A seventeen-year-old sat in my office last month, referred by her school counselor for an attention evaluation. She was bright, articulate, and visibly frustrated. “I know I’m smart,” she told me. “But I can’t focus on anything for more than like a minute. I’ll start reading a chapter for class and then I’m on my phone and twenty minutes are gone and I don’t even remember picking it up.”

Her cognitive testing told a striking story. Her verbal comprehension was in the 88th percentile. Her perceptual reasoning was in the 82nd percentile. Her processing speed was above average. But her working memory, the cognitive system that holds information in mind while you manipulate it, the system that sustains focus and resists distraction, was in the 34th percentile. The gap between her reasoning ability and her working memory was statistically rare, occurring in fewer than five percent of the population.

When I asked about her daily habits, the picture clarified. She spent roughly six hours per day on her phone outside of school, primarily consuming short-form video on TikTok and Instagram Reels. She described a pattern I now see in the majority of adolescents I assess: a near-constant cycle of picking up her phone, scrolling through fifteen to sixty seconds of content, switching to another app, returning to the feed, and repeating. She rarely read anything longer than a caption. She had not finished a book in over two years.

She was not suffering from ADHD. She was suffering from what the internet has named “brain rot,” and what the science is beginning to document as a measurable shift in how digital media environments reshape cognitive habits.

Oxford University Press named “brain rot” its 2024 Word of the Year after more than 37,000 public votes. Usage of the term surged 230 percent in a single year. But beneath the meme lies a genuine scientific question: is excessive consumption of low-quality digital content actually degrading human cognitive ability? And if so, how much damage are we talking about, and can it be reversed?

I have spent the past year reviewing every major study on this question. The answer is more nuanced than either the panic or the dismissal would suggest.

The Term Is 170 Years Old. The Problem Is Brand New.

Henry David Thoreau coined “brain rot” in 1854 in Walden, writing about society’s preference for simple ideas over complex thought. The complaint maps remarkably well onto algorithmic content feeds 170 years later.

What makes the modern resurgence fascinating is its self-awareness. As Casper Grathwohl, President of Oxford Languages, observed: Gen Z and Gen Alpha, the communities largely responsible for the use and creation of the digital content the term describes, have amplified the expression through the very platforms said to cause it. The term beat out finalists including “demure,” “dynamic pricing,” and “slop,” which refers to AI-generated low-quality content.

Yet Oxford’s own Andrew Przybylski, Professor of Human Behaviour and Technology, pushed back on the announcement day with a statement that captures the central tension in this field: “There’s no evidence of brain rot actually being a thing. Instead it describes our dissatisfaction with the online world.”

Is he right? Let us look at what the research actually shows.

The Largest Study to Date: 98,299 Participants

The most comprehensive quantitative assessment of short-form video’s cognitive effects was published in 2025 in Psychological Bulletin, the American Psychological Association’s premier review journal. Nguyen and colleagues conducted a preregistered meta-analysis of 71 studies encompassing 98,299 participants examining the relationship between short-form video use and cognitive outcomes.

The results were consistent and concerning. The overall correlation between short-form video consumption and cognitive performance was r = -0.34, a moderate negative effect. Attention showed a correlation of r = -0.38. Inhibitory control, the ability to resist impulses and suppress automatic responses, showed the strongest effect at r = -0.41.

To translate these numbers into something clinically meaningful: a correlation of -0.34 means that short-form video use explains roughly 12 percent of the variation in cognitive performance. That is not catastrophic, but it is not trivial either. For comparison, the correlation between smoking and lung cancer is approximately r = 0.30. No one would call that negligible.

Critically, age did not moderate these relationships. Adolescents and adults showed comparable negative associations. This is not a “kids these days” problem. It is a human brain problem.

A second major meta-analysis published in 2026 in the Journal of Medical Internet Research, covering 58 studies and 96,676 participants, added an essential nuance. Problematic or compulsive use showed strong negative associations with stress (r = 0.41) and loneliness (r = 0.33). But routine, non-compulsive use showed no significant associations with negative outcomes. The distinction between habitual and compulsive engagement may be the single most important finding for understanding this phenomenon.

Your Brain on TikTok: What Neuroimaging Shows

The behavioral findings are reinforced by neuroimaging research that reveals measurable brain changes in heavy short-form video users.

Studies using functional MRI have found that college students with high short-video addiction scores show increased gray matter volume in reward-processing regions and heightened activity in the dorsolateral prefrontal cortex. This second finding is particularly telling: their brains are working harder to maintain cognitive control, suggesting the prefrontal systems that support attention and executive function are under strain.

EEG research has demonstrated that heavy short-form video users show significantly lower midfrontal theta power, a neurophysiological marker of executive function, even when their behavioral performance on tests appears intact. In other words, the brain changes precede observable cognitive decline. The neural signature of impaired executive control is present before the person or their teachers notice anything wrong.

The first peer-reviewed review explicitly addressing “brain rot” as a clinical phenomenon, published in Brain Sciences in March 2025, examined 35 studies and concluded that the phenomenon involves emotional desensitization, cognitive overload, and impaired executive functioning driven by dopamine-mediated feedback loops. The algorithmic feed is not merely wasting time. It is training the brain’s reward system to expect constant novelty and resist sustained engagement with any single stimulus.

Paper Beats Screens. This Is Not Nostalgia.

One of the most robust findings in this literature has nothing to do with social media specifically. It concerns the basic act of reading.

Delgado and colleagues’ 2018 meta-analysis in Educational Research Review synthesized 54 studies with more than 170,000 participants and found that paper reading produces significantly better comprehension than screen reading, with an effect size of Hedges’ g = -0.21. The paper advantage was larger under time pressure and, counterintuitively, it increased over the years studied (2000 to 2017). Digital natives read worse on screens relative to paper than older cohorts did.

Clinton’s 2019 independent meta-analysis of 33 studies confirmed the finding with an even larger effect (g = -0.25) and demonstrated something I find clinically important: readers systematically overestimate their comprehension when reading on screens. They think they understand the material better than they actually do, which means they are less likely to recognize when they need to re-read or study more carefully.

The mechanisms appear to involve depth of processing. Screen reading encourages skimming and rapid scanning. Paper reading, with its tactile and spatial properties, supports slower, more deliberate engagement. The physical layout of a book, the sense of where you are within it, the act of turning pages, all contribute to what researchers call the “shallowing hypothesis”: screens train shallow processing habits that transfer to non-screen contexts.

The PISA Data: 700,000 Students, 81 Countries

The largest dataset linking screen time to academic performance comes from the OECD’s PISA 2022 assessment of approximately 700,000 fifteen-year-olds across 81 countries.

The findings reveal a striking dose-response curve. Students spending up to one hour daily on digital devices for learning scored 14 points higher in mathematics than non-users, suggesting moderate educational technology use provides modest benefits. But leisure screen time told a very different story: students using screens for one hour or less daily for leisure scored 49 points higher in math than those using screens five to seven hours daily, a gap equivalent to roughly 2.5 years of schooling, even after adjusting for socioeconomic profiles.

One in three OECD students reported being distracted by peers’ digital devices in most or all math classes. The distraction effect alone was estimated to be equivalent to three-quarters of a year of education.

These are not small numbers. They represent real, measurable differences in cognitive outcomes at a scale that encompasses hundreds of thousands of students across dozens of countries. And they align precisely with the pattern I see in my clinical practice: moderate, purposeful technology use is neutral or mildly beneficial, while excessive passive consumption is consistently associated with poorer cognitive performance.

The IQ Connection: Are Screens Actually Lowering Intelligence?

This is the question everyone wants answered, and honesty requires acknowledging that the evidence is more complicated than either side claims.

The Reverse Flynn Effect, the documented reversal of a century-long rise in IQ scores, is real and well-established. As I discussed in detail in my previous article on Gen Z IQ trends, Bratsberg and Rogeberg’s 2018 study of approximately 730,000 Norwegian conscripts found IQ scores declining at roughly 3.3 points per decade for cohorts born after the mid-1970s, and Dworak’s 2023 analysis of 394,378 American adults found declines in verbal reasoning, matrix reasoning, and letter-number series performance between 2006 and 2018.

The temporal correlation with digital technology adoption is suggestive. But no study has established a direct causal link between screen time and population-level IQ decline. The Reverse Flynn Effect began in Scandinavian countries in the 1990s, before smartphones existed and before social media was a factor. Multiple environmental changes occurred simultaneously: education policy shifts, dietary changes, reduced leisure reading, environmental pollutants.

The most counterintuitive finding comes from the ABCD study, the largest longitudinal study of adolescent brain development in the United States. When researchers controlled for genetic cognitive predisposition and socioeconomic status, they found gaming was associated with a 2.55 IQ point gain and video watching with a 1.8 point gain over two years. Social media showed no significant effect on IQ.

However, the same study found concerning structural brain changes: children spending seven or more hours daily on screens showed premature thinning of the cerebral cortex, the region responsible for higher-order thinking, which correlated with lower crystallized intelligence scores. A 2025 analysis of over 10,000 ABCD participants found screen time associated with increased ADHD symptoms and reduced cortical thickness in frontal regions.

The honest clinical summary: excessive screen time is almost certainly one contributing factor to cognitive changes, but it is not the sole cause, and the relationship is more about what screens replace (deep reading, sustained attention, effortful learning) than what screens contain.

The Evidence That Complicates Everything

Before concluding that screens are destroying human intelligence, you need to reckon with a body of evidence that points in the opposite direction.

The most important counterweight comes from Benge and Scullin’s 2025 meta-analysis in Nature Human Behaviour, examining 136 papers covering 411,430 adults aged fifty and older. They found that digital technology use was associated with 58 percent lower odds of developing cognitive impairment. Not one of the 136 studies reported an increased risk. The authors proposed a “technological reserve” hypothesis, analogous to cognitive reserve from education and social engagement.

Andrew Przybylski’s research at the Oxford Internet Institute offers perhaps the most methodologically rigorous challenge to screen-time alarmism. His 2017 study of 120,115 English adolescents found digital technology explained at most 0.4 percent of adolescent well-being variance. To put this in perspective, wearing glasses had a more negative statistical association. His 2019 specification curve analysis covering 355,358 participants concluded that “the outsized weight given to digital screen-time in scientific and public discourse might not be merited.”

Daphne Bavelier’s research on action video games demonstrates that some screen-based activities actively enhance cognition. Her meta-analyses find consistent improvements in perception and top-down attention from action game play, with effects that transfer to non-game contexts. The brain can be trained upward through screens, not only degraded.

The emerging consensus, supported by research across age groups, is that the active-versus-passive distinction matters enormously. Passive consumption of algorithmically curated short-form content is consistently associated with poorer cognitive outcomes. Active, purposeful engagement, including gaming, educational technology use, and complex information seeking, shows neutral or positive associations. The question is not “how much screen time” but “what kind of screen time.”

Your Phone Is Draining Your Brain (But Not How You Think)

The most discussed smartphone study is Ward and colleagues’ 2017 “Brain Drain” experiment, which reported that the mere presence of a smartphone on a desk reduced working memory and fluid intelligence in approximately 800 participants, even with phones silenced and face down.

The reality is more complicated than the headlines suggested. A preregistered direct replication failed to reproduce the effect. Subsequent meta-analyses reached divergent conclusions, with the most comprehensive analysis of 166 effect sizes across 33 studies finding an effect of essentially zero (d = -0.02).

Where the evidence is more robust is in notification effects. A study by Stothart and colleagues found that merely receiving a phone notification, without answering it, made participants more than three times more likely to make errors on a sustained attention task. The effect sizes were substantial: d = 0.54 for call notifications and d = 0.72 for text notifications. These effects were comparable to actually using the phone. The mechanism is mind-wandering: notifications trigger task-irrelevant thoughts that persist long after the brief auditory cue.

Gloria Mark’s two decades of workplace research documents the broader fragmentation pattern. In 2004, average sustained attention on any screen task lasted 2.5 minutes. By her most recent measurements, it had dropped to 47 seconds, with self-interruptions accounting for roughly 49 percent of all attention switches. After an interruption, returning to a task takes an average of 25 minutes.

The practical implication is clear: the most evidence-supported intervention is not reducing total screen time but reducing notifications and environmental cues that trigger attention switching. Silencing your phone, removing it from your workspace, and batching your checking into deliberate intervals may be more effective than arbitrary screen time limits.

Can Brain Rot Be Reversed?

This is the question that matters most to the teenagers I assess, the parents who bring them, and anyone who recognizes their own habits in the research I have described.

The answer, supported by converging evidence from multiple studies, is yes.

A 2025 study in JAMA Network Open involving 373 young adults found that a one-week social media detox produced measurable improvements: anxiety reduced by 16.1 percent, depression reduced by 24.8 percent, and insomnia reduced by 14.5 percent. A separate study of 150 participants found that one week of digital abstention improved attention network test performance, with benefits persisting at one-month follow-up.

Mindfulness training shows particularly promising results for restoring the attention systems that excessive screen use degrades. A 2024 meta-analysis of 111 randomized controlled trials found that mindfulness-based interventions improved sustained attention accuracy, executive attention, working memory, and verbal fluency. Even brief interventions work: a 2024 study demonstrated that just four weeks of meditation training improved sustained attention, with EEG data confirming neural-level changes.

Nature exposure provides a complementary recovery pathway. Attention Restoration Theory, supported by experimental evidence, posits that natural environments engage “involuntary attention” through gentle stimulation, allowing the voluntary directed attention fatigued by screens to recover. Research suggests 20-minute daily nature walks without a phone are associated with measurable increases in BDNF, the neurotrophic factor that promotes neuronal growth and survival.

The emerging timeline for cognitive recovery, drawn from multiple small studies: initial withdrawal effects in the first one to three days, measurable mental health improvements within one week, significant cognitive improvements within two to four weeks, and consolidation of new attentional habits over three or more months. These timelines remain approximate, but the directionality of the evidence is clear. The brain can recover because the same neuroplasticity that allows screens to reshape attention also allows deliberate practice to reshape it back.

What I Tell My Patients

Based on everything the research shows, here is the framework I now use with every patient, parent, and attorney who asks me about screens and cognition.

First, the type of engagement matters far more than total screen time. Compulsive, passive consumption of algorithmic short-form content shows consistent moderate negative associations with attention and cognition. Active, purposeful technology use, including certain video games, can enhance cognitive function. I assess how patients use technology, not merely how much.

Second, the most robustly supported cognitive threat from smartphones is not the content but the interruption architecture. Notification effects are larger and more consistently replicated than the contested “brain drain” effect. Simple environmental modifications, silencing notifications, physically separating from devices during focused work, have stronger empirical support than blanket screen time reduction.

Third, deep reading on paper, sustained attention to single tasks, and handwriting remain the most evidence-supported cognitive practices for building and maintaining the skills that IQ tests measure. These are not nostalgic preferences. They are practices supported by meta-analyses covering hundreds of thousands of participants.

Fourth, the damage is not permanent. Every study that has examined cognitive recovery after reducing screen-based passive consumption has found measurable improvement, typically within two to four weeks. The brain is not rotting. It is adapting to its environment. Change the environment, and the adaptation changes too.

And fifth, understanding your own cognitive profile is the foundation of any meaningful intervention. The seventeen-year-old I described at the beginning of this article did not know her working memory was a relative weakness until we tested it. Once she understood the specific cognitive system that excessive scrolling was degrading, she had a target for deliberate improvement rather than vague anxiety about “brain rot.”

If you are curious about your own cognitive strengths and weaknesses, if you want to know whether your working memory, processing speed, verbal reasoning, and pattern recognition are where you want them to be, a well-designed IQ assessment with a detailed score report gives you exactly that baseline. Not a verdict, but a map: here is where you are, here is what it means, and here is where deliberate practice would make the biggest difference.

Final Thoughts

The term “brain rot” captures a genuine cultural anxiety, but the science demands precision where the meme encourages catastrophizing.

The brain is not rotting. It is adapting, sometimes maladaptively, to an informational environment that rewards rapid switching over sustained depth. That adaptation is measurable, concerning in its extremes, and reversible with intentional intervention.

The research is clear on one point above all others: the distinction that matters is not between “screen time” and “no screen time.” It is between passive consumption, which trains your brain to expect constant novelty and resist sustained engagement, and active cognitive effort, which builds the working memory, attention, and reasoning skills that predict success in every domain of life.

You get to choose which of those paths your brain walks. But making that choice wisely requires understanding where your cognitive abilities stand right now, and what the evidence says about how to protect and strengthen them.

Your brain is remarkably adaptable. The question is what you are asking it to adapt to.