Cutting back on social media helps a little, but breaks under a week may not

Key takeaways

• In a 2018 experiment, Facebook users paid to deactivate for four weeks gained about an hour a day, felt slightly better and kept using Facebook less after the study ended¹.
• Ten randomized trials pooled in 2025 found that cutting back on social media modestly eased depression symptoms, while trials of quitting outright showed no clear benefit⁶.
• When 221 students turned phone alerts off for a week, they reported fewer signs of inattention and hyperactivity than in a week with alerts on⁹.
• In an Oxford analysis of surveys covering more than 355,000 people, technology use explained at most 0.4% of the differences in teens’ well-being²³.
• By mid-2026, 32 states banned phones in class statewide, a JAMA Pediatrics study found, but direct evidence that phones drive youth mental health problems is hard to find¹³.

In the fall of 2018, a team of economists made an unusual offer to Facebook users. Deactivate your account for four weeks before the midterm election, and we’ll pay you $102¹.

About 580 people were paid to deactivate, and the researchers compared them with more than 1,000 people who kept using Facebook¹. The quitters gained roughly an hour a day¹. They spent more of it with friends and family, and more watching TV alone¹.

They also knew a little less about the news and became a bit less politically polarized¹. On surveys, they reported slightly higher happiness and life satisfaction and slightly less anxiety and depression¹.

The most telling result came after the experiment ended. Nobody was paying them anymore, yet they kept using Facebook less than the comparison group and valued it less¹.

That study, led by Hunt Allcott and Matthew Gentzkow and published in the American Economic Review in 2020, captures what the research on screens now shows¹. It was funded by the Sloan and Knight foundations; Allcott was also a paid employee of Microsoft Research, and Gentzkow did paid consulting for Amazon¹. Our figures come from the 2019 working-paper version.

Effects are real but small, and they depend on which habit you change. This guide sorts the evidence and ends with a plan to test your own digital boundaries.

On average, screen time explains little about how people feel

In 2019, Oxford researchers Amy Orben and Andrew Przybylski analyzed three large surveys from the U.S. and Britain, with more than 355,000 participants in all²³. The surveys ran from 2007 to 2016³.

Technology use explained at most 0.4% of the differences in young people’s well-being². That link was negative but tiny. It was about the same as the link with regularly eating potatoes, and wearing glasses had a more negative link³.

Other factors mattered far more. Being bullied had a negative link about 4.3 times as large, and getting enough sleep and eating breakfast were much more strongly tied to feeling well³.

The team tested a huge range of reasonable ways to analyze the data, a method called specification curve analysis³. Depending on the choices, results could look positive, negative or flat³. The authors concluded the effects were too small to justify policy changes².

Not everyone agrees, and an average can hide people who are harmed. A 2026 study of school phone policies, described below, noted that research links phone use with anxiety and depression but that direct evidence of cause is hard to find¹³.

The practical lesson: “screen time” is too blunt a measure. The experiments point to specific habits, such as endless feeds, constant alerts and late-night scrolling. Our article on screens and sleep covers the bedtime piece.

Trimming social media brings a small, real lift

In 2018, psychologist Melissa Hunt and colleagues at the University of Pennsylvania ran an early experiment⁴. They randomly assigned 143 undergraduates either to keep their usual habits or to limit Facebook, Snapchat and Instagram to 10 minutes each per day for three weeks⁴.

Use was checked with phone battery screenshots, not just memory⁴. The limited group reported less depression and loneliness, and the students who started out most depressed improved most⁴.

Hunt pointed to social comparison: scrolling through other people’s highlights can make your own life look dull by contrast⁴. Even so, she didn’t recommend quitting social media entirely⁴.

Larger reviews now pool these experiments. In 2025, Burnell and colleagues combined 32 papers with 5,544 people⁵. Restricting social media improved well-being by a small amount, and the length and type of break made no consistent difference⁵. The authors called this weak support and urged research on why it helps⁵.

A second 2025 review pooled 10 randomized trials with 1,491 people⁶. In a randomized trial, people are assigned to groups by chance. Cutting back eased depression symptoms by an amount researchers call small⁶.

This is the twist. Trials that limited use showed a clear benefit, while trials that required total abstinence did not⁶. The gap between the two kinds wasn’t statistically significant, so it isn’t settled⁶. No trial ran longer than three weeks⁶.

Break length may matter, too. A 2024 analysis by Ferguson pooled 27 social media experiments and concluded the effects weren’t different from zero⁷. In 2025, a team led by Thrul reanalyzed those data using only true reduction or abstinence studies⁷.

Breaks shorter than a week went with slightly worse mental health, while breaks of a week or more went with slight improvement⁷. The team suggested at least one week, with about three weeks as a possible sweet spot, and said this needs confirming⁷.

Why would a short break backfire? One clue: a 2022 review found that craving and withdrawal symptoms rose during abstinence in all three studies that measured them⁸.

Notifications are interruptions you agreed to

In 2016, Kosta Kushlev and colleagues ran a two-week experiment with 221 students at the University of British Columbia⁹. During one week, every alert stayed on and the phone stayed close at hand. During the other, Do Not Disturb was switched on and the phone was tucked away out of sight⁹.

With alerts on, students reported more symptoms of inattention and hyperactivity, measured with questions based on ADHD criteria⁹. The effects were medium-sized⁹. Silencing phones didn’t directly raise productivity or well-being. But students who felt more scattered during the alert week also rated themselves less productive and less socially connected⁹.

The authors were careful. Students knew which week they were in, the measures were self-reports, and the study doesn’t show that phones cause ADHD⁹.

Email showed a similar pattern. In a 2015 study in Computers in Human Behavior, Kushlev and Elizabeth Dunn had 124 adults, about two-thirds of them students, check email just three times a day for one week, with alerts off and the inbox closed in between¹⁰. In the other week, they checked as often as they could¹⁰.

People reported less daily stress in the limited week, a small-to-medium effect¹⁰. The authors called the result groundwork for firmer tests, since they ran many comparisons. They also noted that some jobs might get more stressful with fewer checks¹⁰.

Gray screens and app timers: small nudges, small gains

Phones are designed in bright colors. What happens if you take the color away?

In a 2024 study run through the University of Amsterdam, 84 Android users tracked their phones for a week and then switched the screen to grayscale for a week¹¹. Daily screen time fell by about 20 minutes, from 287 minutes to 268¹¹. The number of times they unlocked their phones didn’t change, staying around 80 a day¹¹.

People reported less stress and felt more in control of their phone use¹¹. Sleep and productivity didn’t change¹¹.

The catch: users rated grayscale as annoying, and about half said they would rarely or never use it again¹¹. The study also had no separate comparison group¹¹. Even with grayscale, people averaged more than four hours a day on their phones¹¹.

Built-in app timers are popular, but we found little rigorous testing of them in our review. The 2022 review’s authors argued that a detox can target specific apps or features, not just whole devices⁸.

Digital detoxes get mixed reviews

A “digital detox” means stepping away from devices or apps on purpose. In 2022, researchers funded by the Swiss National Science Foundation reviewed 21 studies with 3,625 people, most of them students⁸. The breaks ran from 24 hours to four weeks, and most involved social networking sites, especially Facebook⁸.

The results were scattered⁸:

• Life satisfaction: 4 of 8 studies found gains, 2 found no change and 2 found declines.
• Depression: all 3 studies that checked found declines.
• Stress: 3 studies found less stress, and 2 found no change.
• Anxiety: only 1 of 4 studies found a drop.

Only 6 of the 21 studies were rated high quality⁸. The authors’ verdict: no clear answer yet⁸.

These studies mostly asked people to give up an app or device in daily life. They say little about paid, phone-free retreats.

For natural-health readers, there’s a quieter lesson. You don’t need a retreat to protect your attention. A walk without earbuds, a meal without a screen and a garden bed that needs weeding are digital boundaries, too. The research suggests the steady, everyday version may matter more than the dramatic break.

Schools and states are drawing lines

In July 2026, researchers led by Y. Tony Yang of George Washington University reported in JAMA Pediatrics on cellphone rules in all 50 states and the District of Columbia¹³. They found 42 states with statewide laws or policies limiting phones in K-12 public schools¹³.

Of those, 32 banned phones statewide during class time, and 15 of the 32 banned them for the whole school day, often called “bell to bell”¹³. Florida passed the first statewide ban in 2023¹³. Utah’s bell-to-bell ban took effect July 1, 2026¹³.

The JAMA Pediatrics team found broad agreement that phones disrupt learning¹³. But it noted that direct evidence tying phone use to the decline in youth mental health is hard to find¹³. School bans are a reasonable experiment, not a proven treatment.

Warning labels are another front. In 2024, then-Surgeon General Vivek Murthy called for warning labels on social media¹⁵. In 2025, Minnesota passed a law requiring platforms to show a mental health warning and contacts for the 988 Lifeline, set to begin July 1, 2026¹⁵.

NetChoice, a tech industry group, sued to block it¹⁴. On July 1, 2026, Minnesota’s attorney general agreed not to enforce the law while a court considers the state’s motion to dismiss the case¹⁴. The pause was the state’s choice, not a court order, and we could not confirm any later ruling.

For families, our article on busy parents explains the American Academy of Pediatrics’ Family Media Plan, and our screens-and-sleep article covers the Surgeon General’s 2023 bedtime advice.

Your digital boundary plan

This plan is our own, built from the studies above, and it hasn’t been tested as a package. Its four parts take about 20 minutes to set up.

Part 1: Run a notification audit. Open your phone’s notification settings and sort every app into one of three groups.

Group	What goes here	What to do
Keep	Calls and texts from people close to you, calendar reminders, safety alerts	Leave on
Batch	Email, work chat, school apps	Turn off alerts; check at set times, such as three times a day10
Silence	Social media likes, shopping, games, most news apps	Turn off alerts entirely9

Part 2: Pick phone-free zones and times. Start with two:

• the table during meals;
• the bedroom at night, with the phone charging in another room;
• the first 30 minutes after you wake, before you check messages.

These are our suggestions. Choose the ones that fit your home.

Part 3: Set limits on one or two apps. Choose the apps that most often leave you feeling worse. Try a daily cap, such as the 10 minutes per app used in the Penn study⁴. Grayscale is an option if color pulls you in¹¹.

Part 4: Run a seven-day experiment. A week is the shortest break that showed benefit in the reanalysis described above⁷.

When	What to do
Day 0	Write down yesterday’s screen time. Rate your mood, focus, sleep and sense of connection from 1 to 5.
Days 1 to 7	Follow Parts 1 to 3. Each evening, rate the same four things. Note any urge to check.
Day 8	Compare your ratings. Keep what helped, drop what didn’t and decide whether to run a second week.

Expect the first few days to feel restless. Studies that measured cravings found they rose during breaks, one more reason to stick with the full week⁸.

What the evidence shows

Claim	What studies found	Evidence level
Total screen time is a major driver of young people’s well-being	At most 0.4% of differences explained in 355,000+ people23	Evidence says no (for screen time as a major driver)
Limiting social media for 1 to 3 weeks eases depression symptoms	Small effect in 10 randomized trials6	Promising but early
Restricting social media improves overall well-being	Small effect across 32 papers5	Promising but early
Breaks shorter than a week help	Slightly worse outcomes in one reanalysis7	Not enough evidence
Silencing notifications improves attention	Medium effect in one two-week crossover study9	Promising but early
Checking email in batches lowers stress	One two-week experiment10	Promising but early
Grayscale cuts screen time	About 20 minutes a day in one uncontrolled study11	Promising but early
Digital detox raises life satisfaction	Mixed results; few high-quality studies8	Not enough evidence
School phone bans improve youth mental health	Direct evidence hard to find13	Not enough evidence

When it’s more than a phone habit

Sometimes heavy scrolling is a way of coping with low mood, loneliness or stress. Cutting back can help, but it isn’t a treatment for depression or anxiety.

The National Institute of Mental Health (NIMH) gives clear guidance on when to get help¹²:

• If mild symptoms don’t improve with self-care, or they get worse, talk with a health care provider¹².
• If you’ve had severe symptoms for two weeks or more, seek professional help¹². These include feeling sad, tearful or hopeless; trouble finishing tasks; changes in sleep or appetite; losing interest in things; and trouble concentrating¹².
• If you’re having thoughts of suicide or urges to hurt yourself, get help right away¹².

Call or text the 988 Suicide & Crisis Lifeline at 988, or chat at 988lifeline.org¹². In a life-threatening emergency, call 911¹².

Questions readers ask

The bottom line

The average link between screens and well-being is small, and giving up technology isn’t the goal. The research favors targeted boundaries: fewer alerts, less time in the apps that leave you feeling worse and breaks that last at least a week.

The gains in trials were modest, so set your expectations accordingly. Run your own seven-day test, keep what helps and get professional support if low mood lasts.

Related reading

• Blue light gets the blame, but time on screens may cost more sleep
• Too much health advice makes people tune out, so build an information diet
• Health tracking has a helpful dose, and past it, checking feeds worry
• Busy parents need a few anchors more than a perfect routine
• Wellness apps help a little if you keep using them, and most people don’t

Sources

1. Allcott H, Braghieri L, Eichmeyer S, Gentzkow M. “The Welfare Effects of Social Media.” American Economic Review, 2020;110(3):629–676 (figures from NBER Working Paper 25514, revised November 2019). nber.org
2. Orben A, Przybylski AK. “The association between adolescent well-being and digital technology use.” Nature Human Behaviour, 2019;3:173–182 (Oxford Research Archive record). ora.ox.ac.uk
3. University of Oxford. “Technology use explains at most 0.4% of adolescent wellbeing, new study finds.” News release, January 14, 2019. ox.ac.uk
4. Berger MW. “Social media use increases depression and loneliness” (on Hunt MG, Marx R, Lipson C, Young J, Journal of Social and Clinical Psychology, 2018). University of Pennsylvania, November 9, 2018. penntoday.upenn.edu
5. Burnell K, Meter D, Andrade F, Slocum A, George M. “The effects of social media restriction: Meta-analytic evidence from randomized controlled trials.” SSM – Mental Health, 2025;7:100459 (RTI International record). rti.org
6. May W, Malouff JM, Meynadier J. “Reducing social media use decreases depression symptoms: A meta-analysis of randomised controlled trials.” European Journal of Investigation in Health, Psychology and Education, 2025;15(11):222. pubmed.ncbi.nlm.nih.gov
7. Thrul J, Devkota J, AlJuboori D, Regan T, Alomairah S, Vidal C. “Social media reduction or abstinence interventions are providing mental health benefits—Reanalysis of a published meta-analysis.” Psychology of Popular Media, 2025;14(2):207–209. scholar.usuhs.edu
8. Radtke T, Apel T, Schenkel K, Keller J, von Lindern E. “Digital detox: An effective solution in the smartphone era? A systematic literature review.” Mobile Media & Communication, 2022;10(2):190–215. refubium.fu-berlin.de
9. Kushlev K, Proulx J, Dunn EW. “‘Silence Your Phones’: Smartphone Notifications Increase Inattention and Hyperactivity Symptoms.” Proceedings of CHI 2016, 1011–1020. interruptions.net
10. Kushlev K, Dunn EW. “Checking email less frequently reduces stress.” Computers in Human Behavior, 2015;43:220–228. interruptions.net
11. Dekker CA, Baumgartner SE. “Is life brighter when your phone is not? The efficacy of a grayscale smartphone intervention addressing digital well-being.” Mobile Media & Communication, 2024;12(3). doi.org
12. National Institute of Mental Health. “My Mental Health: Do I Need Help?” Revised 2025. nimh.nih.gov
13. Mailloux M. “How many states limit cellphone use in schools?” (on a JAMA Pediatrics study led by Yang YT, July 2026). Deseret News, July 27, 2026. deseret.com
14. Gorham Q. “Minnesota will not yet enforce new law requiring mental health warnings on social media amid lawsuit.” KEYC News (Gray Television), July 1, 2026. keyc.com
15. Ferguson D (Minnesota Public Radio). “Mental Health Warnings on Social Media? Minnesota Will Require Them Next Year.” KQED MindShift, July 28, 2025. kqed.org