Summary: Alston’s singing mice produce both loud, rhythmic, human-audible songs and high-pitched ultrasonic calls used for close-range interactions. New experiments show that both vocal types are created by a whistle-like airflow mechanism rather than by vibrating vocal cords, and that the same midbrain vocal circuit controls both modes.
Unexpectedly, the brain region that governs routine squeaks in laboratory mice—located in the midbrain—also controls the complex songs and ultrasonic vocalizations (USVs) of the singing mouse. These results show how sophisticated vocal behaviors can arise from shared neural circuitry, offering insight into the evolution of communication and potential relevance to human speech disorders.
Key Facts
- Whistle-Based Sound Production: Both long-range songs and short-range ultrasonic calls are generated by airflow-driven whistle mechanics.
- Shared Brain Circuit: The same core midbrain vocal node drives both song and USVs, the same general region used by ordinary mice for vocal behaviors.
- Human Health Relevance: Understanding how a conserved circuit can produce diverse vocal outputs may inform studies of stroke-related aphasia, autism-related communication differences, and improvements in AI sound recognition.
Source: CSHL
All mice make sounds, but only a few species sing. Scotinomys teguina, commonly called Alston’s singing mouse, inhabits the cloud forests of Costa Rica. Researchers at Cold Spring Harbor Laboratory (CSHL) study these animals to learn how complex vocal behaviors evolved and how shared brain circuits can support new modes of communication.
Beyond evolutionary questions, this research may shed light on medical conditions that affect speech and social communication.
“Mice don’t just squeak; many species use ultrasonic vocalizations (USVs) that are too high and too quiet for humans to hear without equipment,” explains Arkarup Banerjee, an assistant professor at CSHL. By contrast, the songs of Alston’s singing mice are loud and patterned enough to be heard by people.
Singing mice appear to rely on songs to carry across the open, foggy environment of cloud forests, while USVs serve short-range social communication. To understand how these different signals are produced and controlled, Banerjee and colleagues combined behavioral assays, physiological manipulations, and targeted neural interventions.
The team developed a behavioral paradigm called PARId (partial acoustic isolation reveals identity) to separate and identify the different vocal outputs during social interactions. PARId showed two distinct vocal modes: soft, variable USVs used in close encounters and loud, rhythmic songs used for long-distance signaling.
To test how the sounds are generated, postdoctoral researcher Cliff Harpole exposed mice to a helium-rich atmosphere. If sounds result from vibrating vocal folds, helium changes would have different effects than on a whistle mechanism. The pitch of both songs and USVs shifted upward in helium, indicating a whistle-like production—airflow-driven tones—rather than classic vocal cord vibration.
Further experiments used viral tools to manipulate specific midbrain regions. Graduate student Xiaoyue Mike Zheng selectively targeted the caudolateral periaqueductal gray (clPAG) and related nodes. These manipulations showed that song amplitude and duration are regulated by this conserved midbrain circuit node; silencing it progressively reduced song strength and length. Crucially, the same node also gates USVs, meaning both vocal modes rely on the same central control point.
Mathematical modeling of song rhythm combined with the neural silencing experiments indicated that song termination can be described by a single parameter influenced by clPAG activity. This mechanism helps explain natural variation in song features, including sex differences, and suggests that tuning a central circuit node can generate new behavioral outputs without evolving an entirely new pathway.
Banerjee notes, “These results highlight what is shared across species and point us to the features that differ. Understanding how a conserved circuit can be repurposed offers a framework for studying how communicative behaviors evolve quickly in mammals.”
Beyond evolutionary biology, these findings may have translational value. Identifying how a central neural node controls diverse vocal behaviors could inform rehabilitation strategies for stroke-related speech loss, shed light on communication differences in autism, and inspire better algorithms for automated sound and speech recognition.
Key Questions Answered:
A: Alston’s singing mice produce long, loud, rhythmic songs using an airflow-driven whistle mechanism that enables long-range signaling, while most mice rely primarily on softer ultrasonic calls for close-range communication.
A: No. Both songs and ultrasonic vocalizations are controlled by the same midbrain vocal node—the caudolateral periaqueductal gray (clPAG)—which is also used by ordinary laboratory mice for vocal behaviors.
A: The study shows how a conserved neural circuit can generate diverse vocal outputs, offering a model for how vocal behaviors evolve and suggesting potential avenues for research into speech recovery after stroke and communication differences in autism.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The full journal paper was reviewed for accuracy.
- Additional context and clarification were provided by the editorial staff.
About this communication and neuroscience research news
Author: Samuel Diamond
Source: CSHL
Contact: Samuel Diamond – CSHL
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Vocal repertoire expansion in singing mice by co-opting a conserved midbrain circuit node” by Arkarup Banerjee et al. Current Biology
Abstract
Vocal repertoire expansion in singing mice by co-opting a conserved midbrain circuit node
Understanding how neural circuits generate a variety of behaviors is a central goal in neuroscience. Distinct outputs may come from dedicated motor pathways or from shared circuits operating in different functional modes. While multifunctional circuits offer efficiency and may enable rapid behavioral evolution, their mechanisms are still poorly understood in mammals.
Using the singing mouse (Scotinomys teguina) as a model, the authors developed a behavioral assay, PARId (partial acoustic isolation reveals identity), that precisely attributes vocalizations during social interactions. This paradigm revealed two vocal modes: soft, variable ultrasonic vocalizations (USVs) used for short-range communication, and loud, rhythmic, human-audible songs unique to singing mice for long-range signaling.
Despite their acoustic and behavioral differences, USVs and songs share the same sound production mechanism—phonatory-respiratory coupling—and are gated by the midbrain caudolateral periaqueductal gray (clPAG). Mathematical modeling of song rhythm, paired with synaptic silencing of clPAG, showed that gradual reduction of clPAG activity decreases song amplitude and duration, and that song termination can be captured by a single governing parameter.
These results explain natural variability, including sexual dimorphism in songs, and identify clPAG as a central node that can be parametrically tuned to produce distinct vocal modes. The work provides a mechanistic explanation for how conserved circuits can be repurposed to drive rapid behavioral evolution in mammals.