What internet searches reveal about human sexuality

Polls never reveal who we really are. Google does.

What internet searches reveal about human sexuality
Photo by Tim de Waele/Corbis via Getty Images
  • In Everybody Lies, Seth Stephens-Davidowitz discusses how Pornhub and Google data provide a window into human sexuality.
  • Large, anonymized data sets are more reliable indicators than polling or other traditional methods.
  • More people report having sex than actually having sex, Stephens-Davidowitz reports.

Polls are not the best indicator of reality, as was shown in the 2016 presidential race. There are many reasons. People might not want to admit the truth. They might purposefully fib in order to skew data. They simply lie. Everybody lies, as data scientist and economist Seth Stephens-Davidowitz claims in his 2017 book.

What polls will not reveal can be discovered with another data source: Google searches. Or, as Stephens-Davidowitz spends a fair portion of his book covering, Pornhub searches. The internet delivers a treasure trove of data impossible to secure with other means (such as polling). Acquiring data sets in the thousands can be prohibitively expensive; with Google, Pornhub, and others, you can secure data sets in the billions, often at no expense beyond time.

Of course, such large data sets come with their own problems. A big one, which Viktor Mayer-Schonberger and Kenneth Cukier address in their 2013 book, Big Data, is that we have to leave behind our fascination with causation—a biologically hard-wired feature—and settle for correlation.

"In a big-data world...we won't have to be fixated on causality; instead we can discover patterns and correlations in the data that offer us novel and invaluable insights."

Everybody lies; Google searches are the truth. In the comfort of our own home angels and demons appear, naked and honest. Search giants anonymize the data, yet these sets pull back the curtain enough to allow us to reckon with the animals that we truly are. As Stephens-Davidowitz puts it,

"Sometimes we need internet data to correct our instinct to pat ourselves on the back."

You have doppelgängers. They’re quietly influencing your life. | Seth Stephens-Davidowitz

While he reports frightening data on white nationalism and racism—"n word" queries and Trump voters match up nearly perfectly across the nation—it is our—yes, male and female, though mostly male—fascination with pornography that offers clues to the nature of human sexuality.

Before continuing, let me state that I take no issue with people's sexual preferences. The point isn't shame; it's curiosity regarding the distance between what we present ourselves to be and who we really are. That said, the increasing numbers of child pornography available online—over 45 million reported cases last year alone—is a serious issue that involves every one of us. Living in a world where this flourishes does not bode well for the future of sexuality and, therefore, the entire race.

The other reason this is important is to further the cultivation of honesty in relationships. Too many people cheat and lie to their partners; others cannot properly express their desires and therefore feel intimately and sexually unfulfilled, a topic beautifully explored in Daniel Bergner's book, The Other Side of Desire. Opening up about desires can be therapeutic and should not be condemned, but condoned. Ambitious, certainly, though possible.

Until then we have a massive influx of pornography, which has long been a technological driver. It pushed forward the development of the camera. It is one of the reasons that VHS crushed Betamax, as well as an influence behind the mass adoption of the internet. Even still, many won't fess up. Stephens-Davidowitz writes that more people search for "porn" than they do "weather" even though only one-quarter of men and 8 percent of women admit to it.

Old school ideology from Alfred Kinsey states that 10 percent of American men are gay. Surveys believe the number is between 2-3 percent. Using a wealth of Facebook data, including location information spread across more tolerant and less tolerant states, Stephens-Davidowitz states that the number is closer to 5 percent, not including closeted homosexuals.

Women are 10 percent more likely to search on Google for "Is my husband gay?" than "Is my husband cheating?" Women wonder if their husband is gay eight times more than if he's an alcoholic and ten times more than if he suffers from depression. The states where this question is most asked are South Carolina and Louisiana.

Speaking of female sexuality, 20 percent of all videos watched by women on Pornhub show lesbian scenes.

Seth Stephens-Davidowitz pauses to have his portrait taken while he works on his elections predictions at his apartment.

Photo by Kayana Szymczak for The Boston Globe via Getty Images

On the animated show, "Big Mouth," Missy loves humping stuffed animals. There is precedent: while it's not a particularly popular search on Pornhub, women do look for such videos.

Twenty-five percent of female searches involving heterosexual porn on Pornhub involve pain and/or humiliation. Five percent search for nonconsensual sex. Women are twice as likely to search for these topics as men. As Stephens-Davidowitz points out, we should differentiate between fantasy and reality when considering such searches. Still, data don't lie.

Using Google data, it turns out that searching why a partner doesn't want sex is sixteen times more popular than a partner unwilling to talk. With unmarried partners, not wanting sex is five-and-a-half times more popular than not texting back. Women are twice as likely to search for boyfriends not wanting sex than the reverse.

Men Google about penis size more than all other body parts combined. Women rarely look it up. In fact, for every one woman that searches, 170 men enter it into the search field. Even then, 40 percent of women complain that it's too big. Yet only 1 percent of men seek information on how to make it smaller.

Men disproportionally ask how to make sex longer. Women search for making their partner climax quicker as much as they want to stretch out the experience.

For women, breast size is a much more common search: seven million breast implant queries a year. When it comes to their behind, until 2014 the searches always dealt with making it smaller. In 2014, that switched: in every state, women searched for a bigger behind more often.

The number one search for women and men when dealing with female genitalia concerns smell.

Finally, what turned out to be one of Stephen-Davidowitz's favorite pieces of data, and one that does not fly in the face of convention but rather confirms it:

"Men make as many searches looking for ways to perform oral sex on themselves as they do how to give a woman an orgasm."


Stay in touch with Derek on Twitter and Facebook.

U.S. Navy controls inventions that claim to change "fabric of reality"

Inventions with revolutionary potential made by a mysterious aerospace engineer for the U.S. Navy come to light.

U.S. Navy ships

Credit: Getty Images
Surprising Science
  • U.S. Navy holds patents for enigmatic inventions by aerospace engineer Dr. Salvatore Pais.
  • Pais came up with technology that can "engineer" reality, devising an ultrafast craft, a fusion reactor, and more.
  • While mostly theoretical at this point, the inventions could transform energy, space, and military sectors.
Keep reading Show less

COVID and "gain of function" research: should we create monsters to prevent them?

Gain-of-function mutation research may help predict the next pandemic — or, critics argue, cause one.

Credit: Guillermo Legaria via Getty Images

This article was originally published on our sister site, Freethink.

"I was intrigued," says Ron Fouchier, in his rich, Dutch-accented English, "in how little things could kill large animals and humans."

It's late evening in Rotterdam as darkness slowly drapes our Skype conversation.

This fascination led the silver-haired virologist to venture into controversial gain-of-function mutation research — work by scientists that adds abilities to pathogens, including experiments that focus on SARS and MERS, the coronavirus cousins of the COVID-19 agent.

If we are to avoid another influenza pandemic, we will need to understand the kinds of flu viruses that could cause it. Gain-of-function mutation research can help us with that, says Fouchier, by telling us what kind of mutations might allow a virus to jump across species or evolve into more virulent strains. It could help us prepare and, in doing so, save lives.

Many of his scientific peers, however, disagree; they say his experiments are not worth the risks they pose to society.

A virus and a firestorm

The Dutch virologist, based at Erasmus Medical Center in Rotterdam, caused a firestorm of controversy about a decade ago, when he and Yoshihiro Kawaoka at the University of Wisconsin-Madison announced that they had successfully mutated H5N1, a strain of bird flu, to pass through the air between ferrets, in two separate experiments. Ferrets are considered the best flu models because their respiratory systems react to the flu much like humans.

The mutations that gave the virus its ability to be airborne transmissible are gain-of-function (GOF) mutations. GOF research is when scientists purposefully cause mutations that give viruses new abilities in an attempt to better understand the pathogen. In Fouchier's experiments, they wanted to see if it could be made airborne transmissible so that they could catch potentially dangerous strains early and develop new treatments and vaccines ahead of time.

The problem is: their mutated H5N1 could also cause a pandemic if it ever left the lab. In Science magazine, Fouchier himself called it "probably one of the most dangerous viruses you can make."

Just three special traits

Recreated 1918 influenza virionsCredit: Cynthia Goldsmith / CDC / Dr. Terrence Tumpey / Public domain via Wikipedia

For H5N1, Fouchier identified five mutations that could cause three special traits needed to trigger an avian flu to become airborne in mammals. Those traits are (1) the ability to attach to cells of the throat and nose, (2) the ability to survive the colder temperatures found in those places, and (3) the ability to survive in adverse environments.

A minimum of three mutations may be all that's needed for a virus in the wild to make the leap through the air in mammals. If it does, it could spread. Fast.

Fouchier calculates the odds of this happening to be fairly low, for any given virus. Each mutation has the potential to cripple the virus on its own. They need to be perfectly aligned for the flu to jump. But these mutations can — and do — happen.

"In 2013, a new virus popped up in China," says Fouchier. "H7N9."

H7N9 is another kind of avian flu, like H5N1. The CDC considers it the most likely flu strain to cause a pandemic. In the human outbreaks that occurred between 2013 and 2015, it killed a staggering 39% of known cases; if H7N9 were to have all five of the gain-of-function mutations Fouchier had identified in his work with H5N1, it could make COVID-19 look like a kitten in comparison.

H7N9 had three of those mutations in 2013.

Gain-of-function mutation: creating our fears to (possibly) prevent them

Flu viruses are basically eight pieces of RNA wrapped up in a ball. To create the gain-of-function mutations, the research used a DNA template for each piece, called a plasmid. Making a single mutation in the plasmid is easy, Fouchier says, and it's commonly done in genetics labs.

If you insert all eight plasmids into a mammalian cell, they hijack the cell's machinery to create flu virus RNA.

"Now you can start to assemble a new virus particle in that cell," Fouchier says.

One infected cell is enough to grow many new virus particles — from one to a thousand to a million; viruses are replication machines. And because they mutate so readily during their replication, the new viruses have to be checked to make sure it only has the mutations the lab caused.

The virus then goes into the ferrets, passing through them to generate new viruses until, on the 10th generation, it infected ferrets through the air. By analyzing the virus's genes in each generation, they can figure out what exact five mutations lead to H5N1 bird flu being airborne between ferrets.

And, potentially, people.

"This work should never have been done"

The potential for the modified H5N1 strain to cause a human pandemic if it ever slipped out of containment has sparked sharp criticism and no shortage of controversy. Rutgers molecular biologist Richard Ebright summed up the far end of the opposition when he told Science that the research "should never have been done."

"When I first heard about the experiments that make highly pathogenic avian influenza transmissible," says Philip Dormitzer, vice president and chief scientific officer of viral vaccines at Pfizer, "I was interested in the science but concerned about the risks of both the viruses themselves and of the consequences of the reaction to the experiments."

In 2014, in response to researchers' fears and some lab incidents, the federal government imposed a moratorium on all GOF research, freezing the work.

Some scientists believe gain-of-function mutation experiments could be extremely valuable in understanding the potential risks we face from wild influenza strains, but only if they are done right. Dormitzer says that a careful and thoughtful examination of the issue could lead to processes that make gain-of-function mutation research with viruses safer.

But in the meantime, the moratorium stifled some research into influenzas — and coronaviruses.

The National Academy of Science whipped up some new guidelines, and in December of 2017, the call went out: GOF studies could apply to be funded again. A panel formed by Health and Human Services (HHS) would review applications and make the decision of which studies to fund.

As of right now, only Kawaoka and Fouchier's studies have been approved, getting the green light last winter. They are resuming where they left off.

Pandora's locks: how to contain gain-of-function flu

Here's the thing: the work is indeed potentially dangerous. But there are layers upon layers of safety measures at both Fouchier's and Kawaoka's labs.

"You really need to think about it like an onion," says Rebecca Moritz of the University of Wisconsin-Madison. Moritz is the select agent responsible for Kawaoka's lab. Her job is to ensure that all safety standards are met and that protocols are created and drilled; basically, she's there to prevent viruses from escaping. And this virus has some extra-special considerations.

The specific H5N1 strain Kawaoka's lab uses is on a list called the Federal Select Agent Program. Pathogens on this list need to meet special safety considerations. The GOF experiments have even more stringent guidelines because the research is deemed "dual-use research of concern."

There was debate over whether Fouchier and Kawaoka's work should even be published.

"Dual-use research of concern is legitimate research that could potentially be used for nefarious purposes," Moritz says. At one time, there was debate over whether Fouchier and Kawaoka's work should even be published.

While the insights they found would help scientists, they could also be used to create bioweapons. The papers had to pass through a review by the U.S. National Science Board for Biosecurity, but they were eventually published.

Intentional biowarfare and terrorism aside, the gain-of-function mutation flu must be contained even from accidents. At Wisconsin, that begins with the building itself. The labs are specially designed to be able to contain pathogens (BSL-3 agricultural, for you Inside Baseball types).

They are essentially an airtight cement bunker, negatively pressurized so that air will only flow into the lab in case of any breach — keeping the viruses pushed in. And all air in and out of the lap passes through multiple HEPA filters.

Inside the lab, researchers wear special protective equipment, including respirators. Anyone coming or going into the lab must go through an intricate dance involving stripping and putting on various articles of clothing and passing through showers and decontamination.

And the most dangerous parts of the experiment are performed inside primary containment. For example, a biocontainment cabinet, which acts like an extra high-security box, inside the already highly-secure lab (kind of like the radiation glove box Homer Simpson is working in during the opening credits).

"Many people behind the institution are working to make sure this research can be done safely and securely." — REBECCA MORITZ

The Federal Select Agent program can come and inspect you at any time with no warning, Moritz says. At the bare minimum, the whole thing gets shaken down every three years.

There are numerous potential dangers — a vial of virus gets dropped; a needle prick; a ferret bite — but Moritz is confident that the safety measures and guidelines will prevent any catastrophe.

"The institution and many people behind the institution are working to make sure this research can be done safely and securely," Moritz says.

No human harm has come of the work yet, but the potential for it is real.

"Nature will continue to do this"

They were dead on the beaches.

In the spring of 2014, another type of bird flu, H10N7, swept through the harbor seal population of northern Europe. Starting in Sweden, the virus moved south and west, across Denmark, Germany, and the Netherlands. It is estimated that 10% of the entire seal population was killed.

The virus's evolution could be tracked through time and space, Fouchier says, as it progressed down the coast. Natural selection pushed through gain-of-function mutations in the seals, similarly to how H5N1 evolved to better jump between ferrets in his lab — his lab which, at the time, was shuttered.

"We did our work in the lab," Fouchier says, with a high level of safety and security. "But the same thing was happening on the beach here in the Netherlands. And so you can tell me to stop doing this research, but nature will continue to do this day in, day out."

Critics argue that the knowledge gained from the experiments is either non-existent or not worth the risk; Fouchier argues that GOF experiments are the only way to learn crucial information on what makes a flu virus a pandemic candidate.

"If these three traits could be caused by hundreds of combinations of five mutations, then that increases the risk of these things happening in nature immensely," Fouchier says.

"With something as crucial as flu, we need to investigate everything that we can," Fouchier says, hoping to find "a new Achilles' heel of the flu that we can use to stop the impact of it."

The misguided history of female anatomy

From "mutilated males" to "wandering wombs," dodgy science affects how we view the female body still today.

Credit: Hà Nguyễn via Unsplash
Sex & Relationships
  • The history of medicine and biology often has been embarrassingly wrong when it comes to female anatomy and was surprisingly resistant to progress.
  • Aristotle and the ancient Greeks are much to blame for the mistaken notion of women as cold, passive, and little more than a "mutilated man."
  • Thanks to this dubious science, and the likes of Sigmund Freud, we live today with a legacy that judges women according to antiquated biology and psychology.
Keep reading Show less
Mind & Brain

Why do holidays feel like they're over before they even start?

People tend to reflexively assume that fun events – like vacations – will go by really quickly.