Study reveals whistling secret of horses’ whinny

Lead

Researchers report in Current Biology (published Monday, 23 February 2026) that the familiar horse whinny combines two distinct sound mechanisms: low-pitched vibration from the voice box and a previously unrecognized high-pitched whistle formed in the laryngeal area. The team filmed live horses with an endoscopic camera, scanned laryngeal anatomy, and ran airflow tests on excised larynges to trace how the sounds are produced. Their results show the high tones originate as a whistle that escapes through a small opening above vibrating tissues, making horses the first large mammals known to generate a laryngeal whistle while also producing vocal-fold sounds. The finding helps explain how horses convey complex social signals such as greetings, contact calls and excited vocalisations around feeding.

Key Takeaways

• The study appeared in Current Biology on Monday, 23 February 2026, and reports a combined mechanism for the horse whinny: simultaneous low- and high-frequency components produced by different structures.
• Researchers used three primary methods: endoscopic filming through the nasal passages of live horses, detailed anatomical scans, and controlled airflow experiments on isolated larynges from deceased animals.
• The low-frequency portion is generated by tissue vibration in the voice box (analogous to human vocal-fold phonation); the high-frequency component is a whistle formed just above the vibrating tissues.
• Horses are the first large mammals documented to produce a whistle from the laryngeal region while also producing voiced sounds; a few small rodents make similar whistles, but the mechanism and scale differ.
• Related species such as wild Przewalski’s horses and some elk show comparable dual-tone calls, while donkeys and zebras do not appear to produce the same high-pitched whistle.
• Authors suggest the two-tone structure could allow horses to convey multiple signals at once—information about identity, distance and emotional state may be carried on different frequency channels.
• Practical implications include new avenues for monitoring horse welfare and refining acoustic models of mammalian communication; basic evolutionary questions about the trait’s origin remain open.

Background

Horses use whinnies and nickers in many social contexts: to locate unfamiliar individuals, to greet familiar companions or to express anticipation at feeding. For decades, researchers recognized that many horse calls contain both low and high spectral components, but the anatomical source of the high tones was not clear. Larger animals typically produce lower sounds because vocal anatomical structures scale with body size, so the presence of conspicuously high pitches in an adult horse’s call challenged existing expectations about vocal production in large mammals.

Animal communication research has documented diverse sound-generating mechanisms across taxa—vocal-fold vibration in most mammals, specialised vocal membranes in some cetaceans, and oral whistling in humans. Small rodents, including certain rats and mice, produce ultrasonic whistles via airflow through specialised upper-airway structures, but such whistling had not been demonstrated in a large mammal’s larynx. The new study therefore sits at the intersection of comparative anatomy, acoustics and social behaviour, testing whether horses use purely vocal-fold mechanisms or an additional airflow-based whistle to create their high tones.

Main Event

The research team introduced a small endoscope through each horse’s nostril to capture real-time images of the larynx as animals produced whinnies and softer nickers. Video frames showed that during whinnies the vocal tissues vibrated while an adjacent laryngeal region narrowed to a small aperture above those vibrating bands. The narrowing was timed with the emergence of the high-frequency component in acoustic recordings.

To test causality, researchers scanned laryngeal anatomy and then performed airflow trials on excised larynges. When air was driven through isolated larynges, the same combination of tissue vibration and a whistle-like tone emerged if the supraglottic area formed a small aperture; altering that aperture changed the whistle’s presence and spectral character. The experiments reproduced the two-component acoustic signature observed in live animals.

The team contrasted whinnies with nickers, finding the nicker lacks the strong whistle component and corresponds to a different configuration of laryngeal tissues. Taken together, the live imaging, anatomical scans and airflow experiments support the interpretation that horses generate the whinny’s high tones via a laryngeal whistle operating alongside conventional vocal-fold vibration.

Analysis & Implications

The discovery reframes how scientists think about information encoding in horse vocalisations. Producing two modes simultaneously gives horses a simple physical way to multiplex signals: low-frequency energy can carry cues about body size or arousal while the whistle can add identity, urgency or other social information without interfering with the low channel. That layered signalling increases the communicative bandwidth available to social animals like horses.

From an evolutionary perspective, the presence of comparable two-toned calls in Przewalski’s horses and some elk suggests either convergent evolution of a useful signaling solution or retention of an ancestral capability lost in other equids. The absence of the whistle in donkeys and zebras raises questions about when and why this capacity evolved or was lost across the equid family tree. Robust phylogenetic sampling and developmental studies will be needed to resolve those possibilities.

There are also applied consequences. Acoustic markers that separate whistle and voiced components could improve remote monitoring of horse welfare—alerts for separation distress, pain or feeding anticipation could be refined by analysing the balance and timing of the two modes. Additionally, the finding challenges acoustic models that treat mammalian calls as single-mechanism signals, prompting revisions in bioacoustic theory and animal-communication modelling.

Comparison & Data

The table summarises mechanisms without assigning numeric frequency bands, because the study emphasises structural origins rather than fixed frequency cut-offs. Contextual factors—individual variation, posture, and ambient acoustics—modulate how the two components are perceived in the field, so practical acoustic monitoring should combine spectral analysis with behavioural observations.

Reactions & Quotes

Outside experts greeted the finding as both surprising and clarifying, noting it reveals a previously hidden dimension of equine communication.

“I’d never imagined that there was a whistling component. It’s really interesting, and I can hear that now.”

Jenifer Nadeau, University of Connecticut (equine researcher, comment)

Nadeau, who was not part of the study, said the live endoscopy footage makes the whistle obvious in ways acoustic data alone could not, and she encouraged further behavioural tests linking whistle features to social contexts.

“Knowing that a ‘whinny’ is not just a ‘whinny’ but that it is actually composed of two different fundamental frequencies created by two different mechanisms is exciting.”

Alisa Herbst, Rutgers Equine Science Center (equine scientist, email comment)

Herbst highlighted the potential to reinterpret earlier acoustic datasets in light of the dual-mechanism model and to explore links with emotional state and group dynamics.

“They can express emotions in these two dimensions.”

Elodie Mandel-Briefer, University of Copenhagen (study author)

Mandel-Briefer, the lead author, framed the whistle as a mechanism that expands the expressive range of horse calls and noted plans for follow-up work examining wild populations and developmental onset of the whistle.

Unconfirmed

• Whether the laryngeal whistle evolved once in equids or multiple times across related species remains unresolved and requires broader phylogenetic sampling.
• The exact communicative contents carried separately by the low and high components—identity, emotional valence or distance cues—are hypothesised but not yet experimentally proven in field contexts.
• Reports that donkeys and zebras cannot produce the high-pitched whistle come from comparative observations but lack exhaustive physiological testing across breeds and populations.

Bottom Line

The study overturns a long-standing assumption that a horse whinny is a single-mode vocalisation by showing it combines vocal-fold vibration with a laryngeal whistle. This dual mechanism enlarges the possible information horses can transmit in a single call and invites researchers to reanalyse previous acoustic data with fresh hypotheses about multiplexed signalling.

Follow-up work should map the trait across wild and domestic equids, test behavioural responses to isolated components, and explore links to welfare monitoring. For horse owners, scientists and bioacousticians, the result opens practical and theoretical avenues: a new acoustic marker for animal state and a reminder that large mammals can sometimes surprise us with novel vocal mechanisms.

Sources

• The Guardian (news report summarising the study)
• Current Biology (peer-reviewed journal where the study was published)
• University of Copenhagen (academic affiliation of lead author, institutional information)
• University of Connecticut (academic comment from Jenifer Nadeau)
• Rutgers Equine Science Center (expert comment from Alisa Herbst)