Yes, and the evidence is better than the usual backyard anecdote. Researchers at the University of Washington put rubber masks on their own faces, caught crows while wearing them, and then walked around campus for years afterwards to see what would happen.
What happened is that the crows scolded the mask. Not the person, not the clothes, not the height or the walk. The mask.
The more interesting result came later, and it is the part most summaries leave out.
What the experiment actually did
The design is simple enough to describe in a paragraph, which is part of why it is convincing. The National Wildlife Federation account of the work describes how "researchers netted and banded sets of crows at various sites on campus and around Seattle while wearing rubber human masks."
Being caught and banded is not injury, but from a crow point of view it is a bad day. The mask was the only thing the birds could reliably associate with it.
- The threatening face
- A rubber mask worn only while catching and banding the birds
- The caring face
- A different mask, worn only by the people who fed the captive birds
- What was tested afterwards
- People wearing each mask walked the same routes and the researchers recorded which birds scolded and how many joined in
- Why masks rather than faces
- A mask can be handed to someone of a different height, build and gait. If the birds still react, the face is what they are reading
That last point is the whole design. And the reported result is blunt: "the crows ignored all but the particular mask that had been worn during the banding, which they greeted with loud scolding cries." One researcher put it as "The interesting part was that not a whole lot mattered except the face."
How long the memory lasts
This is where the study stops being a neat demonstration and starts being strange.
The same account reports that "close to seven years after the study began, the birds continue to harangue the banding mask even though they see it only twice a year."
Twice a year. Most training regimes for domestic animals would not survive that schedule. These birds are maintaining a grudge on two exposures per year, against a face that did them no lasting harm, for the better part of a decade.
The part that is genuinely odd
If the story ended with individual memory it would be impressive but ordinary. It does not end there.
Over the years of the study, "the percentage of birds joining the ruckus has roughly doubled since the experiment began, even though most have never been banded and likely didn't witness the original traumatic event." Some of the birds now scolding the mask are "young birds born in years since."
So the number of crows who object to a particular human face grew, while the number who actually had a bad experience with it stayed the same. Birds that were not present, and in some cases not yet hatched, learned which face to dislike.
The Cornell Lab notes that "American Crows are highly social birds, more often seen in groups than alone," and that "crows often stay together in year-round family groups that consist of the breeding pair and offspring from the past two years." A young crow spends its first years alongside adults who already know which face is the problem.
The crow in your yard is probably a regular
None of the above matters much if the birds outside your window are a different set each week. They are not.
The Cornell Lab describes American Crows as birds that "regularly uses both natural and human created habitats, including farmland, pasture, landfills, city parks, golf courses, cemeteries, yards, vacant lots, highway turnarounds, feedlots" and notes that they stay in "year-round family groups that consist of the breeding pair and offspring from the past two years."
Family groups that hold together across years, in a habitat list that includes yards, is the condition under which a durable face memory does anything at all. The same small set of birds is looking at the same person repeatedly.
The scale can be larger than it looks from a kitchen window. Cornell notes that "winter roosts of American Crows sometimes number in the hundreds of thousands." The half dozen in your garden in July are not necessarily the whole social world those birds belong to.
How the face could spread without anyone teaching anything
The doubling result gets retold as crows telling each other about you. That is one possible mechanism and it is not the only one, and the sources we could open do not settle which is happening.
- Learning by watching
- A young crow sees adults mob a particular face and joins in, the way it would join any mobbing. It learns the face from the reaction, not from being told about it
- Learning by consequence
- A bird that scolds alongside the group is doing what the group does. Nothing about the original event needs to be transmitted at all
- Something more like telling
- This is the interpretation that makes the best headline and the one that would need the most evidence. We did not find that evidence in what we could read
- What is not in dispute
- The count of birds reacting to that face grew while the count of birds that were ever caught did not
We are drawing this line because the difference matters for what you can conclude about your own yard. Birds copying a reaction they can see is ordinary and well within what a social species does. A crow describing your face to another crow is a much larger claim, and it is not one the pages we read actually make.
What the brain scans showed, and what we could not read
A later stage of the work put crows through brain imaging while they looked at the two masks. The University of Washington describes the setup: 12 adult male crows were caught by "investigators all wearing masks that the researchers referred to as the threatening face," and then "for the four weeks they were in captivity, they were fed by people wearing a mask different from the first, this one called the caring face."
On what lit up, the university writes that "it appears that birds have a region of their brain that is analogous to the amygdala of mammals. The amygdala is the region of the vertebrate brain where negative associations are stored as memories."
John Marzluff, who led the work, is quoted as saying "The regions of the crow brain that work together are not unlike those that work together in mammals, including humans."
- What we read ourselves
- The University of Washington news release describing the imaging study, and the National Wildlife Federation magazine account of the mask work
- What we could not open
- The peer reviewed papers themselves. The 2011 journal paper on the mask work sits behind a publisher paywall, and the journal and archive pages for the imaging study returned a bot check rather than the text
- What that means for this article
- Every quoted sentence here comes from a page we opened and read. We are not going to describe the contents of a paper we could not get into
The other mask is the useful one
Almost every retelling of this research focuses on the grudge. The caring mask is the half worth paying attention to if you have a yard.
Marzluff, on the birds who were fed by a masked keeper: "By feeding and caring for birds in captivity their brain activity suggests that the birds view their keepers as valued social partners, rather than animals that must be feared."
Valued social partner is a strong phrase for a bird looking at a person. It is also the other end of the same mechanism. A face that predicts something good gets filed as reliably as a face that predicts something bad.
What this means if you keep water in the yard
Here is where we have to be careful, because the research above is about faces and the question you probably came with is about your bird bath. We are going to separate the two.
- What the research supports
- Crows can tell human faces apart, hold the association for years, and other crows adopt it
- What it does not say
- Nothing in this work is about bird baths, water depth or feeding. Do not let us or anyone else stretch a face recognition study into a product claim
- What is reasonable to infer
- You are a face to the crows in your yard, and you are the same face every day. Whatever category they have put you in, you have been in it a while
The practical half is simpler and is not about recognition at all. Crows are large, they arrive as family groups rather than singly, and Cornell lists yards among the habitats they regularly use. A basin sized around finches and chickadees will be emptied and dirtied faster by a crow family than by the birds it was designed around.
- Expect to clean more often, not differently. The method does not change, the interval does. The procedure is in the guide to cleaning a concrete bird bath.
- Watch the water level rather than the calendar. Several large birds drinking and bathing move a surprising amount of water out of a shallow basin.
- Do not try to exclude them by making the water deeper. Depth is set by what is safe for the smallest birds using it, and that constraint does not move because a crow turned up.
- Keep the approach open. The sight lines that let small birds see a cat coming are covered in the article on cats and bird baths, and they help here too, because a crow arriving is also something the small birds want to see early.
Can you get on a crow's good side
This section is our reading rather than a finding, and we are marking it as such.
The captive birds in the imaging study were fed daily by the same masked people for four weeks, in a cage, with nothing else going on. That is a much denser schedule of association than a bird in a garden gets. We would not promise that leaving water out produces the same filing.
What the research does support is narrower and still worth knowing: the categories exist, they are durable, and they are built out of what you repeatedly do in front of the bird. Being consistently boring and harmless is a real strategy.
Frequently asked questions
Do crows remember faces or just clothing and habit
The mask design exists to answer exactly this. The same mask was worn by different people, and the reported result is that "not a whole lot mattered except the face."
How long is the grudge
Close to seven years in the study described above, on roughly two exposures a year. That is the longest figure we found in a source we could open. We are not going to extrapolate past it.
Do other birds do this
Our site has separate pieces on cardinals, chickadees, sparrows and roadrunners. This article is only about what the crow research showed, and we are not transferring it to species it did not test.
Will crows chase smaller birds away from my bird bath
We have no study to point at for this and are not going to invent one. What we can say is that crows are large and social, so a family group turning up is a different event from a single songbird arriving.
Does the water I put out change anything
Not for recognition, as far as anything we read shows. It changes how often you will be cleaning. What to fill it with is covered in the article on what water to use in a bird bath.
References
- Reference: Crows react to threats in human-like way | University of Washington, news release on the brain imaging study
- Reference: Crows Can Distinguish Faces In a Crowd | National Wildlife Federation, National Wildlife magazine
- Reference: American Crow Life History | The Cornell Lab of Ornithology, All About Birds