Comprehensive Academic Study Guide: Social Media and Adolescent Mental Health

Methodological Framework: Product Safety vs. Historical Trends

• The Product Safety Question:

  • Focuses on whether social media use is reasonably safe for children and adolescents.

  • "Safe" denotes a consumer product whose ordinary use does not place young users at substantially elevated risk of mental health problems (specifically depression and anxiety) or regularly expose them to severe direct harms.

  • "Adolescence" spans the developmental period from roughly age $10$ to $19$. Puberty occurs during the initial phase (typically beginning at ages $9$ to $11$ and concluding by age $15$ or $16$), representing a sensitive period of extreme neural plasticity where environmental inputs alter neuronal growth, pruning, and myelination.

  • "Social media" refers to platforms incorporating user profiles, user-generated content, networking, interactivity, and algorithmic curation (such as Instagram, Snapchat, TikTok, Facebook, YouTube, Reddit, and X).

  • Ordinary common use among United States adolescents averages nearly $5\,\text{hours}$ per day ($4.8\,\text{hours}$ to $5.0\,\text{hours}$: $4.4\,\text{hours}$ for boys and $5.3\,\text{hours}$ for girls), with $25\%$ of $13$- to $14$-year-olds using platforms for $7\,\text{hours}$ or more daily.

  • Direct harms occurring through ordinary platform exposure include graphic pornography, real-life violence, cyberbullying, deepfakes, lethal challenges, contact with sexual predators, and illicit drug distribution.

• The Historical Trends Question:

  • Focuses on whether the global spread of always-available smartphone social media in the early 2010s substantially contributed to population-level increases in adolescent mental illness across Western nations.

  • Between 2015 and 2018, Programme for International Student Assessment (PISA) data across $47$ nations and territories revealed statistically significant declines in adolescent Life Satisfaction in $40$ countries, insignificant changes in $6$, and an increase in only $1$ (South Korea). Girls exhibited significant declines in $41$ nations and boys in $33$.

  • Cross-national adolescent school loneliness increased significantly in $34$ out of $35$ countries between 2012 and 2018 in the same PISA dataset.

• Standard of Proof: Preponderance of the Evidence:

  • In evaluating consumer product safety, the legal standard of "beyond a reasonable doubt" is inappropriate because the cost of a false positive (delaying account access to age $16$ while retaining other online communication channels) is low.

  • The cost of a false negative (falsely assuming safety) is catastrophic, subjecting millions of minors to clinical psychiatric illness, self-harm, and severe exploitation.

  • A civil standard—"preponderance of the evidence" (evaluating whether the product is probably safe or probably unsafe)—provides the proper legal and public policy benchmark, alongside the precautionary principle.

Evidence Line One: Testimony of Affected Youth (Adolescents and Gen Z)

• Adolescent Experiences and Firsthand Regret (Exhibit A):

  • Pew Research Center (2024–2025) surveys establish that nearly one-third of U.S. teens report being online "almost constantly."

  • Among adolescent girls surveyed by Pew, $50\%$ report social media negatively affects their sleep, $25\%$ report it damages their mental health, $20\%$ report it harms their confidence, and $34\%$ report it makes them feel worse about their lives ($20\%$ among boys).

  • Reported subjective benefits (e.g., $74\%$ feeling connected to friends' lives, $63\%$ gaining a creative outlet, $52\%$ finding support) do not establish product safety, just as positive subjective experiences from alcohol do not establish its safety for minors.

  • A Harris Poll survey of U.S. Gen Z adults (ages $18$ to $27$) revealed substantial regret: $50\%$ wished X had never been invented, $47\%$ wished TikTok had never been invented, $43\%$ wished Snapchat had never been invented, and $34\%$ wished Instagram had never been invented.

  • Regret for non-social media platforms was markedly lower: $15\%$ for YouTube, $17\%$ for Netflix, and $19\%$ for messaging applications. For TikTok, Gen Z men expressed significantly higher regret than women ($55\%$ vs. $39\%$).

• The Collective Action Trap and International Survey Data:

  • A 2023 investigation by Leonardo Bursztyn and colleagues demonstrated that $58\%$ of U.S. college students preferred a world without Instagram and $57\%$ without TikTok. Active users required a median compensation of \n  59 to deactivate TikTok and \n  47 to deactivate Instagram for one month.

  • When asked what compensation they would require if all their peers deactivated their accounts simultaneously, the required amount became negative; participants were willing to pay researchers to remove them and their peers from the platforms together.

  • A 2025 More in Common survey in the United Kingdom found that $62\%$ of youth aged $16$ to $24$ ($66\%$ among Gen Z women) believe social media causes more harm than good for children under $16$, with $55\%$ affirming life would improve if social media were banned for under-$16$s, compared to $22\%$ who stated it would be worse.

  • The British Standards Institution (BSI, 2025) reported that $47\%$ of UK youth aged $16$ to $21$ would prefer to have grown up in a world without the internet, and $50\%$ affirmed an online social media curfew would improve their lives.

  • Australian youth data (headspace, 2018) consistently identify social media as a primary driver of youth mental health decline.

• Internal Platform Surveys of Young Users (Exhibit B):

  • Internal research disclosed by Meta whistleblower Frances Haugen (2021) showed that across six nations (Brazil, India, Indonesia, Japan, Turkey, and the United States), $33.3\%$ of teenage girls reported that Instagram worsened their body image issues (against $22\%$ reporting improvement).

  • Internal qualitative findings submitted to Meta leadership concluded: "Teens blame Instagram for increases in the rate of anxiety and depression . . . This reaction was unprompted and consistent across all groups."

  • An internal survey of $50,590$ Instagram users across $10$ countries revealed that $37\%$ of teen girls reported frequently seeing content that worsened their body image, compared to $26\%$ of users overall.

  • Research led by Arturo Béjar via the Bad Encounters and Experiences Framework (BEEF, sampling over $200,000$ users) found that among Instagram users aged $13$ to $15$, $13\%$ received unwanted sexual advances within the previous seven days, $13\%$ witnessed violent, bloody, or disturbing content within the previous seven days, and $28\%$ reported witnessing bullying.

  • Legal filings by the Nebraska Attorney General (2024) revealed internal TikTok studies documenting that compulsive use directly causes "negative emotions," impairs sleep, damages academic performance, and that users found time management exceptionally difficult on TikTok relative to competing platforms.

Evidence Line Two: Eyewitness Testimony of Observers

• Parental Observations and Regret (Exhibit C):

  • A 2025 Pew Research Center survey found that $44\%$ of U.S. parents identify social media as the single most detrimental influence on youth mental health, surpassing technology broadly.

  • The 2025 UK More in Common survey showed that parents ranked "social media use/excessive screen time" ($45\%$) as the primary negative driver of their children's mental distress, followed by exposure to harmful online material ($32\%$) and cyberbullying ($31\%$).

• In a nationwide Harris Poll survey of $1,013$ U.S. parents, majorities expressed regret over the invention of social media platforms: $62\%$ wished TikTok had never been invented, $62\%$ wished X had never been invented, $60\%$ wished Snapchat had never been invented, $56\%$ wished Instagram had never been invented, and $53\%$ wished Facebook had never been invented.

• The parental desire to eliminate TikTok and X ($62\%$) equaled the desire to eliminate guns ($62\%$) and surpassed alcohol ($57\%$), smartphones ($38\%$), and bicycles ($9\%$).

• The 30-country Ipsos Education Monitor (2025) identified mental health as the preeminent parental concern and documented clear majorities across all $30$ surveyed nations supporting statutory bans on social media for youth under age $14$.

• Educator Perceptions of Educational and Psychological Harm (Exhibit D):

  • A 2024 survey of $2,889$ U.S. educators by the National Education Association (NEA) found that following "lack of parental involvement" ($92\%$), the primary factors damaging student mental health were "social media use" ($84\%$) and "cell phone or personal device use after school hours" ($81\%$).

  • An NBC survey of $559$ U.S. school principals revealed that $42.2\%$ viewed smartphones and social media as major drivers of deteriorating pupil mental health, while only $1.3\%$ regarded concerns as overblown; only $3.4\%$ observed academic benefits and $5.6\%$ noted mental health benefits.

  • Pew Research (2024) documented that $72\%$ of U.S. high school teachers identify classroom mobile phone distraction as a severe problem.

  • An EdWeek Research Center survey of $595$ educators found that approximately $90\%$ believe social media harms pupils' interpersonal communication skills, behavioral treatment of peers, and adolescent female mental health.

  • A European Union educational survey established that $71\%$ of educational personnel report mobile devices degrade cognitive focus, $56\%$ report harms to well-being and social interaction, and $54\%$ report deterioration of the overall school environment.

• Clinical Professionals' Evaluations (Exhibit E):

  • Meta conducted an internal mixed-methods study sampling over $1,000$ licensed mental health practitioners (psychiatrists, clinical psychologists, therapists, and licensed clinical social workers, each possessing at least two years post-licensure experience and active caseloads).

  • Discovery records from U.S. school district Multi-District Litigation revealed that $81\%$ of surveyed clinicians stated social media exacerbates anxiety disorders in patients, $78\%$ confirmed it worsens depressive disorders, and $85\%$ identified social media as addictive.

  • No professional medical, psychiatric, or pedagogical organization has produced evidence characterizing social media platforms as net-positive contributors to adolescent psychological health.

Evidence Line Three: Internal Corporate Disclosures and Admissions

• TikTok Corporate Records (Exhibit F):

  • Unredacted briefs from the Attorneys General of Kentucky and Nebraska (2024) disclosed internal assessments from TikTok's research division ("TikTank").

  • TikTank documentation confirmed: "Compulsive usage correlates with a slew of negative mental health effects like loss of analytical skills, memory formation, contextual thinking, conversational depth, empathy, and increased anxiety," while impairing "sufficient sleep, work/school responsibilities, and connecting with loved ones."

  • High-level executive statements recorded in internal memos acknowledged: "The product in itself has baked into it compulsive use," and warned that the algorithm displaces basic biological and developmental needs: "sleep, and eating, and moving around the room, and looking at somebody in the eyes."

  • Corporate memoranda acknowledged minor users possess "minimal ability to self-regulate effectively," lack the requisite executive functioning to curtail screen exposure, and exhibit heightened sensitivity to social feedback rewards.

• Snap Inc. Internal Records (Exhibit G):

  • Filings from the New Mexico Attorney General (2024) revealed that Snap Inc.'s Trust and Safety division received approximately $10,000$ reports of minor sextortion per month in 2022 ($120,000$ annually), with internal acknowledgments that this metric represented a small fraction of actual occurrences due to underreporting.

  • In March 2023, internal communications documented personnel stating: "God I'm so pissed that we're over-run by this sextortion shit right now. We've been twiddling our thumbs and wrung our hands all f…ing year."

  • A February 2022 in-app reporting research deck prepared by Snap consultants revealed that the platform's core ephemeral design ("disappearing messages") systematically facilitated cyberbullying by eliminating accountability and leaving victims without evidentiary recourse.

  • Internal decks acknowledged that platform friending mechanisms ("Quick Add") allowed adult illicit drug dealers to target adolescent users, leading to fatal fentanyl poisoning overdoses directly facilitated through platform architecture.

• Meta Internal Records (Exhibit H):

  • Internal communications sent to Instagram head Adam Mosseri specified: "Our overall goal remains total teen time spent . . . with some specific efforts (Instagram) taking on tighter focused goals like U.S. teen total time spent," reinforced by 2018 strategic guidance: "the young ones are the best ones. You want to bring people to your service young and early."

  • Sean Parker, founding president of Facebook, admitted publicly in 2017: "How do we consume as much of your time and conscious attention as possible? . . . give you a little dopamine hit every once in a while . . . It's a social-validation feedback loop . . . you're exploiting a vulnerability in human psychology. The inventors, creators—it's me, it's Mark [Zuckerberg], it's Kevin Systrom . . . understood this consciously. And we did it anyway."

  • A Meta core data scientist holding a neuroscience doctorate warned internally: "Intermittent rewards are most effective (think slot machines) reinforcing behaviors that become especially hard to extinguish—even when they provide little reward, or cease providing reward at all."

  • Internal user-experience research dialogues recorded staff stating: "IG is a drug . . . We're basically pushers . . . We are causing Reward Deficit Disorder bc people are binging on IG so much they can't feel reward anymore . . . the top down directives drive it all towards making sure people keep coming back for more."

  • Parallel corporate records from YouTube (2016 app strategy targeting "addictive" daily habits) and Roblox (acknowledging internal safety features were suppressed because they degraded active user engagement metrics) demonstrate industry-wide prioritization of user time extraction over child protection.

Evidence Line Four: Epidemiological and Cross-Sectional Studies

• Dose-Response Associations with Depressive Illness (Exhibit I):

  • Kelly et al. (2019) evaluated $10,904$ fourteen-year-olds via the UK Millennium Cohort Study, demonstrating a clear dose-response gradient between daily social media use and clinical depression.

  • Adolescents spending $5\,\text{hours}$ or more daily on social media exhibited double the rate of clinically significant depressive symptoms compared to users consuming under $1\,\text{hour}$ daily.

  • Heavy use ($\ge 5\,\text{hours/day}$) was reported by $25\%$ of $14$-year-old girls and $10\%$ of $14$-year-old boys.

• Relative Risk Calculations:

  • Among adolescent girls, heavy social media use carries an elevated depression relative risk of $RR = 2.65$.

  • This relative risk magnitude exceeds that of severe sleep deprivation ($< 7\,\text{hours}$ per night, $RR = 2.45$) and online harassment victimization ($RR = 2.56$).

  • The depression risk elevation associated with heavy social media exposure markedly exceeds the risk associated with household poverty ($RR = 1.06$).

  • A dose-response meta-analysis by Liu et al. (2022) found that the odds of depression increase by $13\%$ for every additional daily hour of social media use ($OR = 1.13$, $95\%\,\text{CI: } [1.09, 1.17]$). At current average usage levels ($\sim 5\,\text{hours/day}$), this reflects a greater than $50\%$ elevation in baseline depression risk.

• Resolution of Purported "Null" Findings Through Unblending (Exhibit J):

  • Widely cited meta-analyses asserting negligible effects—notably Hancock et al. (2022), which reported an aggregate correlation of $r = 0.01$—systematically obscure harm by aggregating disparate constructs.

  • Hancock et al. pooled internalizing pathology (anxiety, depression) with eudaimonic, hedonic, and external "social" well-being metrics across broad, non-adolescent populations.

  • When unblended to isolate internalizing psychiatric disorders, Hancock et al.'s own data show substantial effect sizes: $r = 0.18$ for clinical anxiety and $r = 0.13$ for clinical depression.

  • Similarly, Orben & Przybylski (2019) claimed digital media effects were trivial by blending screen types and irrelevant behavioral items (e.g., "shares readily," "eats potatoes").

  • Re-analysis of the identical datasets by Twenge et al. (2022) isolating social media use, female adolescents, and internalizing symptoms yielded standardized regression coefficients of $\beta \approx -0.20$ (ranging from $-0.15$ to $-0.25$), compared to $\beta = -0.04$ for boys.

Evidence Line Five: Longitudinal and Prospective Cohort Studies

• Prospective Prediction of Depressive Symptom Onset (Exhibit K):

  • Longitudinal methodologies track cohorts across multiple measurement intervals, establishing whether temporal exposure at Time 1 predicts pathological outcomes at Time 2 while controlling for baseline confounding.

  • Riehm et al. (2019) tracked $6,595$ U.S. adolescents aged $12$ to $15$ and proved that elevated baseline social media time prospectively predicted internalizing psychiatric symptoms at subsequent evaluation.

  • Nagata et al. (2025), utilizing the national Adolescent Brain Cognitive Development (ABCD) multi-year tracking dataset, established that longitudinal increases in social media exposure predict prospective increases in adolescent depressive symptoms.

  • Grund & Luciana (2025), analyzing the ABCD cohort, established that baseline internalizing pathology failed to predict subsequent increases in screen time, disproving the claim that the association is driven exclusively by reverse causality.

  • Vasconcellos et al. (2025) identified bidirectional pathways in younger cohorts, demonstrating that while psychological distress can increase device engagement, heavier baseline screen use robustly predicts downstream depressive symptom onset.

• Refutation of the "Reverse Predictability Only" Hypothesis:

  • Critical claims that longitudinal evidence reflects only pre-existing distress driving platform use (e.g., Odgers, 2022) misstate the underlying literature.

  • McClean & Lebwohl (2025) re-examined the three citations historically deployed to support the reverse-only claim:

  • Puukko et al. (2020) omitted measurements of time spent on algorithmically curated social media, examining instead direct messaging via WhatsApp and email.

  • Hancock et al. (2022) pooled disparate age brackets and screen categories, yet still found a distress-to-media cross-lagged pathway of zero.

  • Heffer et al. (2019) demonstrated reciprocal, bidirectional causality for adolescent girls, which confirms that Time 1 exposure directly contributes to Time 2 clinical distress.

Evidence Line Six: Randomized Controlled Trials (RCTs) of Usage Reduction

• Experimental Reduction of Depressive and Anxious Symptoms (Exhibit L):

  • Burnell et al. (2025) conducted a comprehensive meta-analysis of $32$ randomized controlled trials encompassing $5,544$ participants across $91$ discrete effect sizes.

  • Experimental allocation to restricted social media usage produced statistically significant reductions in psychiatric symptoms and elevations in psychological health:

  • Depression reduction: Hedges' $g = 0.19$

  • Anxiety reduction: Hedges' $g = 0.28$

  • Overall subjective well-being improvement: Hedges' $g = 0.22$

  • Among the $16$ RCTs providing unblended clinical depression inventories, $14$ demonstrated significant depressive symptom declines; among the $10$ measuring clinical anxiety, $8$ recorded marked symptom declines.

  • Davis & Goldfield (2025) executed an RCT targeting $220$ clinically distressed youth aged $17$ to $25$. Restricting social media to $1\,\text{hour}$ daily over three weeks generated significant declines in depression, anxiety, and fear of missing out (FOMO), alongside objective improvements in sleep continuity across male and female participants.

• Re-evaluating Flawed Methodological Designs in Reduction Literature (Exhibit M):

  • Ferguson (2025) reported an aggregate effect size across $27$ reduction experiments of $d = 0.088$ ($95\%\,\text{CI: } [-0.018, 0.197]$), claiming experimental interventions produce null effects.

  • Ferguson's methodology suffered from fatal temporal blending: pooling ultra-short abstinence protocols ($24$ to $48\,\text{hours}$) with multi-week reduction protocols.

  • Re-analysis by Thrul et al. (2025) revealed that acute abstinence under one week induced behavioral withdrawal (irritability, agitation, craving, negative affect; $d = -0.175$), whereas sustained usage reduction lasting longer than one week generated significant psychiatric improvement ($d = 0.156$).

  • Stein (2024) isolated clinical internalizing scales within Ferguson's dataset, deriving an effect size of $d = 0.22$.

  • Lemahieu et al. (2025) documented null findings by restricting inclusion to brief interventions measuring momentary affect and general life satisfaction while excluding clinical anxiety and depression.

• Meta's Internal Intervention Experiment: Project Mercury (Exhibit N):

  • In 2020, Meta partnered with media analytics firm Nielsen to conduct a randomized controlled trial code-named "Project Mercury."

  • Users were randomly assigned to deactivate their Facebook accounts for one week.

  • Meta's internal documentation summarized the causal outcome: "people who stopped using Facebook for a week reported lower feelings of depression, anxiety, loneliness, and social comparison."

  • Internal staff reactions warned that suppressing the results duplicated tobacco industry cover-ups: "doing research and knowing cigs were bad and then keeping that info to themselves."

  • Meta leadership subsequently aborted the research program and publicly dismissed the findings as contaminated by external media narratives.

Evidence Line Seven: Natural Experiments and Broadband Expansion

• Quasi-Experimental Infrastructure Rollouts (Exhibit O):

  • Natural experiments exploit the geographic, staggered timing of high-speed broadband infrastructure expansion during the late 2000s and early 2010s to estimate population-level causal impacts without self-selection biases.

  • Cross-national studies leveraging instrumental variables consistently show marked psychiatric declines following broadband access:

  • Germany (Golin, 2022): High-speed internet access caused sharp declines in self-reported mental health that disproportionately expanded the female psychiatric gender gap.

  • Italy (Donati et al., 2022): High-speed connectivity generated significant increases in hospital admissions for adolescent psychiatric and behavioral disorders.

  • Spain (Arenas-Arroyo et al., 2025): Broadband infrastructure caused a substantial rise in hospital diagnoses of adolescent mental illness, concentrated heavily among girls.

  • United States (Kyung et al., 2021): Broadband penetration caused significant increases in youth suicide rates and elevated clinical depression among college students.

  • McDool et al. (2020) demonstrated that high-speed internet arrival in England reduced adolescent subjective well-being, driven specifically by heavy use ($> 5\,\text{hours/day}$) of social networking applications.

  • Churchill & Johnson (2026) established that U.S. broadband introduction from 2009 to 2019 increased adolescent suicidal behaviors, mediated by cyberbullying and body image distress among girls and acute sleep deprivation among boys.

  • Braghieri, Levy, & Makarin (2022) exploited the staggered rollout of Facebook across U.S. university campuses from 2004 to 2006, demonstrating that platform introduction immediately increased student rates of depression and generalized anxiety disorder.

  • The singular contrary study (Johnson & Persico, 2024/2025), which linked broadband between 2000 and 2008 to reduced suicide, examined an era preceding modern algorithmically curated mobile social networks and attributed findings to local economic revitalization.

Population-Level Impact: Direct and Indirect Harm Quantification

• Scale of Direct Exploitation in the United States:

  • The 2023 U.S. adolescent population (ages $10$ to $19$) is approximately $43\,\text{million}$. With $90\%$ adoption, roughly $39\,\text{million}$ adolescents actively use social media.

  • Compulsive Usage/Addiction: Operating on Meta's internal finding that $12.5\%$ of users suffer compulsive impairment affecting sleep, education, and relationships, a baseline rate of $10\%$ indicates that $3.9\,\text{million}$ U.S. minors experience clinical behavioral addiction.

  • Sleep Deprivation: Pew data (2025) shows $45\%$ of U.S. adolescents ($17.5\,\text{million}$ individuals) suffer chronic sleep reduction caused directly by social media engagement.

  • Minor Sextortion: Snap Inc.'s internal metric of $10,000$ reports/month ($120,000$/year) represents a fraction of total activity; cross-platform annualized victim counts exceed $1\,\text{million}$ minors.

  • Severe Sexual Harassment: Arturo Béjar's BEEF metrics demonstrate that $13\%$ of users aged $13$ to $15$ receive unwanted sexual solicitation in a single seven-day window, indicating an ongoing scale of $5.7\,\text{million}$ adolescent victims weekly in the U.S. alone.

  • Total direct harms—encompassing cyberbullying ($10.8\%$ weekly, $59\%$ lifetime), algorithmic exposure to self-harm content ($8.4\%$ weekly), and illicit drug purchases leading to overdose deaths—impact over $10\,\text{million}$ U.S. children annually.

• Epidemiological Translation of Indirect Harm Effect Sizes:

  • Experimental reduction studies establish an indirect harm effect size of Hedges' $g = 0.22$ on validated psychological scales such as the Warwick–Edinburgh Mental Wellbeing Scale (WEMWBS, normally distributed with $\mu = 51$, $\sigma = 7$, with scores $< 41$ designating probable clinical depression).

  • Translating a sustained shift of $0.22\,\sigma$ across a normal population distribution demonstrates that a two-week reduction in social media lowers the prevalence of clinical depression by approximately one-third ($33.3\%$).

  • Conversely, resuming normative daily platform use expands the clinical depression pool by one-half ($50\%$).

  • For context, childhood physical and emotional maltreatment accounts for increased depression and anxiety risks corresponding to Cohen's $d = 0.22$ and $d = 0.25$, respectively (Grummitt et al., 2024).

• Group-Level Network Effects and Structural Traps:

  • Individual-level experimental models systematically understate population impacts by failing to account for network externalities.

  • The wholesale migration of adolescent communication into virtual networks eradicated unstructured face-to-face peer socialization, physical play, tactile contact, and continuous eye contact beginning in the early 2010s.

  • Non-users are trapped: an individual adolescent who disconnects from social media preserves individual cognitive clarity but suffers severe social isolation and exclusion.

  • The Bursztyn et al. (2023) college experiments confirm social media exists as a structural collective-action trap where users collectively pay to eliminate an ecosystem that traps them individually.

• Sensitive Period Windows in Neurodevelopment:

  • Heightened neural plasticity during puberty renders the adolescent brain uniquely vulnerable to algorithmic intermittent reinforcement and social comparison feedback loops.

  • Orben et al. (2022) identified discrete developmental windows of sensitivity where social media use strongly predicts declining life satisfaction:

  • Girls: Ages $11$ to $13$

  • Boys: Ages $14$ to $15$

  • Imposing algorithmic behavioral feedback during sensitive neurodevelopmental windows risks permanently altering neurological pathways governing attention, emotional regulation, and social bonding, compounding population-level morbidity.