Study: Short-Form Video Isn't Rotting Your Brain
A recent meta-analysis concerning short-form video, mental health, and attention spawned a lot of tech panic. Did critics even read the study?
If you or your kids have spent too much time scrolling short-form videos during your holiday breaks, don't fret—you probably haven't suffered too much brain damage.
Wait, was that even a question? the less paranoid among you may be asking. Oh, my sweet child, I envy your ability to avoid the tech panic world. In certain circles, it's taken as a given that watching short-form videos—TikTok content, Instagram Reels, etc.—will destroy your intellect, attention span, and mental health.
"Bingeing TikTok reels may be hazardous to your well-being," organizational psychologist Adam Grant posted to X in November. In "71 studies, >98k people: The more short-form videos teens and adults watched, the more they struggled with attention, self-control, and stress and anxiety."
You are reading Sex & Tech, from Elizabeth Nolan Brown. Get more of Elizabeth's sex, tech, bodily autonomy, law, and online culture coverage.
AI investor Aadit Sheth put it even more starkly: "The most dangerous addiction today isn't a substance," he posted to X on December 4. "Research on 100,000 people confirms that heavy short-form video use is just voluntary cognitive decline. We are actively training our brains to fail at hard tasks."
And, of course, tech-panic godfather Jonathan Haidt took it even further, describing what's happening as "the global destruction of the human ability to pay attention."
Did anyone read the study?
Because delving beyond the abstract gives us a much less dire picture than all that.
14 Chinese Studies Find TikTok Addicts Are Addicted to TikTok
Titled "Feeds, feelings, and focus: A systematic review and meta-analysis examining the cognitive and mental health correlates of short-form video use," the study was published in Psychological Bulletin in September. And while the review overall included 71 previous studies, the analysis about cognitive traits, like attention—that is, the finding everyone seems to have glommed onto—only relied on 14 studies, all of which were conducted in China.
A review of 14 studies is not nothing, but it's a lot less impressive and conclusive than a review of 71 studies with nearly 100,000 participants. And the fact that all of these studies come from China at least opens the possibility that the results aren't generalizable to the U.S. or the whole world.
There's a bigger issue: Almost all of these studies were specifically concerned with people with problematic or compulsive short-form video habits.
The negative effects of short-form video on cognition were largely linked to "addictive" use and strong emotional attachment to the habit, not short-form video consumption broadly and not even to how much time people spent watching short-form video.
Overall, more short-form video "engagement" was linked to poorer cognition, with a moderate, negative mean effect size. The correlation coefficient—known in statistics as r—was -0.34.
(A mean effect size of 0.10 represents a weak correlation, the review authors explain, while 0.30 represents a moderate correlation and 0.50 a strong correlation. "A positive correlation indicates that higher [short-form video] engagement is associated with better health indicators, whereas a negative correlation indicates the reverse.")
But the effects weren't equally distributed across cognitive domains or video engagement metrics. The analysis found no link between video engagement and reasoning ability, and only a weak negative association between video engagement and language skills or memory. There were no strong associations. The only moderate associations between short-form video engagement and cognition concerned attention (-0.38) and inhibitory control (-0.41), a measure of self-control and the ability to avoid impulsive behavior.
What's more, negative cognitive effects were very weakly associated with short-form video duration—that is, with the amount of time people spent watching short-form video. Poorer cognition was only strongly associated with intensity of consumption (a measure of "emotional/psychological attachment" to short-form video apps) and moderately associated with addiction (defined as "habitual, uncontrollable use").
This seems like a really important point (and hints at which way causation may flow here). The negative cognitive traits were at least moderately linked to compulsive watching of short-form videos (with a correlation coefficient of -0.37) and strongly linked to emotional dependency (-0.55), but only very weakly linked to the amount of time spent watching videos (-0.20).
Tech critics love to ignore the fact that we don't know which way causation flows with studies like these. Their interpretations suggest that all people start at an equal level of cognitive ability and mental stability, and upon watching too many short-form videos (or spending too much time on social media, or whatever the crisis du jour is), this ability and stability starts to decline. But we all know that stability and intellect are not distributed equally. And it seems perfectly plausible that people already suffering from stress, social anxiety, poor self-control, and so on, may turn to digital distractions more frequently than do their more well-adjusted, less stressed, and less impulsive peers.
If the tech-panic interpretation were correct here, we would expect to see duration at least moderately tied to negative traits. The amount of time spent watching short-form videos would be a big deal, with a sort of dose-dependent effect—the more short-form videos you consume, the worse your cognition gets.
Or, to put it in popular parlance: The more scrolling, the more brain rot. But that's not what we see.
Rather, we see people with problematic attachments to short-form video also suffering from poorer focus and inhibitory control.
And the fact that people with poor self-control are more likely to be compulsive or habitual scrollers seems like a pretty commonsense and nonalarming finding.
But nonalarming findings don't get a lot of attention on social media and don't lend themselves to regulatory intervention. So instead, we see people distorting, obfuscating, and exaggerating this review's findings to make things appear much darker and more catastrophic.
Short-Term Video Use, Duration Not Significantly Linked to Mental Health
What about mental health—surely there must at least be some big doom and gloom finding here?
Nope.
The vast majority of the source studies included in this review—61 studies—looked at how short-form video consumption correlated with mental health. And the association was a negative one, with higher short-form video engagement linked to poorer mental health correlates. But the average effect size was quite weak, with a correlation coefficient of just -0.21.
Looking at specific mental health factors, the only moderate correlations concerned stress (-0.34) and anxiety (-0.33). (There were no strong associations.)
Short-form video engagement was just barely associated with well-being (-0.14) and weakly associated with depression and loneliness (-0.23 for each). It showed no significant association with body image or self-esteem.
And there was no significant association between the amount of time spent watching short-term videos or frequency of engagement and mental health correlates. The correlation coefficients here were -0.10 and -0.05, respectively.
The association between mental health and whether one watched short-form video at all was also trivial (-0.13), as was the tie between mental health and intensity of usage (-0.14).
The only moderate effect size was found when looking at the link between short-form video "addiction" and poor mental health (-0.32).
Again, we've got what's not exactly an earth-shattering finding: People who feel addicted to short-form videos are also more likely to be anxious and stressed.
And, again, we've got little to suggest that watching short-form videos generally makes one worse off than those who don't, or that poorer mental health is tied to duration or frequency of time spent doing so.
Big-Picture Problems
Tyler Cowen wrote earlier this month about another study on video watching. This one also raised a lot of alarm over what were, ultimately, pretty small effects: "for each daily hour of video watching, a child experiences (on average) a reduction of non-cognitive skills of 0.091 standard deviations…less than a tenth of a standard deviation" and "likely smaller than the change in your cognitive ability over the course of a day," as Cowen pointed out. He suggests that we shouldn't ignore such findings ("this is a matter of genuine concern, and I believe many parents would be wise to limit their children's video watching") but, also, we should avoid Haidt-style theatrics about "the global destruction of the human ability to pay attention."
Dramatic pronunciations about humanity's doom seem to be weirdly appealing to people here, perhaps because they play to people's sense of superiority or intrigue or nostalgia. But they're rarely, if ever, borne out by the data and effect sizes we actually see.
There are also big-picture problems with claims like Haidt's, including the fact that most people don't watch short-form videos at all. Only about a third of U.S. adults say they use TikTok, and that tells us nothing about how often they do. Even among U.S. teenagers—where use of TikTok and Instagram stands around 60 percent and use of YouTube around 90 percent—a much smaller fraction (between 12 percent and 16 percent) report truly problematic use. If these apps are out to destroy human attention spans globally, they need to up their game.
We should also be careful about generalizing the "feeds, feelings, and focus" results globally because they're not globally representative. Nearly three-quarters of the studies included in the review (74 percent) come from Asia, with just 11 percent from North America, 11 percent from Europe, 3 percent from Africa, and 1 percent from Central America.
And, of course—one more time, everybody!—correlation is not causation. Even when we find associations between tech use and negative traits, we cannot be sure tech use caused these negative traits. Maybe stress and short attention spans cause people to be more avid or enthusiastic users of social media or online video. Maybe some third factor, like life circumstances or an underlying mental condition, spurs both app addiction and stress or poor inhibitory control.
Bingeing TikTok reels may be hazardous to your well-being.
71 studies, >98k people: The more short-form videos teens and adults watched, the more they struggled with attention, self-control, and stress and anxiety.
Read a book. Watch a movie. Long live longform. pic.twitter.com/Yzyv68kBDh
— Adam Grant (@AdamMGrant) November 14, 2025
A substacker who writes under the pseudonym Owen Kellogg has been developing what he calls the Compensatory-Use Model of screen time. Data suggests that "phones do matter, but their role is often misunderstood," he writes. "Instead of operating as a primary source of distress, heavy phone use appears to function as a compensatory behavior. When young people lack reliable sources of support or connection, they turn to tools that provide stimulation or regulation. Heavy screen use fills gaps left by unmet material and psychological needs."
Of course, we have heard all sorts of "tech is rotting your brain" rhetoric before. "This is your brain on computers," The New York Times intoned in a 2010 piece on how "juggling e-mail, phone calls and other incoming information" was undermining our ability to focus, with "deadly consequences." Psychology Today warned, in 2009, that "Twitter makes you stupid." The Guardian was ahead of the game, warning way back in 2000 that "computers rot our children's brains."
And before social media and email and iMacs, there was television to blame. In 1990, Time magazine warned about "every parent's worst nightmare: the six-year-old TV addict" and how television was making children "less well informed, more restless and poorer students." And "a senate sub-committee…heard a psychiatrist testify Thursday that the stupidity of a child is in direct ratio to his stock of comic books," the Associated Press reported on April 23, 1954.
The "this time it's different" people are going to need better data to convince me.
More Sex & Tech News
"The modern Internet is cooked": Law professor Eric Goldman takes a look "at legal challenges to a dozen state censorship laws," many of which "are headed towards the Supreme Court," he writes.
Department of There Are Not Enough Eye-Rolls: New York Gov. Kathy Hochul has signed into law a measure requiring warning labels on social media.
The U.S. government is using real censorship to fight fake censorship, notes Mike Masnick. It has blocked five people from getting U.S. visas because of their efforts to "coerce American platforms to censor, demonetize, and suppress American viewpoints they oppose," as the State Department put it. "That theory relies almost entirely on fabricated or grossly misrepresented evidence," Masnick suggests.
App store age verification halted: A federal court has temporarily blocked a Texas law—which was set to take effect January 1—that required app stores to verify user ages before allowing them to download apps, and to block minors from downloading apps without parental permission.
Ditch that embarrassing old moniker: Gmail will now let users change their email addresses while still retaining all their data and services.