Skip to content
NatureHQ

Claim check

Do butterfly eyespots scare birds by looking like owl eyes?

MisleadingThe words are defensible; the impression they create is not.

The display genuinely works — butterflies with eyespots painted out were attacked far more. But when researchers varied eye-likeness and conspicuousness separately, equally conspicuous bars and squares protected prey just as well as circles.

The claim as it circulates

“A butterfly flashes the eyespots on its wings and a small bird flees, because it thinks it is suddenly facing an owl.”

Where you may have met it: Documentary narration and museum captions; School materials about animal adaptations; Popular natural history writing

Taking it apart

What was claimed
That eyespots deter predators specifically by mimicking the eyes of a larger animal, causing the predator to react as though confronted by a threat.
What was actually observed
Peacock butterflies presented to blue tits survived far more often with eyespots intact than with them painted over, establishing that the display deters attack. Separately, artificial prey were presented to wild birds with markings that varied independently in how eye-like they were — circles versus bars versus squares — and in how conspicuous they were. Survival was predicted by conspicuousness; eye resemblance added nothing detectable, and equally conspicuous non-circular marks protected prey as well as circles.
What the evidence supports
That eyespots deter predators, and that the mechanism is salience — a large, sudden, high-contrast pattern producing hesitation — rather than resemblance to a vertebrate eye.
What it does not support
It does not show eyespots are ineffective; the deterrent is real and measured. It does not exclude eye mimicry contributing in particular species with unusually detailed eyespots. And it does not address deflection, a separate proposed function in which eyespots draw attacks towards a survivable part of the wing.

Why the owl explanation lasted a century is the instructive part, and it is not that anybody was careless. Every observation commonly offered in its support — that birds startle, that the spots look strikingly eye-like, that larger spots work better — is equally predicted by the conspicuousness account. Two hypotheses making the same predictions cannot be separated by collecting more observations. They need an experiment that pulls them apart, and nobody ran one until 2008.

The rest of the answer

The claims underneath

Each one carries its own evidence, scope and caveats. Expand any of them to reach the studies.

Butterfly eyespots do deter birds — but apparently not by looking like eyes. When conspicuousness and eye-likeness were varied separately, equally conspicuous bars and squares protected prey just as well as circles.

Well supported

Good evidence backs this, though some details remain open.

Field experiments varying eye resemblance and conspicuousness independently found survival predicted by conspicuousness, with highly conspicuous non-circular markings conferring protection equivalent to eye-like circles. The intimidation effect is real and attributable to salience rather than to mimicry of vertebrate eyes.

Who this applies to
Demonstrated with artificial prey against wild birds; the intimidation effect itself is separately demonstrated in a real butterfly.
Studied in
Lepidoptera, Aves
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

The independent manipulation is exactly the right test and gives a clear answer. It rests on artificial prey with abstract marks, and does not exclude eye mimicry operating in particular species, which is why this is well-supported rather than established.

How far it can be extended

The conspicuousness result comes from artificial prey and applies to how avian predators respond to salient markings generally.

Caveats

  • The deterrent effect itself is well established; what is not supported is the owl-face explanation for it.
  • Deflection — drawing attacks to a wing edge rather than the body — is a separate proposed function of eyespots and is not addressed by these experiments.

Still unanswered

  • Whether eye mimicry contributes in specific cases with unusually detailed eyespots, which the general result does not exclude.

Last reviewed 2026-09-03

The evidence (3 studies)

Go deeper

Eyespots

The spots do work — birds really do leave the butterfly alone. They just do not seem to work by looking like an owl.

Related things people ask

Last reviewed 2026-09-03