When a dog freezes at another dog’s whine or backs away from a growl, most owners assume their pet is reading the caller’s emotional state. But a new study from Eötvös Loránd University suggests the picture is more specific than that.
Researchers played agonistic, playful, and distress vocalizations to dogs and tracked how they responded. What dogs appeared to decode wasn’t simply whether a sound was positive or negative — it was something else entirely. And when the team swapped in human and chimpanzee calls, the dogs’ responses shifted in ways that neither emotional valence nor motivational state could predict.
When dogs hear dogs: motivation trumps emotion
In Study 1, researchers played three categories of dog vocalizations to pet dogs and measured how quickly they approached or withdrew. The sounds represented agonistic calls (negative emotional valence, hostile motivation), play and comfort sounds (positive valence, non-hostile motivation), and distress calls (negative valence, non-hostile motivation). That last category was the critical test: distress and agonistic calls share the same emotional tone — both are negative — but differ in whether the caller poses a threat.
The results were unambiguous. Dogs approached distress calls faster and withdrew from them more slowly than they did from agonistic calls. Since emotional valence was identical in both cases, tone alone can’t account for the difference in behavior.
What did account for it was motivational state — specifically, whether the caller was hostile or non-hostile. A distressed dog isn’t a threatening one, and the listening dogs appeared to recognize that distinction. The social message they extracted wasn’t “this sounds bad” but something closer to “this caller needs help, not avoidance.”
Across species lines, the signal breaks down
Study 2 asked whether the same decoding ability extends beyond dogs. Researchers presented the same subjects with vocalizations from chimpanzees and humans — including both non-linguistic emotional calls and speech. If dogs were applying a general rule for reading vocal intent, their responses should have tracked the motivational state or emotional valence of these heterospecific callers as well.
They didn’t. Neither motivational state nor emotional valence predicted how dogs responded to chimpanzee or human sounds. The pattern that had held consistently for conspecific calls simply vanished.
This is a notable null result. It suggests that whatever dogs are doing when they decode another dog’s call, they’re not applying some transferable formula that works across species.
Challenging the idea of a universal vocal code
The finding runs against one of the more durable ideas in animal communication research. In 1977, biologist Eugene Morton proposed what became known as motivation-structural rules: the hypothesis that low-frequency, harsh sounds signal aggression or hostility across birds and mammals, while high-frequency, tonal sounds signal submission or friendliness. The prediction is that these acoustic patterns should be broadly readable — even across species lines.
There’s supporting evidence for a weaker version of this idea. Previous research has shown that humans can judge emotional arousal in vocalizations from a wide range of vertebrate species, suggesting at least some shared acoustic cues. The new study indicates, though, that when it comes to how dogs actually respond behaviorally — not just perceive — the mechanisms underlying conspecific call decoding appear to be species-specific.
This doesn’t invalidate Morton’s rules entirely. It does, however, complicate the picture considerably. Decoding social messages from vocalizations, the researchers suggest, may rely less on universal rules than the field has sometimes assumed.
Evolution shaped dogs to read intent, not feeling
Why would motivation matter more than emotion? The researchers offer an evolutionary argument. In close-contact social situations — the kind that characterize life within a pack or group — knowing whether a caller is hostile or non-hostile may have been more immediately useful than knowing whether they feel good or bad. A distressed companion calls for a different response than an aggressive rival, even when both sound negative.
The distress calls in this study triggered behavior consistent with prosocial responding: faster approach, slower withdrawal. This pattern aligns with earlier research documenting empathy-like behavior in dogs, including studies showing that dogs orient toward humans and conspecifics in apparent distress.
The study’s limitations are worth acknowledging. Sample sizes were constrained, and a controlled playback experiment differs substantially from the complexity of actual dog interactions, where body language, scent, and context all shape how signals get read.
Still, the motivation-versus-emotion distinction carries practical weight. If dogs are primarily reading intent rather than affect, that has implications for how we interpret dog behavior in social and welfare contexts — and perhaps for how we think about what dogs are actually picking up from us when we speak to them.
The broader question the study leaves open is worth sitting with: how much of what we think animals share, perceptually and socially, is genuinely universal, and how much reflects the particular evolutionary pressures each species faced? Dogs may be remarkably attuned to other dogs — but that attunement, it seems, was shaped for dogs specifically. What looks like a general capacity for reading social intent may, in many cases, be a finely tuned species-specific skill wearing the costume of something universal.