The evidence
What the research actually says
Peer-reviewed studies only, with the design, the sample and the caveats next to each one.
How to read this page
Most of this is correlational: people who do more of one thing report more of another, which isn't the same as cause. Studies that randomly assigned people get an Experiment stamp. Effect sizes are noted where they're small, which is often.
Screens before bed, and sleep
The most consistent finding here. Screens in the hour before bed go with later bedtimes and shorter sleep in nearly every study, most of it correlational. One controlled lab experiment shows the mechanism: evening screen light suppresses melatonin and pushes the body clock later.
- Review 2016
Association Between Portable Screen-Based Media Device Access or Use and Sleep Outcomes: A Systematic Review and Meta-analysis
Systematic review and meta-analysis of cross-sectional studies · 20 studies, 125,198 children and adolescents, mean age 14.5
Children who used a phone or tablet at bedtime had roughly twice the odds of inadequate sleep (OR 2.17) and of daytime sleepiness (OR 2.72). Simply having a device in the bedroom, even without reported use, was also associated with worse sleep.
but: All included studies were cross-sectional, so the review cannot establish that devices cause poor sleep. Sleep and device use were mostly self reported.
- Experiment 2015
Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness
Lab experiment (randomized crossover, inpatient) · 12 healthy young adults, 2 weeks in a sleep lab
Reading on a light-emitting tablet for four hours before bed, compared with a printed book, delayed sleep onset by about 10 minutes, suppressed melatonin, shifted the circadian clock later, reduced REM sleep and left people less alert the next morning.
but: Very small sample and an unusually long, bright exposure. It shows a mechanism under controlled conditions rather than the size of the effect in everyday life.
- Review 2015
Screen time and sleep among school-aged children and adolescents: A systematic literature review
Systematic review · 67 studies published 1999 to early 2014
In 90 percent of the studies reviewed, more screen time was associated with worse sleep, mainly shorter sleep and later bedtimes. The pattern held across televisions, computers, video games and mobile devices.
but: Most underlying studies were cross-sectional. The authors call for experimental and longitudinal work before drawing causal conclusions.
- Survey 2015
Sleep and use of electronic devices in adolescence: results from a large population-based study
Cross-sectional population survey · 9,846 Norwegian adolescents aged 16 to 19
Using any electronic device in the hour before bed was associated with taking longer to fall asleep and sleeping less, and the more daytime screen time teenagers reported, the higher their odds of short sleep, in a dose-response pattern.
but: Cross-sectional and self reported, so it cannot separate cause from effect. Teenagers who sleep badly may reach for devices more.
Notifications, interruptions, and the morning check
A notification costs attention even when you ignore it, and interruptions raise stress even when the work gets done. Nobody has studied the first minutes after waking specifically; what exists is evidence about alerts and interruptions in general.
- Experiment 2016
"Silence Your Phones": Smartphone Notifications Increase Inattention and Hyperactivity Symptoms
Within-subjects field experiment (two one-week periods, counterbalanced) · 221 adults, mostly university students
During a week with notifications turned on and the phone kept in reach, people reported more symptoms of inattention and hyperactivity than during a week with alerts off and the phone put away. More notifications also went with lower self-reported productivity and well-being.
but: Outcomes were self reported over one week each. Peer-reviewed conference paper rather than a journal article.
- Experiment 2015
The attentional cost of receiving a cell phone notification
Lab experiment · Undergraduates performing a sustained attention task
Participants who received a call or text notification during an attention task, without answering it, made more errors than a control group. The authors report the size of the disruption as comparable to earlier estimates for actively texting or talking.
but: Short lab task with students. It measures the cost of a single unanswered alert, not the effect of checking a phone on waking, which has not been studied directly.
- Experiment 2008
The cost of interrupted work: more speed and stress
Lab experiment · 48 adults, mostly German university students
People who were interrupted while writing emails finished the task faster and with no more errors, but reported higher stress, frustration, time pressure and effort than people who worked without interruptions.
but: Small sample and a simulated office task, not phone use specifically. The often-quoted claim that it takes 23 minutes to refocus after an interruption comes from a separate observational study by the same lead author, not from this paper.
Does a phone on the desk make you dumber?
A well-known 2017 study found a phone on the desk lowered working memory scores. A preregistered replication found no effect, and the pooled estimate across studies is small. Listed because the claim is everywhere.
- Replication 2022
Reexamining the "brain drain" effect: A replication of Ward et al. (2017)
Preregistered direct replication (lab experiment) · 383 students
Using the same tasks and phone-location conditions as the original Brain Drain study, this replication found no difference in working memory or attention performance whether the phone was on the desk, in a bag or in another room.
but: A single replication does not settle the question, but it shows the original effect is not robust and should not be presented as established.
- Experiment 2017
Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity
Two lab experiments · 520 and 275 undergraduates
Students whose silenced phone sat on the desk scored lower on working memory and fluid intelligence tasks than students whose phone was in another room, even though nobody used the phone. The effect was strongest among people who said they depended on their phone most.
but: A preregistered direct replication with 383 students (Ruiz Pardo and Minda 2022) found no effect of phone location. A 2023 meta-analysis found a small pooled effect (g about -0.14), mainly on memory tasks. Treat the size of this effect as uncertain.
Screens and early childhood
More screen time in young children goes with slightly weaker language and developmental scores, and heavy use at age one predicts delays at two. What's on the screen, and whether a parent watches too, matter as much as the hours. Parents' own phone use counts as well.
- Cohort 2023
Screen Time at Age 1 Year and Communication and Problem-Solving Developmental Delay at 2 and 4 Years
Prospective birth cohort · 7,097 mother-child pairs in Japan
One year olds with four or more hours of daily screen time had about 4.8 times the odds of communication delay and 2.7 times the odds of problem-solving delay at age 2 compared with under one hour, with a dose-response pattern. Some associations, especially for communication and problem solving, were still present at age 4.
but: Observational and parent reported. Only about 4 percent of children were in the heaviest group, and gross motor, fine motor and personal-social domains showed no consistent association.
- Review 2020
Associations Between Screen Use and Child Language Skills: A Systematic Review and Meta-analysis
Systematic review and meta-analysis · 42 studies, 18,905 children under 12
More time on screens was associated with slightly weaker language skills (r = -0.14), and background television with weaker skills as well (r = -0.19). Educational content and watching together with a parent were associated with better language (r = 0.13 and 0.16), and starting screen use later was also associated with better language.
but: Mostly correlational studies, and the associations are small. What is watched and whether a parent is involved appear to matter as much as the amount.
- Cohort 2019
Association Between Screen Time and Children's Performance on a Developmental Screening Test
Longitudinal cohort (random-intercepts cross-lagged panel) · 2,441 mother-child pairs in Calgary, assessed at 24, 36 and 60 months
More screen time at age 2 predicted slightly lower developmental screening scores at age 3, and more screen time at 3 predicted lower scores at 5. The reverse direction (poorer development leading to more screen time) was not supported.
but: The effects were small (standardized coefficients of about -0.06 to -0.08) and both screen time and development were reported by parents. Observational, so unmeasured family factors could contribute.
- Cohort 2018
Technoference: longitudinal associations between parent technology use, parenting stress, and child behavior problems
Longitudinal (four time points over six months, cross-lagged models) · 183 couples with a child aged 0 to 5
Parents who reported that devices more often interrupted time with their child tended to report more child acting-out behavior at later time points, and children with more difficult behavior led to more parental stress and more device interference in turn. The relationship ran in both directions.
but: Small, mostly white and well-educated US sample, all measures parent reported. The effects were modest and the design cannot rule out other family factors.
- Guidance 2016
Media and Young Minds
Policy statement
The AAP recommends no screen media other than video chat for children under 18 months, choosing high-quality programming and watching together for 18 to 24 months, and limiting screen use to about one hour a day of high-quality content for ages 2 to 5, while keeping screens out of bedrooms and mealtimes.
but: Guidance from an expert body, not a primary study. The statement itself notes that research on interactive media in young children is still limited.
What happens when people cut back
The encouraging part. Blocking mobile internet for two weeks improved attention, mood and well-being in a preregistered trial. Cutting use by an hour a day held up better over four months than quitting outright. A review of "digital detox" studies is more cautious.
- Experiment 2025
Blocking mobile internet on smartphones improves sustained attention, mental health, and subjective well-being
Preregistered randomized controlled trial (delayed-intervention control) · 467 iPhone users in the US and Canada, two-week block
People assigned to block all mobile internet on their phone for two weeks, while keeping calls and texts, showed improvements in mental health (about 0.56 SD), subjective well-being (about 0.45 SD) and objectively measured sustained attention (about 0.23 SD) relative to the waiting group. Part of the benefit was explained by more time socializing in person, exercising and being outdoors.
but: Only about a quarter of participants kept the block active for the required 10 of 14 days, so the intention-to-treat effects mix compliers and non-compliers. Participants were paid volunteers willing to try going offline.
- Experiment 2023
Finding the "sweet spot" of smartphone use: Reduction or abstinence to increase well-being and healthy lifestyle?! An experimental intervention study
Randomized controlled trial with four-month follow-up · 619 adults in Germany: 200 abstinence, 226 reduction, 193 control
Cutting smartphone use by one hour a day for one week, or giving it up entirely for a week, both reduced problematic use and symptoms of depression and anxiety and increased life satisfaction and physical activity. The one-hour reduction group kept its gains more steadily over four months than the abstinence group.
but: All outcomes self reported and the intervention lasted only one week. Published online in 2022, print issue 2023.
- Review 2022
Digital detox: An effective solution in the smartphone era? A systematic literature review
Systematic review · 21 intervention studies, 3,625 participants, 12 of them randomized trials
Digital detox interventions produced inconsistent results: some studies found gains in well-being, self control or relationships, others found no benefit, and a few found worse well-being during abstinence. The authors conclude the evidence does not yet support blanket claims that detoxes work.
but: Included studies varied widely in design, duration and what was restricted, so the review could not pool effects. Useful mainly as a corrective to strong claims in either direction.
Phones and the people in the room
Phubbing, looking at your phone while with your partner, goes with lower relationship satisfaction across dozens of studies. Correlational, but large and consistent.
- Review 2025
A meta-analytic study of partner phubbing and its antecedents and consequences
Meta-analysis of correlational studies · 52 studies, 58 samples, 19,698 participants
Across 52 studies, partner phubbing was associated with lower relationship satisfaction (r = -0.22), lower intimacy (r = -0.27), lower partner responsiveness (r = -0.29) and more conflict. Problematic phone or media use was the strongest predictor of phubbing.
but: Nearly all underlying studies are cross-sectional self-report surveys, so this establishes a consistent association, not causation. Frontiers journals are peer reviewed but have a mixed reputation; the pooled estimates are in line with earlier reviews.
- Survey 2016
My life has become a major distraction from my cell phone: Partner phubbing and relationship satisfaction among romantic partners
Two cross-sectional surveys · 308 adults (scale development) and 145 adults in relationships (model test)
People who reported that their partner often looked at a phone during time together reported more conflict over phone use and lower relationship satisfaction, which in turn went with lower life satisfaction and more depressive symptoms.
but: Cross-sectional and self reported, so direction of effect is unknown; unhappy partners may notice or resent phone use more. The scale it introduced has been widely reused, and a 2025 meta-analysis of 52 studies confirms a small to moderate association.
Children's eyes
Screen time is associated with short-sightedness in children, rising above about an hour a day. Time outdoors is the best-established protection, and many studies didn't adjust for it.
- Review 2025
Digital Screen Time and Myopia: A Systematic Review and Dose-Response Meta-Analysis
Systematic review and dose-response meta-analysis · 45 studies, 335,524 participants, mean age 9.3
Each additional hour of daily screen time was associated with 21 percent higher odds of myopia. The curve was flat below about one hour a day and rose steeply between one and four hours, suggesting a possible safety threshold under an hour.
but: Mostly observational studies with self-reported screen time; the association is not proof of cause. The authors say the threshold finding needs confirmation before it guides clinical advice.
- Review 2021
Association between digital smart device use and myopia: a systematic review and meta-analysis
Systematic review and meta-analysis · 11 studies, participants aged 3 months to 33 years
Smartphone or tablet use was associated with higher odds of myopia (OR 1.26, borderline significant), and combined smart device plus computer use with higher odds still (OR 1.77). The authors describe the evidence as mixed and of variable quality.
but: High heterogeneity, and only 40 percent of studies measured refraction objectively. Many did not adjust for time outdoors, which is the best-established protective factor for myopia.
How this list was put together
Every entry was checked against the publisher's page or its DOI record for the exact title, authors, journal and year, and the finding was written from the abstract or the paper itself rather than from press coverage. Studies that failed replication are marked as such rather than dropped, because you'll run into them elsewhere. A few well-known papers were left out on purpose: media-multitasking work whose effects didn't hold up in later meta-analyses, and a 2024 meta-analysis of social media restriction experiments that is still being argued over in print. Last audited September 2026. If a link breaks or a newer study changes the picture, tell me.
None of this is medical advice. If sleep, mood or a child's development worries you, that's a conversation for a doctor, not a website about flip phones.
Social media, mood, and depression
The best causal studies find real but modest effects. The largest correlational datasets find associations too small to matter on their own. Deactivating Facebook for a month helped a little; so did capping social apps at ten minutes a day. Neither the catastrophe nor the nothing that the loudest voices claim.
Social Media and Youth Mental Health: The U.S. Surgeon General's Advisory
Office of the U.S. Surgeon General. U.S. Department of Health and Human Services .
Public health advisory (evidence summary and recommendations)
The advisory concludes there is not yet enough evidence to say social media is sufficiently safe for children and adolescents, notes that up to 95 percent of 13 to 17 year olds use it, and cites a study in which adolescents using social media more than three hours a day had about double the risk of depression and anxiety symptoms.
but: A government advisory, not a peer-reviewed study. It summarizes correlational research and emphasizes uncertainty. The three-hour figure comes from a single cohort analysis.
Social Media and Mental Health
Braghieri L, Levy R, Makarin A. American Economic Review , 112(11):3660-3693. doi:10.1257/aer.20211218
Quasi-experimental (staggered rollout, difference-in-differences) · More than 430,000 student survey responses across 775 US colleges, 2000 to 2008
Using the college-by-college rollout of early Facebook as a natural experiment, the authors find that Facebook's arrival worsened a student mental health index by about 0.085 standard deviations and increased reports of depression and of academic impairment due to mental health. They estimate this at roughly 22 percent of the effect of losing a job.
but: Applies to Facebook in 2004 to 2008, a very different product from today's platforms. The design supports a causal reading but the effect is modest.
Social media use and its impact on adolescent mental health: An umbrella review of the evidence
Valkenburg PM, Meier A, Beyens I. Current Opinion in Psychology , 44:58-68. doi:10.1016/j.copsyc.2021.08.017
Umbrella review (review of reviews) · 25 reviews published 2019 to mid 2021, including 7 meta-analyses and 9 systematic reviews
Most reviews of social media and adolescent mental health describe the associations as weak or inconsistent, although some reviews label the same small associations as substantial. The evidence base is dominated by cross-sectional self-report studies.
but: A narrative synthesis, not a new meta-analysis. It summarizes the state of the literature rather than testing an effect.
The Welfare Effects of Social Media
Allcott H, Braghieri L, Eichmeyer S, Gentzkow M. American Economic Review , 110(3):629-676. doi:10.1257/aer.20190658
Randomized field experiment · 1,661 US Facebook users randomized (2,743 recruited), four weeks
Paying people to deactivate Facebook for four weeks raised a combined measure of subjective well-being by about 0.09 standard deviations, a small but real improvement, and freed up about an hour a day. Deactivation also reduced political polarization and news knowledge, and people used Facebook less after the study ended.
but: Participants were self-selected adults willing to deactivate for a payment. The effect is small in absolute terms; the authors compare it with 25 to 40 percent of the effect of typical psychological self-help interventions.
The association between adolescent well-being and digital technology use
Orben A, Przybylski AK. Nature Human Behaviour , 3(2):173-182. doi:10.1038/s41562-018-0506-1
Cross-sectional, specification curve analysis of three large datasets · 355,358 adolescents (YRBS, Monitoring the Future, Millennium Cohort Study)
Across thousands of possible analyses, digital technology use explained at most 0.4 percent of the variation in adolescent well-being. The association is negative but tiny, smaller than the association with wearing glasses, and the authors argue it is too small to justify policy on its own.
but: Correlational and based on self-reported hours, which are unreliable. Critics argue that averaging across all technology use can hide larger effects for specific groups, such as heavy social media use among girls.
No More FOMO: Limiting Social Media Decreases Loneliness and Depression
Hunt MG, Marx R, Lipson C, Young J. Journal of Social and Clinical Psychology , 37(10):751-768. doi:10.1521/jscp.2018.37.10.751
Randomized controlled trial · 143 undergraduates, three weeks
Students randomly assigned to cap Facebook, Instagram and Snapchat at 10 minutes each per day reported less loneliness and fewer depressive symptoms after three weeks than students who used social media as usual. Both groups showed lower anxiety and fear of missing out, which the authors attribute to self monitoring.
but: Small, single-campus sample and self-reported outcomes over a short period. Meta-analyses of similar restriction experiments disagree about how large and reliable such effects are.