A 2025 meta-analysis in PLOS ONE reviewing dozens of studies found that social media addiction is significantly associated with higher anxiety, depression, and loneliness in students.^[1] A separate 2024 JAMA Pediatrics meta-analysis of over 1 million adolescents confirmed a positive link between social media use and internalizing symptoms like anxiety and depression.^[2] A comprehensive 2024 review covering 182 studies and over 1.1 million participants found significant associations between social media use, depression, anxiety, and sleep problems.^[3]

Rates of teen depression, anxiety, and self-harm were stable throughout the 2000s, then shot upward right around the time smartphone adoption crossed 50% and social media went mobile. Among U.S. college students, diagnosed anxiety rose 134% and depression 106% between 2010 and 2018.^[4] ER visits for self-harm among 10-to-14-year-old girls rose 188%.^[5] Similar trends appear across the U.S., UK, Canada, Australia, Scandinavia, and beyond – in a study spanning 37 countries, nearly twice as many adolescents in 2018 vs. 2012 had elevated levels of school loneliness.^[6]

A 2025 study in JAMA Network Open found that young adults who took just a one-week break from social media saw a 24% decrease in depression symptoms, 16% reduction in anxiety, and 14.5% decrease in insomnia.^[7] Georgetown University research found that improvements from digital detox were comparable in magnitude to cognitive-behavioral therapy and larger than the typical effect of antidepressants.^[8] That’s one week.

A note on the debate: Not all researchers agree on the strength of the causal link. A 2024 review in Nature by psychologist Candice Odgers argued that the evidence is more mixed than some have presented, and that economic hardship and other factors also contribute.^[9] We think the honest answer is: social media is almost certainly not the only cause. But a 2025 consensus statement signed by dozens of researchers concluded it is a substantial contributor.^[10] Either way, trying one day off per week is a pretty low-risk experiment.

A 2025 Karolinska Institutet study following over 8,000 children for four years found that social media specifically – not TV, not video games – was associated with a gradual decline in the ability to concentrate.^[11] A 2025 NTU Singapore study found 68% of young people report social media harms their ability to focus, with many unable to engage with content longer than one minute.^[12]

A systematic review of 23 studies found that excessive social media use is associated with impaired attention, reduced working memory, and diminished executive functioning.^[13] Heavy users (5+ hours daily) are 33% more likely to experience attention fragmentation symptoms.^[14] The average teen receives roughly 192 notifications per day from social media apps – about one every five waking minutes.^[5]

This is the good news. Georgetown University research found that a two-week digital detox improved sustained attention by an amount comparable to reversing roughly 10 years of age-related cognitive decline – and 91% of participants improved on at least one major outcome.^[8] A separate study of 150 participants found measurably faster reaction times and improved cognitive efficiency after just one week, with 65% maintaining reduced usage a month later.^[15] Your focus can come back. You just have to give it a chance.

A report from NYU Stern’s Center for Business and Human Rights, based on 50+ studies and 40+ expert interviews, concluded that while social media didn’t create political polarization, it significantly intensifies it.^[16] Internal Facebook documents revealed that a 2018 algorithm change inadvertently heightened anger and divisiveness – but Zuckerberg resisted fixes because they might hurt engagement.^[17]

A 2025 study published in Science found that simply reranking posts on X to reduce partisan hostility caused participants’ feelings toward the opposing party to warm by about 2 points on a 100-point scale – in just one week. That’s equivalent to roughly 3 years of attitude change in the general U.S. population.^[18] The algorithm isn’t neutral. It’s actively shaping how you feel about your neighbors.

A systematic review of 94 articles consistently found that partisan media exposure exacerbates polarization.^[19] Content that triggers outrage, moral indignation, and us-vs-them thinking generates more likes, comments, and shares. The algorithms learn this and serve it up. When you step away for a day, you break that cycle – even briefly.

Meta, TikTok, and X all collect your name, email, phone number, date of birth, location, contacts, payment information, every post, message, photo, video, and audio you create, plus detailed behavioral data on how you interact with the platform.^[20] Meta integrates this data across Facebook, Instagram, and WhatsApp for comprehensive profiling. TikTok collects biometric data including faceprints and voiceprints.^[21]

A URL Genius study found that TikTok and YouTube contact 14 different tracking domains during a single session – before you even log in. 13 of TikTok’s 14 contacts were third-party trackers, meaning outside companies you’ve never heard of are collecting data about your behavior. These trackers can follow your activity across other websites even after you leave the app.^[22]

Cambridge Analytica harvested up to 87 million Facebook users’ data to build psychological profiles and target political messaging.^[23] Meta was fined €1.3 billion by the EU for data transfer violations. Facebook’s own internal research found Instagram worsens body image issues for 1 in 3 teen girls who already struggle with body image – and the company kept that research private until the Wall Street Journal’s 2021 exposé.^[24] More recently, Meta and X have used public posts to train their AI models, opting everyone in by default.^[21]