But there may be more to it than that. People who have tried to simply tame individual foxes often speak of a stubborn wildness that is impossible to get rid of. This suggests that foxes harder to tame than other animals.
In the late 1950s, a Russian geneticist called Dmitry K. Belyaev attempted to create a tame fox population.
Through the work of a breeding programme at the Institute of Cytology and Genetics at Novosibirsk, in Russia, he sought to trace the evolutionary pathway of domesticated animals. His test subjects were silver-black foxes, a melanistic version of the red fox that had been bred in farms for the colour of their fur.
Belyaev died in 1985, but the project is still ongoing. It is now overseen by Lyudmila Trut, now in her 80s, who started out as Belyaev's intern.
First, Belyaev and Trut travelled to various fur farms in the Soviet Union, from Siberia to Moscow and Estonia. There, they chose foxes to take to their own farm in Novosibirsk.
They selected the animals based on how they responded when their cage was opened. About 10% of the foxes displayed a weak "wild-response", meaning they were docile around humans.
Animals that were friendlier and tolerant to human touch, even to a small degree, were picked out.
Of those friendly foxes, 100 vixens and 30 males were chosen as the first generations of parents.
When the cubs were born, the researchers hand-fed them. They also attempted to touch or pet the foxes when they were two to two-and-a-half months old, for strictly measured periods at a time.
In each selection, less than 10% of tame individuals were used as parents of the next generation.
"As a result of such rigorous selection, the offspring exhibiting the aggressive and fear avoidance responses were eliminated from the experimental population in just two to three generations of selection," Trut wrote in a study published in 2009.
BioEssays -Animal evolution during domestication: the domesticated fox as a model
There was no domestication training. Only selection for breeding was used.
"Belyaev had one main goal at the beginning of experiment: to reproduce the process of historical domestication at the experiment, during a short time," says Trut. "This goal didn't change. But during the experiment the understanding of evolutionary process changed."
By the fourth generation, the scientists started to see dramatic changes.
The cubs were beginning to behave more like dogs. They wagged their tails and "eagerly" sought contact with humans. They whined, whimpered and licked researchers just like puppies would.
The process was surprisingly quick. "By intense selective breeding, we have compressed into a few decades an ancient process that originally unfolded over thousands of years," wrote Trut in 1999.
These foxes were called the "elite of domestication", and as the generations passed the proportion of these elite cubs grew. By 2005-2006, almost all the foxes were playful, friendly and behaving like domestic dogs. The foxes could "read" human cues and respond correctly to gestures or glances. The vocalisations they made were different to wild foxes.
The fox experiment showed that just by selecting for friendliness, all these other changes, including an increase in social skills, happened by accident."
The domesticated foxes had floppier, drooping ears, which are found in other domestic animals such as dogs, cats, pigs, horses and goats. Curlier tails – also found in dogs and pigs – were also recorded.
As of August 2016, there are 270 tame vixens and 70 tame males on the farm.
In the 1990s, the institute supported itself by selling fox pelts. At the end of the 1990s, they started to sell the foxes as house pets. At present, a Florida-based company called the Lester Kalmanson Agency Inc imports foxes for those who want to keep them as pets. Each fox costs $8,900, because of the delivery costs.
With the foxes now tame, the researchers are trying to identify the genes that change under selection for tameness. "The main current goals are focused on molecular-genetics mechanisms of domestic behavior," says Trut.
Could this be applied to Artificial Intelligence
There has been considerable concern about the need to develop friendly AI. Could the "personality" of AI that is not more intelligent than humans be selected for friendliness first before the AI is developed for more intelligence ?
SOURCES- BioEssays, BBC News