels human conversational timing.
- Brains look the same at a glance: The overall anatomy of singing mouse brains is visually indistinguishable from that of lab mice; there are no new gross brain regions.
- High-resolution mapping: Using molecular barcoding and high-throughput tracing, researchers mapped thousands of individual neurons and pinpointed changes to a specific cortical hub: the orofacial motor cortex (OMC).
- Triple-strength connections: Singing mice show roughly three times as many projections from the OMC to two key targets:
- The auditory cortical region, important for hearing and coordinating turn-taking.
- The midbrain periaqueductal gray, a conserved vocal control center across mammals.
- Minimal changes, major effects: These findings suggest that new, sophisticated behaviors can evolve through targeted amplification of existing wiring rather than by creating entirely new circuits.
Source: CSHL
Speech and complex vocal communication are often assumed to require dramatic brain changes. A new study published in Nature challenges that assumption by showing how modest, specific increases in cortical projections can support sophisticated vocal behavior.
Alston’s singing mouse is a small rodent from the cloud forests of Central America. It produces audible, elaborate songs that can be heard across a room and frequently performs precisely timed duets. Because of this turn-taking and adjustable timing, these mice provide one of the closest non-human parallels to conversational exchange.
Scientists at Cold Spring Harbor Laboratory (CSHL) set out to identify what neural changes underlie this singing behavior. Rather than finding a larger brain or novel structures, they discovered a specific, large-scale expansion of projections from the orofacial motor cortex to two vocal-relevant regions. Outside of this targeted augmentation, the brain’s wiring closely resembles that of the common laboratory mouse.
Graduate student Emily Isko led the cellular-resolution mapping, using a molecular barcoding technique developed at CSHL to trace thousands of individual neurons across the whole brain. “When you compare brains side by side, they look nearly identical,” Isko explained. “The differences appear only when you map where single neurons send their signals.”
Associate Professor Arkarup Banerjee emphasized the importance of high-resolution wiring maps: “You might expect that novel vocal communication would require large-scale circuit reorganization. Instead, we found a couple of targeted changes to existing wiring patterns. This gives researchers a practical playbook: to understand new behaviors, find closely related species with big behavioral differences and then map their neural wiring at high resolution.”
The implications extend beyond mice. Key regions amplified in the singing mouse—motor cortex projections to auditory cortex and the periaqueductal gray—are central components of human vocal circuits. Comparative brain-imaging studies already indicate that humans have stronger connections between motor and auditory regions than do other primates. The singing mouse may illustrate a similar evolutionary shortcut by which modest changes in connectivity yield substantial gains in vocal control, a potential preadaptation on the path toward language.
Anthony Zador, whose lab developed the barcoding approach, noted the translational potential: “If only a few specific wiring changes separate singing mice from ordinary mice, then in principle those changes might be engineered experimentally. That raises the exciting question: could a lab mouse be made to sing?”
While the idea of lab mice competing with pop stars is fanciful, the broader discovery could one day inform therapies for speech disorders and deepen our understanding of how vocal communication evolved.
Key Questions Answered:
A: Researchers describe this as an exciting possibility. Because only a few specific connections differ between singing and non-singing mice, future genetic or circuit-level interventions might reproduce those changes to alter vocal behavior experimentally.
A: It provides a strong clue rather than a definitive answer. Humans show amplified motor–auditory connections compared with other primates, and the singing mouse appears to use a similar pattern of selective projection expansion that could support increased cortical control over vocalizations.
A: No. Lab mice produce mostly ultrasonic, reflexive squeaks. Singing mice exhibit cortical control over their songs, allowing them to modify timing and tempo in response to partners—a hallmark of flexible, advanced communication.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full.
- Additional context was added by editorial staff.
About this language and evolutionary neuroscience research news
Author: Samuel Diamond
Source: CSHL
Contact: Samuel Diamond – CSHL
Image: The image is credited to Neuroscience News
Original Research: Open access. “Specific expansion of motor cortical projections in a singing mouse” by Emily C. Isko, Clifford E. Harpole, Xiaoyue Mike Zheng, Huiqing Zhan, Martin B. Davis, Anthony M. Zador & Arkarup Banerjee. Nature
DOI: 10.1038/s41586-026-10458-y
Abstract
Specific expansion of motor cortical projections in a singing mouse
Understanding how changes in neural circuit architecture drive behavioral innovation is a central challenge in neuroscience and evolutionary biology. The neocortex is thought to facilitate rapid behavioral change in mammals, but quantitative tests of long-range connectivity differences across species have been limited by the lack of high-throughput, single-neuron resolution projection data.
This study examined the Alston’s singing mouse (Scotinomys teguina), which displays a pronounced vocal behavior absent in the laboratory mouse (Mus musculus). Using bulk tracing, serial two-photon tomography, and high-throughput DNA sequencing of more than 76,000 barcoded neurons, the authors discovered a pronounced expansion of orofacial motor cortical projections to an auditory cortical region and to the midbrain periaqueductal gray—both implicated in vocal behavior.
Individual neuron analyses revealed a preferential expansion of exclusive projections from the orofacial motor cortex to the auditory cortical region in the singing mouse. The results support the idea that selective enlargement of ancestral motor cortical projections can drive behavioral divergence on relatively short evolutionary timescales and provide a framework for probing enhanced cortical control over vocalizations—a feature important to the evolution of human language.
The comparative approach used—mapping high-resolution wiring differences between closely related species with divergent behaviors—can be generalized to other model clades to reveal quantitative rules governing neural circuit evolution.