A new twist in the debate over pornography’s role in romantic relationships.

porn does to men failed to replicate.

The latest research on pornography suggests that earlier work may be off the mark.


It’s strange timing for a pivot into prudery, but the GOP—now led by our pussy-grabbing, naked-interview–giving, sex-tape–watching, Playboy-video–starring president-elect—is poised to launch a crackdown on pornography. If the last few months are any guide, the new attack on naked pictures will have less to do with morals than with mental health. Last summer, the party wrote a plank into its platform calling online smut a “public health crisis”—a phrase that hints at underlying epidemiology. The same phrase had turned up back in April, when the heavily Republican government of Utah passed a resolution by unanimous vote citing “recent research” to declare pornography a “public health crisis” and “hazard leading to a broad spectrum of individual and public health impact and societal harms.”

Writing in the Washington Post, sociologist Gail Dines supported this idea, endorsing Utah’s declaration on scientific grounds. “The science is there,” she wrote. Forty years of peer-reviewed publications have shown the many ways that porn can hurt us: It makes men more aggressive and more inclined to sexual violence; it reduces women’s self-esteem; it weakens our relationships. “By taking a health-focused view of porn and recognizing its radiating impact not only on consumers but also on society at large,” said Dines, “Utah’s resolution simply reflects the latest research.”

Yet the very latest research on pornography suggests that earlier work may be off the mark. Psychologists in Canada just put out their attempt to replicate one of the field’s most famous studies of pornography’s harmful effects, a textbook finding dating from the porn-war years of the 1980s, made by the prominent evolutionary psychologist Douglas Kenrick. According to that work, viewing naked pictures from Playboy makes heterosexual men less attracted to their wives or girlfriends, even less in love with them. The mere sight of such beautiful and apparently licentious women seems to distort the way that men relate to their partners, Kenrick concluded. But according to the new results, this may not be true at all. If that’s the case then the theory that there is a “public health crisis” of pornography will have run afoul of the replication crisis in psychology.

The Kenrick study does not show up in every disquisition on the dangers of pornography, but its central message—that porn addles the male brain and causes men to lose interest in their partners—has been passed along quite freely. “The more they watch, the more they want porn sex because they become habituated to that kind of industrial-strength sex,” said Dines, who started the group Stop Porn Culture. “What I find is that some men lose interest in their partners altogether and use more pornography.” Kenrick, for his part, had not set out to show the social cost of viewing naked pictures. “I wasn’t really interested in the pornography end of it,” he said last week. Rather, he was studying a perceptual phenomenon called the “contrast effect” and extending it to social situations.

When you stick your hand in a bucket of cold water, he told me, the next thing you touch will feel somewhat warmer than it really is. So Kenrick thought: What if the same principle applied to judgments of attractiveness—when you stare at someone beautiful, will the next person you see seem plainer in comparison? For a 1977 study on what could be called the scientific version of the “ugly friend effect,” he had asked a bunch of male students at Montana State University to rate the beauty of an average-looking woman in a photograph after having watched an episode of Charlie’s Angels. The guys said the woman was less attractive, on average, than a control group who had not been watching television. The Charlie’s Angels viewers, wrote Kenrick, may have been more negative on account of “the immediate influence of the beautiful media stimuli.”

His original Playboy centerfold study from 1989—the subject of the recent replication—aimed to test the same idea with naked pictures. Kenrick told a group of heterosexual participants he was studying judgments of controversial art, then asked them to give opinions of either abstract paintings (by Jackson Pollock, Josef Albers, and the like) or nude centerfolds (drawn from Playboy for the men and Playgirl for the women). Finally he had them answer a bunch of questions about the desirability and sexual attractiveness of their romantic partners and the degree to which they felt devoted to that person.

The men seemed to end up less attracted and devoted to their partners after looking at the centerfolds. No effect was recorded in the women. Kenrick offered up an evolutionary explanation: By looking at these beautiful women, the men might have been led to believe they had more and better options for mating. “When I see all of these beautiful women in Playboy, it changes my assumption about the universe of possibilities,” he explained. In light of their new, porn-distorted understanding of the world, the men were less inclined to stay committed to their wives or girlfriends. Women, he theorized, are less susceptible to this effect because they’ve evolved to be less attuned to physical attractiveness. “The lesson I took out of my own research is that I stopped looking at Playboy,” Kenrick said, “and I didn’t really miss it.”

In the years that followed, other researchers tried to figure out if looking at pornography does lasting harm to couples. Survey studies—many citing Kenrick’s work—found a somewhat consistent negative correlation between porn viewership and various measures of relationship satisfaction. On average, the men in these surveys who said they looked at pornography were less likely to report being happy, stable, communicative, or sexually satisfied with their partners.

One recent study, for example, from a pair of researchers at religious universities (Notre Dame and Brigham Young), used data from 20,000 adult participants in the General Social Survey. In every year since 1973, that survey has asked people about their marital stability and happiness, as well as whether they had watched an “X-rated movie” at some point in the past 12 months. (In 2000 they added a more up-to-date question about “visiting pornographic websites.”) Overall, 26 percent of the men said they had watched a pornographic film in the last year, versus 17 percent of the women. When the researchers compared the porn-watching men with porn abstainers, they turned out to be 60 percent more likely to be divorced, 80 percent more likely to have had an extramarital affair, and 15 percent less likely to describe their marriage as “very happy.”

But it still wasn’t clear if the association between porn use and relationship quality would be the same for everyone or if it would apply to every kind of porn. A few studies hinted, for example, that if porn caused harm, it would be more damaging to men than women. The context for viewing pornography could also make a difference in how it affected relationships. A recent mail-in survey of 1,291 unmarried people found that people who reported no viewing of sexually explicit material also said they were a bit more dedicated to their partners, more sexually satisfied, and better able to communicate than the porn users. But the researchers found that these small differences mostly disappeared among those who said they looked at pornography with their partners, as opposed to by themselves.

It could be that pornography leads relationships to falter, but what if the surveys had it backwards? A solitary porn habit, for example, could be taken as a symptom of a pre-existing problem in a relationship. As a rule, these studies can’t tell us whether using porn really messes up a marriage, or if people in busted relationships are more likely to start using porn (or admit to using porn when asked). Either scenario would produce the same correlation in a survey. It’s only by appealing to work like Kenrick’s, showing clear effects from viewing naked pictures in the lab, that one can draw a more definitive, anti-porn conclusion.

A few researchers tried to get around this correlation-or-causation problem by tracking porn habits and measures of relationship satisfaction over several years. Sociologist Sam Perry of the University of Oklahoma has been analyzing data from the Portraits of American Life study, a set of in-person interviews focused on religious values that began in 2006 and was followed up on in 2012. More than 1,000 subjects were asked to rate the happiness of their marriages, their level of satisfaction with the love and affection they receive from their partners, and the frequency with which they viewed pornographic materials over the previous 12 months.

Perry found the old association: People who said they looked at porn also said they were less happy in their marriages. But when he compared ratings from 2006 and 2012, he found that earlier pornography consumption seemed to predict lower relationship ratings down the line. That held true even when he matched people up according to the quality of their relationships at the beginning of the study. (In other words, it didn’t seem as though bad relationships, in particular, were driving people to use porn.)

Still, this study doesn’t offer much in the way of explanation. If it’s true that using pornography can lead to relationship dysfunction, is that because it produces an automatic, psychological response—something like Kenrick’s contrast effect—deforming the male perspective and biasing his behavior? In that case, we’d be better off with no porn whatsoever or even spending time in careful study of unattractive men and women. That way we’d make our own, average partners seem more special by comparison. “I do think there’s a tendency for hedonism to backfire on you,” Kenrick told me. “My guess is that people are disappointed in their real lives.”

But what if pornography only harms relationships to the extent that it’s taboo? If your spouse thinks of masturbation as a form of cheating, or disapproves of naked pictures, or even worries (à la Kenrick) that porn will make you less devoted, then you might feel anxiety and shame at masturbating over naked pictures. That anxiety and shame could in turn degrade your marriage. You might also be discovered using porn, yielding conflict more directly. So viewing porn can end up being harmful, as shown in Perry’s study, but only on account of the values that we place on it. (By the same token, the consumption of shrimp cocktails could be shown to break families apart—and characterized as a “public health crisis”—if researchers happened to be studying a community of Orthodox Jews.)

Lorne Campbell, a psychologist at the University of Western Ontario and senior author on the Kenrick replication, notes several other problems in pornography research. For one thing, most studies look at individuals, rather than couples. They tend to make gendered assumptions about the frequency of viewing porn and its purpose. (For example, very little attention has been paid to women who make solitary use of porn while in relationships.) And most studies, Campbell argues, have set out to find evidence of harm, neglecting the possibility that the use of porn could make some relationships stronger. “The associations are out there, but they’re not that strong,” he said, referring to the published links between using porn and lower-quality relationships. “They leave room for moderation. Maybe it’s different for different people.”

The lack of solid evidence hasn’t stopped the message that pornography is damaging to relationships. Kenrick’s work has been used to push the idea that women should be afraid of their partners using porn, said the lead author on the replication paper and one of Campbell’s graduate students, Rhonda Balzarini. The plan to redo Kenrick’s experiment came out of Campbell’s seminar on research methodology. For a class project, three students from his lab—Balzarini, Kiersten Dobson, and Kristi Chin—hoped to demonstrate the same effect that Kenrick had, so they could extend it. One idea was to show people naked pictures and then to measure their physiological responses to their partners. (If a guy who looks at Playboy centerfolds rates his wife as less attractive, will he also be less aroused by her in person?)

Balzarini reached out to Kenrick last September and asked him for advice on how to replicate his work. He suggested using recent centerfolds, in place of the ones he’d used in the 1980s. (You can see the NSFW experimental stimuli here.) The students recruited 223 subjects from the internet for their study—more than triple the number from the original—and ran the experiment remotely, through the internet. (Kenrick had invited campus subjects to come into his lab.) Then they ran the study and analyzed their data—and came up with nothing. The replication found no sign of the original effect.

Kenrick described himself as being “shocked” by the Canadian group’s inability to reproduce his result. “I trust this finding,” he said, referring to the original study. “I analyzed this data myself.” Still, he respects the recent effort, calling it “one of the cleaner, nicer nonreplications.” He even served as a peer reviewer for the students’ final paper.

So why might the replication have failed?  Kenrick has a few ideas. It could be that the internet subjects weren’t as engaged with the materials. Some may have been looking at the pictures in public places—sitting with their laptops in cafés, perhaps—which made them so uncomfortable that they tried to stifle their responses. Or else it could be that the subjects, drawn from a relatively small pool of workers through Amazon’s Mechanical Turk, have gotten jaded from participating in so many online psychology studies.

Kenrick also said that the original effect may have dulled over time, as porn has grown ubiquitous. (At the peak of its popularity in the early 1970s, Playboy sold 7 million copies per month. The website Pornhub claims to get 100 million visitors every day.) “Maybe the damage has been done,” Kenrick said. “It used to be that looking at erotic magazines was a special event. People did it once a month if they had a subscription to Playboy.” Now that porn is all around us, he wonders if men’s attraction and devotion to their partners has not been dampened across the board.

It could also be that erotic centerfolds were never making men fall out of love with their wives and girlfriends to begin with. Maybe Kenrick’s finding from the 1980s was a false-positive result that happened to fit so well with pre-existing theory that it made its way into the canon. If that’s the case, then maybe porn was only ever bad for relationships because of how ashamed it made us feel (and not because of the “contrast effect”). Perhaps its recent spread could even end up as a boon for families. “It’s more understood that people are looking at porn. It’s joked about on sitcoms. It’s more accepted,” said Sam Perry, the University of Oklahoma sociologist. That means that whatever harms it may have done before on account of the taboo could begin to dissipate.

In any case, one thing is clear: There isn’t solid evidence that supports the claims, in the Utah resolution and elsewhere, that pornography has perpetuated a “sexually toxic environment” that degrades the family unit and leads to infidelity. Until existing research has been re-examined, extended, and updated, and the textbooks changed to match, fears of widespread harm ought to be on hold.