How Social Media Rewires the Brain

How social media affects the brain is a question with a loud popular answer and a quieter scientific one. The popular answer says apps hijack your dopamine system and permanently alter your wiring. The scientific answer is narrower: reward circuits do respond to likes and notifications, adolescent brain scans show patterns linked to habitual checking, and experiments on cutting back give mixed results. This article walks through the mechanisms, the best-documented studies, and the places where the evidence is thinner than the headlines.

This article is for educational purposes, not medical advice. If you are concerned about your relationship with social media or your mental health, speak with a qualified healthcare professional.

What "rewiring" can and cannot mean

Every repeated experience changes the brain. Learning a piano piece, a language, or a route through a new city strengthens some connections between neurons and weakens others. That capacity is called neuroplasticity, and it is the reason the brain can rewire itself throughout life, not only in childhood.

So the claim that social media "rewires" the brain is trivially true in the same sense that reading rewires it. The meaningful questions are different. How large are the changes? Do they last? Are they harmful? And did social media cause them, or did people whose brains already differed end up using it more?

Hold on to those four questions, because the research tends to answer the first one better than the others. Brain imaging can show that two groups differ. It cannot, by itself, say which came first.

A useful picture is a footpath across a lawn. Walk the same route daily and the grass wears away into a visible trail, which makes the route easier to take tomorrow. The trail is real, and it is also not a rebuilt landscape. Whether heavy use of an app wears a deep trail or a faint one is exactly what researchers are still measuring.

The slot machine in your pocket

In the 1950s, B.F. Skinner and Charles Ferster showed that behavior depends heavily on the schedule of rewards. Their work established that organisms behave differently under different schedules, and one in particular, the variable ratio schedule, stood out. A variable ratio schedule rewards a response after an unpredictable number of tries, averaging some fixed value. Behavioral summaries note that it produces both the highest rate of responding and the greatest resistance to extinction (the tendency to keep going after rewards stop).

Slot machines are the standard example. They pay out on a variable ratio schedule, which is why each pull feels like it might be the one. The uncertainty is not a flaw in the machine. It is the mechanism.

Social feeds share the structure. When you refresh, you do not know whether you will find something delightful, irritating, or nothing at all. A notification may announce a like, a reply, or a message from someone you were hoping to hear from. Rewards arrive irregularly, and checking is cheap, so the checking can repeat many times a day.

There is a caution worth stating. A feed is not literally a laboratory reinforcement schedule, and nothing here shows that every app was built to mimic one. The slot-machine comparison is an analogy that explains why the behavior can feel compulsive, not a proof of intent or a diagnosis. What the analogy does capture accurately is the logic: unpredictable payoffs plus low effort tend to produce persistent checking.

Once a checking routine runs often enough, it starts to look like what researchers call a habit, a behavior triggered by a cue rather than a decision. The psychology of habits explains how the brain moves a behavior from deliberate to automatic, and the same logic applies when a buzz in the pocket produces a reach for the phone before you have decided anything.

Dopamine: anticipation, not pleasure

Dopamine is often described as the brain's pleasure chemical, which is a misleading simplification. A foundational 1997 paper in Science by Wolfram Schultz, Peter Dayan, and Read Montague reviewed evidence that dopamine neurons in primates signal errors in predictions about rewarding events. In plain terms, these neurons respond to rewards that are better or sooner than expected and to cues that predict rewards, and they go quiet when an expected reward fails to arrive.

That matters for feeds. A like is a small reward, and whether it appears is uncertain, so the system is built around prediction and surprise. The feeling of wanting to check, the pull toward the phone, is closer to anticipation than to enjoyment. For a longer treatment of the chemistry, the dopamine system explained covers reward, motivation, and why the popular story gets it wrong.

What the reward-circuit studies show

Several imaging studies have looked at how the brain's reward regions respond to social feedback. Two stand out.

In a 2013 study in Frontiers in Human Neuroscience, Dar Meshi, Carmen Morawetz, and Hauke Heekeren scanned 31 healthy adults aged 19 to 31. Activity in the left nucleus accumbens (a reward-related structure deep in the brain) when people received positive feedback about themselves, compared with feedback about someone else, correlated with their scores on a Facebook Intensity Scale (r = 0.400, p = 0.026). Activity in response to money did not predict Facebook use. The study suggests a link between how strongly someone's reward system responds to social reputation and how much they use social media, but it was a single scan session in a small sample and cannot show that use changed the response.

In 2016, Lauren Sherman and colleagues published "The Power of the Like in Adolescence" in Psychological Science. They built a task that simulated Instagram and scanned adolescents as they viewed photos with many or few likes. Teens were more likely to like photos that already had many likes, including photos depicting risky behavior. Viewing photos with many likes was linked to greater activity in regions involved in reward processing, social cognition, imitation, and attention, while viewing risky photos reduced activity in the cognitive-control network.

Study	Design	Sample	Main finding	Key limit
Meshi et al., 2013, Frontiers in Human Neuroscience	fMRI plus Facebook use scale	31 adults, ages 19 to 31	Nucleus accumbens response to self-related reputation gains correlated with Facebook use (r = 0.400)	Small sample; single time point
Sherman et al., 2016, Psychological Science	fMRI with a simulated Instagram feed	Adolescents	Many-like photos engaged reward and social-cognition regions; risky photos reduced control-network activity	Simulated feed in a scanner, not real use
Maza et al., 2023, JAMA Pediatrics	Three-year longitudinal fMRI	169 early adolescents	Habitual checking linked to different anticipation responses that changed over time	Observational; one region of the US
Achterberg et al., 2022, Developmental Cognitive Neuroscience	Longitudinal structural MRI	Dutch adolescents, small community sample	Social media use and well-being linked to brain structure in distinct ways; the social media results did not survive correction	Small sample; structure, not function

Notice the shared limit. Each study shows that brain responses relate to social media behavior. None of them can show that the behavior reshaped the response, because people who are more sensitive to social reward may simply choose to spend more time where social rewards are on offer.

What brain scans show over time

The strongest evidence on change comes from studies that follow the same people for years. The best-known recent example is a 2023 JAMA Pediatrics paper by Michael Maza, Kathryn Fox, Seh-Joo Kwon, Jessica Flannery, Kristen Lindquist, Mitchell Prinstein, and Eva Telzer.

The researchers followed sixth and seventh graders from three public middle schools in rural North Carolina for three years, scanning them annually while they performed a task that measured brain activity as they anticipated social rewards, punishments, or neutral feedback. Of 178 participants recruited at age 12, 169 met the inclusion criteria. At the start, the habitual checkers (adolescents who reported frequent checking of Facebook, Instagram, and Snapchat) showed lower neural sensitivity to social anticipation in several regions, including the amygdala, insula, ventral striatum, and left dorsolateral prefrontal cortex. Over the following years, the habitual group showed increasing activity in those regions, while the non-habitual group showed decreasing activity.

Two points from that study are easy to miss. First, the task measured the response to social incentives in a laboratory, not a response to social media itself, so the findings concern how the brain handles social reward and punishment generally. Second, the authors concluded that checking "may be associated with" changes, and called for more research on long-term links. The study is longitudinal, which is stronger than a snapshot, but it is still observational. Adolescents were not randomly assigned to check or not check.

A study that points the other way

A 2022 paper in Developmental Cognitive Neuroscience by Michelle Achterberg, Eveline Crone, and colleagues followed Dutch adolescents with structural MRI (scans of brain anatomy rather than activity). They found that social media use and mental well-being were each linked to structural development, but in distinct ways. Adolescents who used more social media had higher baseline cortical thickness in parts of the prefrontal cortex and steeper declines in lateral prefrontal thickness, but none of those social media results survived correction for running many statistical tests. Some well-being findings did survive, and they involved surface area rather than thickness. The sample was a community one of fewer than 200 adolescents.

Why mention a study that complicates the story? Because a fair reading of this literature cannot rest on the one that fits the narrative. When one team finds shifting reward-region responses and another finds that structure and well-being relate differently from what headlines imply, the honest summary is "the picture is unsettled."

The scale of the research effort

Several large projects are trying to settle the matter by following many children over time. The Adolescent Brain Cognitive Development (ABCD) Study invited 11,880 children aged 9 to 10 at 21 research sites across the United States, and it tracks their development through adolescence into young adulthood, including how experiences such as social media and sleep relate to brain development. Studies of that size can detect smaller effects than a 31-person scan, and findings from it will matter for years.

Why teenagers get the most attention

The reason adolescents dominate this research is developmental timing. The National Institute of Mental Health notes that the brain finishes developing and maturing in the mid-to-late 20s, and that the prefrontal cortex, the region behind the forehead that supports planning, prioritizing, and decision-making, is one of the last parts to mature. The Sherman and Maza studies both sit on this point: teens respond strongly to peer approval at an age when the circuits that help weigh long-term consequences are still developing.

NIMH also notes that the teen years are a period of "fine-tuning" how the brain works, and that many teens do not get enough sleep, which can affect attention and impulse control. How sleep affects the brain shows why late-night scrolling can matter beyond the time it takes up.

The size of the effect is the argument

Here the evidence turns more cautious. In a 2019 Nature Human Behaviour paper, Amy Orben and Andrew Przybylski analyzed three large datasets with a combined sample of 355,358 participants. They found that the association between digital technology use and adolescent well-being was negative but small, explaining at most 0.4% of the variation in well-being, and they argued that effects at that scale were too small to warrant policy change.

Other researchers disagree about how to interpret small averages, noting that averages can hide larger effects in particular subgroups or for particular types of use. That disagreement is live. A reasonable reading is that population-wide effects on average well-being appear modest in these data, while individuals vary and some use patterns may matter more than others. Neither view supports the claim that social media universally damages teenage brains, and neither supports the claim that it does nothing.

Anxiety, comparison, and the fear of missing out

Apps differ from slot machines in one important way: they are social. A like carries information about standing in a group, and humans are highly sensitive to that. This is where a mental-health link is most plausible, and where the best causal evidence exists.

A 2022 American Economic Review paper by Luca Braghieri, Ro'ee Levy, and Alexey Makarin used the staggered rollout of Facebook across US colleges as a natural experiment. Because the platform arrived at different colleges at different times, the researchers could compare students at schools that had it with students at schools that did not yet. They found that rollout had a negative impact on student mental health and raised the likelihood that students reported poor mental health harming their academic performance. The authors link these effects to unfavorable social comparisons. The study concerns the early version of Facebook, which was a different product from today's feeds, so it speaks to mechanism more than to any current app.

The concept often used to describe the feeling is fear of missing out (FOMO), an unpleasant sense that others are having rewarding experiences without you. A 2024 scoping review in PLoS One by Groenestein and colleagues screened 4,121 articles and included 106. Researchers broadly agree that FOMO is driven by comparing your activities with others', but the review found the literature fragmented, with unclear conceptual boundaries and a need for causal research.

Why would this register as anxiety? One piece of the answer is that people give bad social information more weight than good. Negativity bias and memory explains how a harsh comment can outweigh several kind ones. The neuroscience of anxiety describes the threat-detection circuits involved. Combining those general findings with a feed is a hypothesis, not a demonstrated pathway. The psychology of loneliness covers a related thread, how the brain handles social disconnection.

Comparison also does not need a screen to work. Social proof psychology shows how strongly people use others' behavior as a guide, and feeds display other people's behavior continuously, which is why like counts alter choices as in the Sherman study.

Attention: the cost of being interrupted

Attention is the third area where claims run ahead of data. The popular version says phones destroy focus. The supported version is narrower and older.

In 2008, Gloria Mark of the University of California, Irvine, Daniela Gudith, and Ulrich Klocke published "The cost of interrupted work: more speed and stress" at the ACM CHI conference. The title states the core result: people who were interrupted compensated by working faster, at the price of more stress. That study concerned workplace tasks, not social media specifically, so it supports a principle rather than a number for how long a phone check costs.

What follows from it is a principle. Interruptions have costs beyond the seconds spent on them, and a notification that arrives mid-task is an interruption whatever its content. The myth of multitasking lays out why switching between tasks is not free.

Feed design adds a second attention effect. Content that never ends removes the natural stopping points that books, newspapers, and television once provided. Infinite scroll psychology looks at how that design affects the decision to continue. The behavior is related to procrastination too: why we procrastinate describes how the brain favors immediate, low-effort reward over distant goals, which is exactly the trade a scroll offers.

The ranking systems behind feeds are worth understanding on their own terms. How recommendation systems rank choices explains that these systems optimize for a measurable signal, such as engagement, and the signal shapes what appears. That is a statement about how the technology works, not evidence about its effects on any individual brain.

What happens when people cut back

If a reward-driven habit is the problem, reducing exposure should help. Experiments test this directly, and the results are more mixed than either side usually admits. None of the three studies below imaged the brain; all measured self-reported or behavioral outcomes.

Study	Intervention	Sample	Result
Allcott, Braghieri, Eichmeyer, Gentzkow, 2020, American Economic Review	Deactivated Facebook for four weeks before the 2018 US midterm elections	US Facebook users	Subjective well-being rose, political polarization fell, factual news knowledge fell; Facebook use stayed lower afterward
Brown and Kuss, 2020, Int. J. Environmental Research and Public Health	Seven days of abstaining from social media	61 participants	Higher well-being and social connectedness; lower FOMO and smartphone use; notifications flagged as a trigger
Collis and Eggers, 2022, PLoS One	Limit Facebook, Instagram, and Snapchat to 10 minutes per day	University students over three academic quarters	No significant effect on well-being or academic success; participants shifted to instant messaging

The four-week deactivation study is the most rigorous of the three, and it found benefits to well-being alongside costs to news knowledge. The seven-day abstinence study used a within-subject design (everyone served as their own comparison), which is weaker than a randomized trial with a control group. The 10-minute limit study found nothing, and its authors noted that participants did not reduce overall device time, they moved to messaging.

Taken together, these results suggest that what people do on a screen, and what they do instead, matters more than the raw presence of an app. Messaging with friends and scrolling a stranger-filled feed are not the same activity, and trials that treat them as one will average away real differences.

Does the brain recover?

This is the question most readers want answered, and the honest answer is that it has not been tested well. The studies above measured how people felt and behaved, not how their brains responded. Neuroplasticity implies that circuits can change when input changes, and that is why habits can be disrupted and rebuilt through changes in cues and routines. But "plasticity exists" is a general principle, not a measured finding that a specific reduction in social media use reverses specific brain changes.

Cutting back also carries trade-offs. In the four-week deactivation study, people knew less about the news afterward. A person who uses an app to keep up with friends may lose something real by dropping it. That is one reason the net effects in the trials above are mixed.

Common misconceptions

"Social media is as addictive as drugs." The comparison overreaches. Substance use disorders involve direct pharmacological action on brain chemistry, and social media does not work that way. Researchers do discuss problematic use patterns, and some describe them as behavioral addictions, but the field has not reached a consensus definition, and the Groenestein review shows how unsettled even a single related concept like FOMO remains.

"A dopamine detox resets your brain." Dopamine is not a substance that accumulates and needs to be flushed. It is a signaling system involved in learning, motivation, and movement. Taking a break from a habit may help people notice and change it, but the mechanism is not that a chemical has been cleared.

"Brain scans prove harm." A scan that differs between two groups proves a difference. Whether the difference is harmful, whether it was caused by use, and whether it persists are separate questions, as the Maza and Achterberg papers illustrate in different ways.

"Willpower is the problem." Willpower is not the central variable in the research. What the studies document is that cues, notifications, and design shape behavior, which is why people often try changing the environment rather than relying on self-control. The science of willpower discusses what the evidence says about self-control, and cognitive biases explained covers the shortcuts that make habitual checking feel like a choice.

Open questions

Researchers still lack answers to several basic questions, and a careful reader should hold the following as unresolved.

• Direction of causation. Do reward-sensitive adolescents seek more social media, does social media increase reward sensitivity, or both? Longitudinal studies like Maza's narrow this down but cannot randomize.
• Type of use versus amount. Passive scrolling, active posting, and private messaging may have different effects, and total time is a crude measure.
• Individual differences. Averages can mask groups who are more vulnerable. Identifying them is a major research goal.
• Durability. None of the trials above follow people long enough to show whether benefits from cutting back persist.
• Product change. A study of early Facebook, or of a 2018 deactivation, may not describe today's short-video feeds.

For readers who want practical context rather than a mechanism, digital minimalism and how to do a digital detox discuss approaches people use. Those are general discussions, not a prescription, and anyone struggling with compulsive use or low mood should talk to a clinician.

Frequently Asked Questions

Q: Does social media change the brain? A: Research suggests that how often adolescents check social media is linked to changes in how their brains respond to social rewards and punishments. A three-year study of 169 early adolescents in the US found shifting activity in reward and emotion regions. The findings are observational, so they show an association rather than proof that social media caused the changes.

Q: Why does social media feel addictive? A: Feeds deliver unpredictable rewards, such as likes, replies, and novel posts, at almost no effort. Unpredictable rewards produce persistent responding in behavioral research, and slot machines work the same way. "Feels addictive" is not the same as a clinical diagnosis, and researchers have not agreed on a formal definition for social media specifically.

Q: Does social media cause anxiety or depression? A: The evidence is mixed. A natural experiment using Facebook's college rollout found negative effects on student mental health, and a four-week deactivation experiment found higher well-being. Large surveys find a negative but small average association with adolescent well-being, explaining at most 0.4% of the variation in one analysis. Individual outcomes vary widely.

Q: Are teenagers more vulnerable than adults? A: Teens are the focus of most research because the prefrontal cortex is one of the last brain regions to mature, with development continuing into the mid-to-late 20s according to the National Institute of Mental Health. That timing makes plausible a greater sensitivity to peer approval. It does not show that every teen is harmed, and effects on average appear small.

Q: Does taking a break from social media help? A: Results differ across trials. A four-week Facebook deactivation raised subjective well-being, a seven-day break improved well-being and reduced fear of missing out in 61 participants, and a 10-minute daily cap showed no significant effect on well-being. None of these trials measured the brain, so they cannot say whether brain changes reversed.

Q: Is the slot machine comparison accurate? A: It is a useful analogy for the logic of unpredictable rewards, but it is not a claim that apps are identical to gambling machines or that designers intended gambling-like effects. Slot machines pay on a variable ratio schedule, which produces high response rates and resistance to stopping, and feeds share that structure in a looser form.

Q: Can scans show whether someone is addicted to social media? A: No. Imaging studies show group-level differences in brain responses, not individual diagnoses. A clinician diagnoses problems from behavior, distress, and impairment, not from a scan.

The Takeaway

Social media does not hijack the brain in a way that bypasses ordinary learning, and it does not leave the brain untouched. Its feeds present unpredictable social rewards at almost no cost, which fits what behavioral science has long known about persistent checking, and imaging studies show that adolescents who check habitually differ from those who do not in how their reward and emotion circuits anticipate social outcomes.

The step from that finding to "social media damages the brain" is where the evidence gets thin. Effects on average well-being look small, experiments on cutting back give mixed results, and no trial has yet shown that the brain changes reverse. The most useful stance is neither alarm nor dismissal: the reward mechanism is real, its size is debated, and how people use the apps likely matters more than whether they do.

Sources

• Meshi, Morawetz, and Heekeren, Nucleus accumbens response to gains in reputation for the self relative to gains for others predicts social media use, Frontiers in Human Neuroscience, 2013.
• Sherman et al., "The Power of the Like in Adolescence: Effects of Peer Influence on Neural and Behavioral Responses to Social Media," Psychological Science, 27(7), 2016, DOI 10.1177/0956797616645673.
• Maza et al., "Association of Habitual Checking Behaviors on Social Media With Longitudinal Functional Brain Development," JAMA Pediatrics, 177(2), 2023, DOI 10.1001/jamapediatrics.2022.4924.
• Achterberg et al., Longitudinal associations between social media use, mental well-being and structural brain development across adolescence, Developmental Cognitive Neuroscience, 2022.
• Schultz, Dayan, and Montague, "A neural substrate of prediction and reward," Science, 275(5306), 1997, DOI 10.1126/science.275.5306.1593.
• Orben and Przybylski, "The association between adolescent well-being and digital technology use," Nature Human Behaviour, 3, 2019, DOI 10.1038/s41562-018-0506-1.
• Braghieri, Levy, and Makarin, Social Media and Mental Health, American Economic Review, 112(11), 2022.
• Allcott, Braghieri, Eichmeyer, and Gentzkow, The Welfare Effects of Social Media, American Economic Review, 110(3), 2020.
• Brown and Kuss, Fear of Missing Out, Mental Wellbeing, and Social Connectedness: A Seven-Day Social Media Abstinence Trial, Int. J. Environmental Research and Public Health, 17(12), 2020, DOI 10.3390/ijerph17124566.
• Collis and Eggers, Effects of restricting social media usage on wellbeing and performance: A randomized control trial among students, PLoS One, 17(8), 2022, DOI 10.1371/journal.pone.0272416.
• Groenestein et al., The relationship between fear of missing out, digital technology use, and psychological well-being: a scoping review, PLoS One, 19(10), 2024, DOI 10.1371/journal.pone.0308643.
• Mark, Gudith, and Klocke, The cost of interrupted work: more speed and stress, Proceedings of CHI 2008, DOI 10.1145/1357054.1357072.
• National Institute of Mental Health, The Teen Brain: 7 Things to Know.
• Adolescent Brain Cognitive Development Study, About the ABCD Study.
• Wikipedia, Reinforcement, for the summary of variable ratio schedules and the Ferster and Skinner (1957) work.

---

Read next:

• Digital Minimalism: How to Reclaim Your Attention in 2026
• The Psychology of Stress: How Cortisol Reshapes the Brain
• Why Conspiracy Theories Spread Online