Hacker News new | past | comments | ask | show | jobs | submit login
Belyayev's Fox Experiment (wikipedia.org)
109 points by deogeo on Jan 25, 2020 | hide | past | favorite | 30 comments

A 2017 study showed that hypersocial dogs have structural changes in genes responsible for Williams-Beuren syndrome in humans. It speculates that this is what might have allowed domestication.



There has been some doubt about the validity of this experiment [1]:

"A widespread misconception maintains that the Farm-Fox Experiment started with wild foxes and recapitulated the entire process of domestication. Belyaev himself accurately described the founders as fur-farm foxes, but by referring to the unselected population as ‘wild controls’, contributed to this misconception. In reality, the experiment started with a fox population from eastern Canada that had been captive and purpose-bred since the late 1800s, something Belyaev and his colleagues may have been initially unaware of."

[1] http://dx.doi.org/10.1016/j.tree.2019.10.011

The purpose of the experiment wasn't to show "wild foxes can be domesticated in just 30 years of selective breeding", though that is the most famous and perhaps the most important takeaway from it.

The hypothesis under test was something like "expression of tameness is genetically linked to expression of physical characteristics, such as spotted fur and floppy ears". It's quite valid in that context: Belyayev started with a population with a low (perhaps non-zero) level of domestication, ended with a moderately domesticated population, and observed the physical differences.

It is doubtful, though, that the spotted fur, floppy ears, and other physical characteristics (domestication syndrome) are a result of the experiment [1]:

"The history of the Farm-Fox population undermines the commonly repeated narrative that a suite of domestication syndrome traits emerged solely as a result of selecting on tameness. There is no temporal link between most of the syndrome traits, which first appeared in Prince Edward Island (PEI) fur farms, and the later behavioral selection in Russia."

On the spotted fur [1]:

"The farm-fox breeders of PEI intentionally selected for white spotting and other unusual coat patterns (Figure 3). They noticed that crossing two white-marked foxes occasionally resulted in animals that held their heads askew, a phenomenon Belyaev would later describe in his population, suggesting shared genetic etiology. White spotting was more common in Belyaev’s selected than unselected populations, but has not been associated with less fearful behavior in individuals (Figure 2D)."

On the floppy ears [1]:

"The farm foxes of PEI occasionally had floppy ears, even as adults (Figure 3C). In the Farm-Fox Experiment, ‘delayed ear raising’ was noted (ears floppy past 3 weeks of age, but not necessarily into adulthood). While slightly more common in the selected population, the trait is extremely rare, and no association between delayed ear raising and less fearful behavior in individuals has been described (Figure 2D)."

[1] https://doi.org/10.1016/j.tree.2019.10.011

Even so, the change in behavior seems genuine, when compared to what it was initially. The experimental fault also seems easy to fix - just compare behavior to other fur-farm foxes, instead of wild ones.

The experiment specifically selected for behaviour, though. And the fur-farm foxes seem to have been quite tame already before the start of the experiment. Therefore, I don't find the changes in behaviour very interesting on their own [1]:

"The Farm-Fox Experiment selected for a behavioral trait that already existed in the population (Figure 3), essentially recapitulating a selection experiment performed on numerous occasions with dog breeds. [...] The change in the critical period of socialization in the selected foxes is more like the difference between more and less easily socialized dog breeds than between dogs and wolves."

[1] http://dx.doi.org/10.1016/j.tree.2019.10.011

I don't really understand this rebutal... the change in behaviour from the initial population (even if it had already been partially domesticated) was not just interesting, it was absolutely amazing.

You can check for yourself how the foxes brought to the US behave on Youtube[1]!

Also, the similar study with mice[2] by the same Russian Institute seems to have had even more striking results.

The rebutal seems to be using only theoretical objections, without any new original, empirical findings, to counteract tangible evidence that had been accepted by (and astonished) numerous other researchers who actually came into contact with the animals.

[1] https://www.youtube.com/watch?v=4dwjS_eI-lQ [2] https://www.nytimes.com/2006/07/25/health/25rats.html

The foxes are great proof that behaviour has a genetic component. But the change in behaviour over the course of Belyaev's experiments does not seem that remarkable. Some foxes of the initial population were already happy to approach humans and be held by them, before the experiment even started. The pictures in [1] show farm foxes from Prince Edward Island in 1922 which seem to behave extremely similar to the foxes in the youtube video you linked to.

[1] https://doi.org/10.1016/j.tree.2019.10.011

You are basically claiming that the difference in behaviour that was reported between the initial population and later generations was exaggerated because there seems to be a photo from 1922 showing a tame individual. I hope you see how claiming any level of tameness based on a few photos cannot be considered good evidence. Photos can be highly misleading, it's easy to see something that's not there. From what I understand, all sources reporting from this experiment and other similar ones claim huge differences in behaviour not only betweeen the initial generation and the selected ones, but also between control groups of animals which were kept in very similar conditions, except they were not submitted to "tameness selection". This difference of course over a large amount of animals, not over just a few. So, maybe the animals in the photo were indeed fairly tame (which as I said before, cannot be concluded from old photos), but even then, that does not mean that they were nearly as common as in the "evolved" animals from the study, where the rate of highly tamed individuals reached above 80% already. Which means that pointing out a counter-example is like claiming global warming is fake because it's been cold in the Midwest.

It is almost impossible to show how tame the fur farm foxes on Prince Edward Island really were, but in addition to the photos - of which there are more than one [1] - there are also written reports about how tame the foxes where [2]:

"Dr. Frank, among other things at Montreal, had a pair of tame foxes lead about on the main streets, driven around in automobiles and taken to a public dance, where the girls did the fox trot with these foxes around their necks, and all this was photographed, moving pictures taken, etc., and to cap the climax Dr. Frank hired two airplanes, took up a pair of foxes in one and from the other airplane a photograph ground out several reels to be shown in the movies in the near future."

How common these traits were is another difficult question. But given that there was prior selection for tameness and that Belyaev specifically looked for those tame foxes to include in his experiment [3], it does not seem so surprising that this trait very quickly became more common.

[1] https://www.washingtonpost.com/science/2019/12/03/tame-foxes...

[2] https://babel.hathitrust.org/cgi/pt?id=nyp.33433090788674&vi...

[3] https://doi.org/10.1002/bies.200800070

See also "Wolf Pups That Play Fetch" at https://news.ycombinator.com/item?id=22155843

The woman who took over the experiment after Belyayev died wrote a book about if a few years back: https://www.goodreads.com/review/show/2116251372

This is a horribly edited Wikipedia article yet it is such a delight to read.

I think https://blogs.scientificamerican.com/guest-blog/mans-new-bes... has much of the same content, and might be easier to read.

A handful of websites claim to offer pups from this specific colony as pets, which sounds wonderful. Though it's hard to validate they weren't just taken from a fur farm (thus not actually tame).

I was expecting much more content about the implications of these results on humans

are we domesticated our-self? how does society (morals, e.g.) acts as Belyayev selection?

Relating it to humans gets far more awkward than you might guess. Ponder any implications of what Wikipedia dares to say about the matter:

The changes manifested by the tame foxes over the generations, moreover, were not only behavioral but also physiological, just as Belyayev had expected. The first physiological change detected in the tame foxes was a lower adrenaline level. Belyayev and his team "theorized that adrenaline might share a biochemical pathway with melanin, which controls pigment production in fur", a hypothesis that has since been confirmed by research.[2] As they became tamer, more and more of the foxes began showing "signs of the domestication phenotype"; Trut later recalled that in the early 1980s "we observed a kind of explosion-like change of the external appearance."[5] After eight to ten generations, the tame foxes began to develop particolored coats, a trait found more in domesticated animals than in wild ones

Belyayev's experimental animals and their descendants have been said to "form an unparalleled resource for studying the process and genetics of domestication".[1] Brian Hare, a biological anthropologist, wanted to study "the unusual ability of dogs to understand human gestures". Hare "wanted to know if dogs' powerful rapport with humans was a quality that the original domesticators of the dog had selected for, or whether it had just come along with the tameness, as implied by Belyaev's hypothesis". He discovered "that the fox kits from Belyaev's domesticated stock did just as well as puppies in picking up cues from people about hidden food, even though they had almost no previous experience with humans."[1] In a 2005 paper for Current Biology, Hare suggested that selection for tameness "may have been sufficient to produce the unusual ability of dogs to use human communicative gestures" and that the inability of wild wolves to pick up human cues is caused by their fear of humans. While Belyayev and his team "didn't select for a smarter fox but for a nice fox", Hare said, "they ended up getting a smart fox." Belyayev's research, Hare further argues, has implications for the origins of human social behavior: "Are we domesticated in the sense of dogs? No. But I am comfortable saying that the first thing that has to happen to get a human from an apelike ancestor is a substantial increase in tolerance toward one another. There had to be a change in our social system."[5]

Short documentary from a few years ago:


The experiment is not designed well. I would guess that the tameness comes from the non stimulating environment more than the process of selection. It is known that mother imprints a lot of hormonal/stress activities to children while pregnant. Just like starving pregnant mothers have an effect on generations that come through hormonal gene activation similar thing happens to animals in wilderness. Is there a line of foxes bread for generations in captivity with purely random selection?

I believe Belyayev had a control group, yes.

Still, the nature of the control group might have a smaller taming effect needing longer time to turn into tame.

We've discussed this a few times on HN, but old discussions are a bit hard to search for. Here are a couple: https://news.ycombinator.com/item?id=10517631 https://news.ycombinator.com/item?id=12493059

An interesting aspect of this experiment often overlooked: the same process influences us.

The way that we behave in a corporate setting seems closely related to domestication. It would be worth examining the traits associated with animal domestication and compare them to human behavior in a variety of contexts.

It's not really a happy thought that we're domesticated, but power structures exist, and it's in most of our genetics to serve. It's often possible to trace certain types of ambition to a young age, for example. It would be neat to see a more rigorous exploration of whether it's true that humans are domesticated, and if so, how much, and what it means in a precise way.

Human reproduction is based on choices made by individual men and women. "Corporate setting" or not, no central authority is selecting us for mating based on some characteristic.

With domesticated animals, often a stud is selected and can have hundreds of offspring. For example, all border collies are related to a single dog (Old Hemp). There's no comparison to humans.

Many of us seem to be related to Genghis Khan: https://www.discovermagazine.com/the-sciences/1-in-200-men-d...

For me, the question of human domestication goes beyond breeding choices. We are often a product of our genetics. But once we have those genetics, what then? Can societal structures influence behavior in the same direction of animal domestication? And if not, what’s the crucial distinction, since animal domestication seems closely related to many human behaviors we see today?

Imagine the strong selective pressure as humans moved to larger social groups and agrarian societies. These were ruled by strongmen interested above all else in most cases in collecting the highest taxes from a compliant population. Anyone too fierce, fearful or combative would be ostracized at best, killed at worst.

Even today, prisons and mental institutions are filled with non-conformist, anti-social people who in many cases are not allowed to reproduce.

OTOH, "undomesticated" males tend to be good as generating more offspring with diverse partners than "domesticated" ones with 2.5 children.

Human self-domestication is certainly widely discussed: https://www.sciencedaily.com/releases/2018/02/180215110041.h...

I figure that most humans alive today are the round-faced, more sociable, less aggressive strains. We've domesticated outselves because those more sociable and able to get along with their fellow humans have more children with better survival rates.

> The relation between the controller and the controlled is reciprocal. The scientist in the laboratory, studying the behavior of a pigeon, designs contingencies and observes their effects. His apparatus exerts a conspicuous control on the pigeon, but we must not overlook the control exerted by the pigeon. The behavior of the pigeon has determined the design of the apparatus and the procedures in which it is used. Some such reciprocal control is characteristic of all science. As Francis Bacon put it, nature to be commanded must be obeyed. The scientist who designs a cyclotron is under the control of the particles he is studying. The behavior with which a parent controls his child, either aversively or through positive reinforcement, is shaped and maintained by the child's responses. A psychotherapist changes the behavior of his patient in ways which have been shaped and maintained by his success in changing that behavior. A government or religion prescribes and imposes sanctions selected by their effectiveness in controlling citizen or communicant. An employer induces his employees to work industriously and carefully with wage systems determined by their effects on behavior. The classroom practices of the teacher are shaped and maintained by the effects on his students. In a very real sense, then, the slave controls the slave driver, the child the parent, the patient the therapist, the citizen the government, the communicant the priest, the employee the employer, and the student the teacher.

-- B F Skinner

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact