Skip to content
NatureHQ

Birdsbehaviour

Mobbing

A bird flying at an owl is doing something dangerous. The interesting question is what makes it worth the risk.

Mobbing is prey approaching a predator and harassing it — which is dangerous, and therefore needs explaining. It drives the predator away, advertises that it has been detected, and teaches watching birds what to be afraid of, including things that are not dangerous at all.

Everything else in this family is about avoiding predators. Mobbing runs the other way: small birds fly towards an owl, call loudly, and harass it until it leaves. That is a real risk — mobbers are occasionally killed — so the behaviour cannot be explained as panic, and a working account has to say what is worth that risk. Several things are, and they are not exclusive. Harassment often does drive a predator out of the area, which removes it from where the mobber lives and nests. Calling loudly at a predator also advertises that it has been seen, and an ambush predator whose position is known has lost the thing it depends on, so leaving is its best move. And mobbing is a public event, watched by every bird nearby. The teaching function has the most striking evidence in this whole sprint, from an experiment run in the 1970s. Blackbirds were arranged so that one bird could see a stuffed owl while another, seeing only the first bird, was shown a harmless object — in some trials a plastic bottle. The observer, watching apparently frantic mobbing directed at a bottle, began mobbing bottles. And it passed that on: the response transmitted along a chain of naive birds, each learning from the last. Predator recognition can be acquired socially, and so, demonstrably, can a mistake. The information travels between species too. Chickadee mobbing calls encode how dangerous the predator is, and nuthatches listening to those calls adjust their own response to match the encoded threat — reading not just that something is wrong but how alarmed the chickadee is. A woodland alarm network is a shared information system rather than a set of parallel private ones.

Early coverage · 38% complete · reviewed 2026-09-03

What this page covers

Mobbing is best documented in birds and occurs in mammals and some fish. Much of the experimental work uses playback, which is why bird systems dominate the evidence.

Often confused with: Territorial aggression, which is directed at rivals rather than at a predator; Panic, when mobbing involves approaching the danger rather than fleeing it; A single behaviour with one purpose, when several functions operate at once

Quick facts

It is genuinely risky
Mobbers are sometimes killed, so the behaviour needs an explanation
Several functions at once
Driving the predator off, advertising detection, teaching onlookers
Recognition is learned
Blackbirds taught to mob a bottle passed it along a chain of birds
Information crosses species
Nuthatches read how alarmed a chickadee is, not just that it is alarmed

Why approach the thing that eats you

Mobbing costs something real, so the benefits have to be worth it.

  • Moving it on. Sustained harassment frequently causes a predator to leave the area, which is worth a great deal to a bird with a nest in it.
  • Advertising detection. An ambush predator that has been spotted and announced has lost its advantage, and its best move is to go elsewhere.
  • Teaching. Mobbing is public, and birds watching learn what the object of it is — which is where predator recognition partly comes from.
  • Recruiting. Mobbing calls draw other birds, including other species, so the harassment escalates.

These are not competing explanations to be adjudicated between. They operate simultaneously and their balance differs by species and situation, which is the conclusion the founding review reached and which later work has not overturned.

Learning what to fear, including the wrong things

The experiment that taught blackbirds to mob a bottle.

Predator recognition is partly learned by watching. Blackbirds shown another bird mobbing a harmless object began mobbing it themselves — and passed the mistake along a chain of further birds.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Enemy recognition can be acquired socially through observation of conspecific mobbing. In a paired-cage design, observer blackbirds shown a conspecific mobbing while themselves viewing a harmless object acquired a mobbing response to that object, and the acquired response was transmitted through a chain of naive individuals.

Who this applies to
Demonstrated in blackbirds; social acquisition of predator recognition is documented more widely in birds and mammals.
Studied in
Turdus merula, Aves
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

The deliberately false association is what makes it decisive: the observers respond to an object that has nothing predatory about it, so the response can only have been acquired from watching another bird.

How far it can be extended

Socially acquired predator recognition has been demonstrated in several bird and mammal species by comparable methods.

Caveats

  • Captive birds in an artificial apparatus, with a transmission chain shorter than a real population.
  • It shows a false association can propagate, without establishing how often incorrect recognition arises in the wild.

Still unanswered

  • How wild populations correct mistaken predator associations, given that this experiment shows how readily they can spread.

Last reviewed 2026-09-03

The evidence (2 studies)

How we know

Teaching blackbirds to mob a bottle

Do birds inherit knowledge of what a predator looks like, or learn it from each other — and if they learn it, can they learn something false?

Blackbirds were placed in adjacent cages arranged so that a "teacher" bird could see a stuffed owl while an "observer" bird, able to see only the teacher, was simultaneously shown a harmless object — a honeyeater, and in some trials a plastic bottle. The observer therefore watched mobbing apparently directed at the harmless object. Observers were then used as teachers for further naive birds.

What happened

Observers began mobbing the harmless object, and the response was transmitted along a chain of several further birds, persisting through the chain.

What it shows

That predator recognition can be acquired socially, and that an incorrect association can propagate through a population. The bottle is what makes the result decisive: there is nothing about it to respond to, so the response can only have been learned by watching.

What it does not show

Captive birds in an artificial apparatus, with a chain shorter than a real population. It demonstrates that a false association can spread without establishing how often mistaken recognition arises in the wild, or how populations correct it.

The controls — what makes this evidence rather than a story
  • The observer never sees the owl, so any acquired response must come from watching the teacher.
  • A plastic bottle as the object in some trials — nothing about it is predator-like, which excludes the possibility that observers responded to genuine predator features.
  • A transmission chain of successive naive birds, testing whether the response propagates rather than merely appearing once.

From Cultural transmission of enemy recognition: one function of mobbing

The deliberately false association is what makes this decisive rather than merely suggestive. A bird that learns to mob an owl might be responding to something genuinely owl-like. A bird that learns to mob a plastic bottle cannot be — there is nothing about a bottle to respond to — so the response can only have been acquired by watching another bird. And it then propagated through a chain of birds, none of whom had seen an owl at all.

Diagram

What one bird mobs, the next bird learns

A mistake travels as readily as the truth.

What one bird mobs, the next bird learnsTeacher seesan owlObserver seesonly a bottleNext birdlearns it tooand onNothing about a bottle to react to — so it was learned by watching.Predator recognition is partly cultural, and a mistake travels as readily as the truth.How wild populations correct such mistakes is not known.
The same explanation in words

A chain of three boxes connected by arrows. The teacher bird sees a stuffed owl and mobs. The observer bird, able to see only the teacher, is meanwhile shown a harmless object — a plastic bottle — and so watches mobbing apparently directed at it. The next bird learns from the observer, and the response continues along the chain. Because there is nothing about a bottle to react to, the response can only have been acquired by watching another bird. Predator recognition is therefore partly cultural, and how wild populations correct such mistaken associations is not known.

The alarm network is shared

And the information crossing between species keeps its detail.

Nuthatches listening to chickadee alarm calls do not just hear that something is wrong. They read how alarmed the chickadee is, and adjust their own response to match the threat encoded in the call.

Well supported

Good evidence backs this, though some details remain open.

Sitta canadensis responds to playback of Poecile atricapillus mobbing calls in a manner graded to the threat level encoded in call structure, indicating extraction of specific threat information from heterospecific signals rather than generalised arousal.

Who this applies to
Demonstrated between two species in one North American woodland community.
Studied in
Sitta canadensis, Poecile atricapillus
Why we rate it this way, and what the caveats are
Well supportedHigh confidence

A playback experiment with graded stimuli and a graded response, which is the design needed to distinguish reading information from simply being alarmed by noise.

Caveats

  • Playback removes the caller and the surrounding context, so the response is to sound alone.
  • Two species in one community; how general this level of cross-species reading is remains open.

Still unanswered

  • How widely species in a community read each other’s graded alarm signals, which would make the local alarm network a shared information system.

Last reviewed 2026-09-03

The evidence (2 studies)

Chickadee mobbing calls carry graded information about how dangerous the predator is — smaller, more manoeuvrable predators, which are more dangerous to a small bird, get a more urgent call. Nuthatches, an entirely different species, respond to playback of those calls in proportion to the threat encoded in them. They are not simply alerted by an alarm; they are reading its content.

Related

The research behind this page

4 studies, newest first. Each one has a page explaining what it found and what it could not show.

2007Proceedings of the National Academy of Sciences

Nuthatches eavesdrop on variations in heterospecific chickadee mobbing alarm calls

Nuthatches adjusted their mobbing response to match the threat level encoded in the chickadee calls, responding more strongly to calls that chickadees produce for more dangerous predators.

2005Science

Allometry of Alarm Calls: Black-Capped Chickadees Encode Information About Predator Size

Smaller, more manoeuvrable raptors — the ones that actually threaten a chickadee — produced calls with more "dee" notes.

1978Science

Cultural transmission of enemy recognition: one function of mobbing

Observer birds began mobbing the harmless object, and passed that response on to further birds through a chain of several individuals, with the acquired response persisting through the transmission chain.

1978Zeitschrift für Tierpsychologie

The adaptive significance of avian mobbing

Mobbing serves several non-exclusive functions, and the balance differs by species and context.

This page is a stop on a longer route

A guided journey reads several subjects in a deliberate order, with an argument for why one follows another. You can join in the middle.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 38% completeness against what we would call a finished subject, and was last reviewed on 2026-09-03. It carries 2 claims and answers 5 mapped search questions.

  • no research from the last few years is attached — check for newer work
  • more experiments could be explained in plain English
  • no popular claim about this subject has been checked yet
  • Mammalian and fish mobbing are mentioned rather than covered.
  • How wild populations correct mistaken predator associations is unknown, given how readily they spread experimentally.
  • The costs of mobbing are known to be real and are rarely quantified.