If your name is Brittany, it’s likely that you were born around 1998, plus or minus five years. That’s when about 75 percent of Brittanys were named, in a surge of parents deciding on the same first name for their child. Dalmations were also popular pets in the 1990s, but not so much any more, and the last big swell in greyhound dogs was in the 1940s.
Why do certain cultural traits, like names or pets, go through cycles of being immensely popular at some times but not others? It’s not necessarily because of some intrinsic quality of a name like Brittany or the breed of a dog. How frequently that cultural trait occurs can impact how much it spreads, or if it stops spreading, said Mitchell Newberry, an assistant professor of complex systems at the University of Michigan. In a recent study in Nature Human Behavior, Newberry and his co-author, Joshua Plotkin, demonstrated this by looking at large datasets on baby names and dog breeds.
Their findings suggest that baby names and dog breeds are subject to something called “frequency-dependent selection,” which is when the spread of a name or dog is dependent on how much of it is out there. This means that we, in part, choose to name our kids or pick the dogs to take into our homes depending on how many other people are using that same name or adopting those dogs.
“Frequency-dependent selection is just a fancy way of saying that there is a pressure to be the same or different,” Newberry said. “If a name gets too popular, then people don’t want to use it anymore, and there’s a pressure to be different.”
Besides offering some insights into our names and pets, understanding frequency dependence selection can provide clues into larger evolutionary patterns. Frequency dependence is a concept that comes directly from evolutionary biology, where it’s thought to play a role in a wide variety of outcomes like genetic diversity, immune escape, sex ratios, mating preferences, and speciation. Alex Stewart, a senior lecturer in mathematical biology at University of St Andrews, said understanding frequency dependence has relevance in other areas too, like how we understand misinformation, for example.
“It’s well documented that one of the things that causes pieces of misinformation to spread is novelty,” he said. “It strikes people as exciting and new, even if it's not particularly believable, it's much more likely to spread.” That’s both a function of the content of the information, but it's also a function of its frequency—if everybody's making the same kinds of outrageous claims, then it's less novel. The methods in this research, which model the relationship between selection and frequency, could be used on data about online behavior, belief in conspiracy theories, radicalization, or the rise and fall of political parties. With baby names and dog breeds, the researchers had a cleaner and easier data sample to examine, which offered them the chance to study frequency dependent selection works.
Using names recorded by the Social Security Administration since 1935, the authors analyzed 245 million names that occured in at least one in 10,000 births. Newberry and Plotkin found that when a name is rare, it grows by around 1.4% per year, but that common names (one in 100) decrease by 1.6% a year. This is called negative frequency dependence, when selection goes down with too much frequency. Names that were around 1/1000 births don’t tend to change frequency, finding a sweet spot between too common or rare. These findings were consistent across multiple countries they looked at, despite countries having different popular names; they also examined name trends in France, the Netherlands, and Norway.
There are some popular names like John and William that tend to sit around at a higher frequency, but are still relatively lower one compared to a trendy name that’s at its peak.
Trendy names tend to go through “boom-bust” cycles, where they rise steeply in popularity and then drop off. Some names crest and fall multiple times throughout different generations: Emma was popular in the late 1800s, and is rising again now. But after the name Linda became popular in the 1940s, it has so far stayed out of favor. Biblical names are an exception: they tend to spread well, despite their frequency.
There can be other cultural factors at play too that cause an explosion in a name: After the 1957 TV show Maverick came out, people began to name their babies Maverick. But when Top Gun was released in 1986, it caused an explosion in babies named Maverick, including female babies named Maverick. “I started to see the number of Mavericks born every year exponentially increasing, and it continued exponentially increasing until today when it's in the top 50 names in the US,” Newberry said. (Now, with a new Top Gun movie out, Newberry said it could receive another push.)
The dog data came from a database of purebred dog registrations from the American Kennel Club. In dogs too, they found that dogs that are rarer become popular faster than dogs that are more common. In the 1990s, dalmatians were a popular breed of dog; now they are a tenth as popular. They saw a greyhound spike in the 1940s and a rottweiler trend in the 1990s. “People may have reasons they think they chose their dog, but basically they're just copying what other people are doing,” Newberry said.
This was found in a paper from 2013 as well—researchers looked at the characteristics of dog breeds and their popularity between 1926 and 2005. They found that the popularity of a breed had no relationship to a breed’s “health, longevity, and behavioral qualities such as aggressiveness, trainability, and fearfulness.”
“Our results support the hypothesis that dog breed popularity has been primarily determined by fashion rather than function,” the authors wrote—meaning social influence was a bigger factor than a person picking a dog because they liked its qualities.
“Frequency dependent selection is a really fundamental part of evolution,” said Mark Pagel, an evolutionary biologist at the University of Reading who wasn’t involved in the new study. Pagel has done work on how frequency dependence impacts language evolution: “A real puzzle of language is how we agree on the words we use,” he said.
He and his colleagues modeled the frequency with which a word is used to convey a meaning, and found that there is a positive frequency dependence. As a group of people start using a word to convey meaning, other people observe that and start using that word; more people are likely to use a word to convey meaning if others are. This is the opposite of baby names. Pagel said to think of it like this: Negative frequency dependence generates diversity, whereas positive frequency dependence generates agreement or cooperation.
"You could look at the manifest content of their paper and say, 'Oh, baby names, dog breeds, who cares?'” Pagel said. “But it tells us something really fundamental about human sociality and human cognition.” Not only does it show us that we want our children to be distinctive and stand out, but it’s a lesson in how we integrate information from the world around us.
For instance, the paper shows the point at which a baby's name stops growing in frequency, but isn’t yet declining. “That means your brain is searching around all these people, and when you're thinking about naming your child you're saying, 'Wow, there's three people out there out of those thousand; I think I'm not going to name my child [that name],” Pagel said. “That tells us that we are extraordinarily sensitive to these stimuli in our environment.”
When we learn about evolution, there can be a focus on the idea that evolution is an optimization process, Stewart said—that it’s finding the thing that’s the best fit for the environment, and is going through a process to search that out and produce it in a population. “That can be true,” Stewart said. But there are other factors, especially when taking into account social behaviors, that are dependent more on qualities like frequency—or simply how common something is—that cause it to spread or not spread.
We often think that we make decisions like naming children or picking a dog based on our own individual agency or preference, but this work shows how even intimate decisions are influenced by others, and in a systematic way.
“When I lecture about this, I say to people they should be astonished by this, that something that seems so individualistic, that you think you're a free agent and you make up your mind and so on,” Pagel said. “In fact, when we embed you in a larger society, you behave according to rules that we can point out.”
Follow Shayla Love on Twitter.