What Does a Screen Do?

The dangers linked to screen time for babies, kids, and teens are well-known. But is screen time really what’s causing them?

Screens with skull and crossbones images on them.
Photo illustration by Slate. Photos by Getty Images Plus.

Early reviewers praised the original iPhone, calling it a “revolutionary” device that “could change the future of computing.” Now, more than a decade later, the cultural conversation has shifted: Consumers fret that the smartphone era has changed our behavior and ruined our attention spans. One study found that just having a phone nearby distracts adults; anecdotally, people report that constant internet access has ruined their ability to read books and that they imagine their phone is ringing or vibrating even when it’s not.

This concern is magnified when applied to the youngest generations, who have grown up with screens in hand and all around them. If device use distracts adults, what could it be doing to young minds as they develop? UNICEF warns that the “negative effects of screen time for babies and toddlers range from shorter attention span to lower empathy,” and organizations like the World Health Organization and the American Academy of Pediatrics recommend that parents avoid all screen time—TV watching or device use—for infants and limit exposure to an hour a day for toddlers. As those kids grow into teens, they acquire their own devices and spend hours a day on them; in a 2017 Atlantic feature, psychologist Jean Twenge wrote that “the arrival of the smartphone has radically changed every aspect of teenagers’ lives, from the nature of their social interactions to their mental health.” (Other experts, as Lisa Guernsey wrote for Slate, are wary of Twenge’s alarmism.) Studies have linked heavy smartphone use in teens to decreased mental well-being and increased depressive symptoms; time looking at screens also takes time away from other healthful activities, like exercise or sleep, so organizations like the Australian Department of Health recommend kids between 5 and 17 limit their screen time to two hours a day.

These conclusions and recommendations come from dozens of studies probing correlations between screen time and behavior. But as any scientist will be quick to tell you, correlation is not causation, and it’s not clear that screen time actually causes, say, depression or anxiety. That’s because studies that show causal links are difficult to come by; they’re difficult to design and difficult to execute, leaving us with fewer causal conclusions and more associative studies to rely on for decision-making and policy.

We know, for instance, that smartphone use is associated with depression in teens. Smartphone use certainly could be the culprit, but it’s also possible the story is more complicated; perhaps the causal relationship works the other way around, and depression drives teenagers to spend more time on their devices. Or, perhaps other details about their life—say, their family background or level of physical activity—affect both their mental health and their screen time. In short: Human behavior is messy, and measuring that behavior is even messier.

Many of the current studies that probe associations simply ask participants to report things like their device use and their mood, an approach that is easy and relatively cheap to do because it requires minimal in-person intervention from researchers. But to probe causes, researchers must exert tighter control over the variables in their study, like how much and what kind of screen time people are engaging in. To establish a causal link, researchers would need to carefully design a study that dictates subjects’ device use while also tracking other factors like exercise and family socioeconomic status (which could then be accounted for through analysis of the data). But this sort of study would be both expensive and logistically difficult. Unlike medical researchers, who can assign different medications or treatments to patients and study the effects of each treatment, screen time researchers would have a hard time finding people willing to limit themselves to a researcher’s prescribed device use. (Ethics boards might also find such prescriptions unethical, especially for children’s research—given results from correlational studies showing links between heavy device use and negative outcomes, it would be hard to approve assigning children to a “high screen time” group.)

Even a medical trial style of screen time study might have flaws when adapted to study screen time. Human behavior is messy, and measuring that behavior is even messier; when humans know they’re being studied, that can change how they act. In medical research, it’s possible for both researchers and participants to remain in the dark about who got what treatment, a type of design researchers call a “double blind” study. But with behaviors, it’s impossible to keep things double blind; parents know what treatment—how much screen time—their kids are getting, and that alone could affect the results. In that case, just doing the study in the first place can skew researchers’ results.

Additionally, careful research takes precious time, and smartphones and tablets have only been around a decade; compared with a technology like television, which has been around for decades, there have just been fewer opportunities to run comprehensive studies on devices, let alone probe into causality. But the jury is still out even on many causal effects of TV, as well, though researchers have more confidence in links between TV watching and outcomes like sedentary behavior, given the sheer number of in-depth, longitudinal studies that suggest those connections with more confidence. It takes years for researchers to design a project, receive approval for it from their university’s ethics board, recruit participants, collect the data, write up the paper, and publish it—and by then, some fundamental aspect of the technology may have changed. Jenny Radesky, an assistant professor of pediatrics at the University of Michigan, jokes that she would never write a grant about, say, Pokémon Go: “The biggest challenge for media researchers right now is that the pace of scientific discovery is so much slower than the pace of technological evolution.”

Still, researchers are trying their best. Eric Rasmussen, an associate professor of media and communications at Texas Tech, has designed a series of studies that carefully manipulates what media children are consuming and how they are consuming it, and examines those variables’ effects on children’s social and emotional development. In a 2015 study, he and his team assigned some children to watch 10 episodes of Daniel Tiger’s Neighborhood, a PBS show inspired by Mister Rogers’ Neighborhood, whereas others watched 10 episodes of Pure Nature, a documentary show. Of the families asked to watch Daniel Tiger, some parents were assigned to watch the show with their child and talk about the episodes as much as possible, whereas others were told to watch together and not to talk about it, or not to watch the episodes with their child at all. Rasmussen and his collaborators found that kids assigned to watch Daniel Tiger with parents who talked about it with them were more likely to exhibit empathy (according to a questionnaire filled out by children’s parents) and to score higher on tests of self-efficacy and emotion recognition. By manipulating these variables, Rasmussen and his team were able to see how TV watching could result in a positive outcome for children, and that the effect appears to be driven mostly by quality time chatting with a parent. They found similar effects in a more recent study where parents and children interacted around Daniel Tiger’s app.

Radesky’s research has also carefully manipulated variables to get a better glimpse at the potential effects of screen time. Trained as a pediatrician, Radesky got interested in screen time research when she was working with children and families in the early 2010s. “I was seeing a lot of kids [who] for some reason were just harder to parent—fussy, more intense, more overwhelmed by the world,” she says. Around that same time, portable devices had become more widely available; phones and tablets could keep these fussy kids entertained at the dinner table, or as they settled down for bedtime. “I was hearing in clinic: [The screen] is the only thing that calms them down! I can’t take this away at bedtime, it’s the only thing that helps them fall asleep.’” Her hypothesis was that children’s temperament and behavior affected the amount of screen time they were exposed to, and screen time, in turn, affected their temperament and behavior; for a fussy toddler or preschooler, screen time might be a must to keep parents sane, but that screen time might also displace opportunities to learn self-regulation skills that would help them be less fussy over time.

To test this hypothesis, Radesky is collecting data from preschoolers three times over the course of six months, including their self-regulation deficits, how they use media, and family demographics. Her team can then create models of how children’s self-regulation and media use change over time to better understand how one might affect the other. “It’s not evidence of causality, but it’s stronger evidence than one-time correlation,” says Radesky.

More exhaustive study design means more time, effort, and funding costs from researchers. Rasmussen estimates that their TV study cost about $10,000, and the app study cost $35,000. (It’s worth noting that the Fred Rogers Co. funded the latter study.) And it requires time and buy-in from families, too; for Rasmussen’s studies, parents had to commit to two weeks of TV watching or app engagement, plus follow-up meetings on campus for additional testing. “If we had more money, we could do a lot more experimental research,” says Rasmussen, but with tightening budgets for basic science, correlational research is a much more feasible option.

And even well-designed lab studies have their drawbacks. It’s possible that any contrived study scenario can produce results that aren’t generalizable to real life, where parents and kids aren’t taking part in a study. (Researchers would say that such a study lacks “ecological validity.”) For example, one of Rasmussen’s studies looked at interactions between kids and parents while they watch videos in Rasmussen’s lab, which is set up like a living room with a couch, TV, and chairs. “Ideally, in order to establish ecological validity, we would go into their home and do the same thing, but when we go into their home and do that, we’re disrupting the normal flow of things,” says Rasmussen. Just being observed can affect behavior, which only makes it more difficult to design a screen time study from which researchers might draw conclusions about devices’ causal powers.

Until more large-scale projects like Rasmussen’s and Radesky’s are completed, we’re left mostly with correlational data to guide our decisions about screen time. No matter what, kids are going to spend time on their phones and tablets—and parents of the youngest users want guidance. Radesky wants to focus on solutions that can help families make the right decisions for themselves. “You can’t solve a problem just by taking the kid’s phone away,” says Radesky. “You need to solve all the underlying problems that contribute to the drama you’re having.” For instance, sometimes you just need to put an episode of Daniel Tiger on the tablet so you can have a few minutes to prep dinner, or draw the bath. Rather than feeling guilty about screen use and getting wrapped up in what Radesky calls “the competitive cult of intense parenting,” parents might benefit from taking a problem-solving approach: “So this doesn’t feel good to you. How can you change that? How can you find replacement activities? How can you ask for help?” There’s a strong consensus, she says, that certain activities—talking, playing, reading, being bored—are opportunities for parents and children to spend quality time together and for children to develop and learn. Those things can happen in the presence of screens, but we often let the screens distract us instead. It’s not so much the screen that causes problems, but the fact that it often supplants opportunities to engage in richer activities instead.

Rasmussen also says he gets questions from concerned parents asking what they should do with their kids in light of the uncertainty about screen time’s causal effects. Parents can set boundaries on use but can also teach their kids about how to set their own boundaries around screen time and what they choose to do on their devices. “A parent’s job is to give our kids tools to deal with the media content themselves,” he says. “Protecting our kids is good, but empowering them is great.”

Future Tense is a partnership of Slate, New America, and Arizona State University that examines emerging technologies, public policy, and society.