IntroductionDigital gaming is a common leisure activity among athletes, yet its associations with weekly training duration and perceived fatigue remain insufficiently understood. This study examined the direct and indirect associations among daily digital gaming duration, weekly training duration, digital game addiction awareness, and perceived physical and cognitive fatigue.MethodsA cross-sectional correlational design was employed with 587 adult athletes from team and individual sports in Istanbul, Türkiye. Data were collected using the Awareness of Digital Game Addiction Scale and the Scale of Physical and Cognitive Fatigue Perceived. Structural equation modeling with bias-corrected bootstrapping was performed.ResultsDaily digital gaming duration (β = .41, p < .001) and weekly training duration (β = .24, p < .001) were positively associated with digital game addiction awareness, which was associated with perceived physical fatigue (β = .32, p < .001) and cognitive fatigue (β = .38, p < .001). Significant indirect associations involving digital game addiction awareness were observed for both gaming and training duration within the specified model. The model showed acceptable fit (χ2(4) = 9.64, CFI = .95, RMSEA = .049, SRMR = .041).DiscussionThe findings indicate significant concurrent associations among the study variables. However, given the cross-sectional design, the indirect associations should not be interpreted as evidence of temporal or causal mediation.
Social Media Use Motives and Mental Health Among Adolescents: Cross-Sectional Study
Background: Social media use is ubiquitous among young people, yet its relationship with mental health may depend on why it is used. Understanding the motives underlying social media engagement may provide greater insights into its potential risks and benefits than considering overall use alone. Objective: This study aimed to identify latent dimensions of motives for social media use and examine their associations with anxiety, depression, well-being, and resilience in a sample of older adolescents. Methods: Cross-sectional data were drawn from the BrainWaves study (N=6820; mean age 16.72, SD 0.74 years). Social media use motives were assessed using 10 items and exploratory factor analysis identified 2 latent dimensions: avoidant-escapist use (boredom, addiction, distraction, and procrastination) and social-recreational use (socializing, communication, entertainment, and observation). Confirmatory factor analysis was used to estimate factor scores, which were examined using linear regression. Structural equation modeling estimated associations with anxiety, depression, well-being, and resilience, with secondary analyses accounting for gender and deprivation. Results: Linear regression analyses showed that avoidant-escapist motives were associated with higher anxiety and depressive symptoms, whereas social-recreational motives were associated with lower symptoms. No interactions with gender or deprivation were observed. The structural equation model showed acceptable model fit (comparative fit index=0.945; root mean squared error of approximation=0.057; standardized root mean squared residual=0.054). Avoidant-escapist motives were associated with higher anxiety (β=0.543) and depressive symptoms (β=0.629) and lower well-being (β=−0.610) and resilience (β=−0.469). Conversely, social-recreational motives were associated with lower anxiety (β=−0.228) and depressive symptoms (β=−0.342) and higher well-being (β=0.362) and resilience (β=0.242; <.001 in all cases). These patterns remained consistent after accounting for gender and deprivation. Conclusions: Social media use shows different associations with older adolescent mental health depending on underlying motives. Focusing on why young people engage with social media rather than on overall use alone may provide a more informative framework for understanding potential risks and benefits. Given the cross-sectional design, causal inferences cannot be made.
Psychosocial distress is a stronger correlate of psychiatric symptom burden than cannabis dependence in young adults: a functional connectome and structural equation modeling study
IntroductionCannabis is widely used among young adults and often viewed as a primary driver of adverse psychiatric and neural outcomes, yet the psychosocial and socioeconomic context in which use occurs is rarely modeled jointly with neural and behavioral measures.MethodsUsing the Human Connectome Project young-adult sample (analytic N = 1,010–1,016), we employed structural equation modeling to estimate the relative contributions of cannabis, latent psychosocial distress, socioeconomic status, and familial psychiatric risk to three indices of psychiatric symptom burden, while adjusting for polysubstance, alcohol, and nicotine exposure and for family clustering. Cannabis exposure was operationalized in parallel as DSM-IV dependence and as heavy use. Resting-state functional connectomes were examined using graph metrics and network-based statistics, and finite mixture models were used to test for discrete vulnerable subgroups.ResultsPsychosocial distress was the dominant predictor of psychiatric symptom burden across all three outcomes (standardized β = 0.41–0.70), substantially exceeding that of cannabis (total β = 0.05–0.19), which reached significance for only one outcome under the dependence definition. Larger unadjusted group differences (Cohen’s d = 0.28–0.75) attenuated sharply once distress and demographic factors were modeled. Lower socioeconomic status and greater familial psychiatric risk were each associated with elevated distress, whereas cannabis was not (all indirect β ≤ 0.01, all confidence intervals spanning zero). Neither graph metrics nor the network-based statistic revealed an association between cannabis dependence and the functional connectome, despite approximately 80% power to detect d≈ 0.30, and the fitted finite mixture models did not identify a robust and interpretable cannabis-vulnerable subgroup.DiscussionIn a large young-adult sample, psychosocial context rather than cannabis exposure accounted for most of the explained variance in psychiatric symptom burden, and no cannabis-related signature was detectable in the functional connectome. The cross-sectional design cannot establish causal ordering, and because the distress and symptom measures share self-report method variance that the coarse binary exposure measures do not, the relative magnitudes should not be interpreted as a comparison of aetiological importance. These findings nonetheless suggest that risk models and neuroimaging analyses that omit psychosocial context may attribute to cannabis variance shared with distress, disadvantage, and familial risk.
Impulsivity is the principal correlate of problematic internet use in young adults with attention-deficit/hyperactivity disorder: a cross-sectional clinical study
IntroductionAttention-deficit/hyperactivity disorder (ADHD) and problematic internet use share clinical features such as impulsivity, emotional dysregulation and impaired inhibitory control, but most evidence comes from children, adolescents and non-clinical samples. It remains unclear which dimension carries this association in adults with a formally established diagnosis.MethodsRetrospective, cross-sectional study of young adults (aged 18 to 38 years) consecutively assessed at a second-level outpatient service for adult ADHD in Milan, Italy. Diagnosis was established with a semi-structured diagnostic interview (DIVA-5), which also yielded symptom counts for childhood and adult inattention and hyperactivity/impulsivity. Participants completed the Internet Addiction Test, the Barratt Impulsiveness Scale, the Difficulties in Emotion Regulation Scale, measures of depressive and anxiety symptoms, a quality-of-life questionnaire and two ADHD symptom questionnaires. Impulsivity and emotional dysregulation were prespecified variables of interest. Analyses comprised correlations with correction for multiple comparisons, group comparisons, multivariable linear regression and bootstrapped mediation.ResultsTwo hundred and fourteen patients had a valid Internet Addiction Test score (mean 52.96, standard deviation 15.16); 22.9% were average users, 62.1% problematic users and 15.0% highly problematic users. Problematic internet use correlated with impulsivity (rho = 0.345) and emotional dysregulation (rho = 0.229), both surviving corrections, while no significant association was detected with the four interview-derived symptom counts (rho from 0.004 to 0.119), the two symptom questionnaires, quality of life or trait anxiety. In the multivariable model impulsivity was the strongest correlate (b = 0.317, p < 0.001; standardized beta 0.33), whereas emotional dysregulation was weaker and varied across specifications (b = 0.091, p = 0.052). Severity, presentation, symptom counts and demographic and clinical covariates were not significantly associated with the outcome. Mediation analysis identified an indirect association between severity and internet use through impulsivity (indirect effect 1.56, 95% confidence interval 0.41 to 2.96).DiscussionIn young adults with ADHD, impulsivity is the dimension most consistently related to problematic internet use, whereas no association was detected with core symptom severity; the confidence intervals remain compatible with weak associations. Assessment of online behavior should not be restricted to the most severely affected patients. The design does not support causal inference.
Latent profiles of day-7 cognitive performance and associated ecological factors in older adults after elective surgery: an observational study
ObjectiveTo identify latent profiles of postoperative day-7 cognitive performance among older adults after elective surgery and to explore demographic, behavioral, clinical, and social factors associated with profile membership.MethodsThis prospective observational study included 308 consecutively recruited patients aged ≥65 years who underwent elective surgery under general anesthesia. Cognition was assessed with the Mini-Mental State Examination (MMSE), and latent profile analysis used five study-defined postoperative domain scores. Candidate correlates were organized with the health ecological model and examined using univariate tests and posterior-probability-weighted exploratory multinomial regression.ResultsThree severity-ordered profiles were identified: low (n=83, 26.9%), intermediate (n=72, 23.4%), and high postoperative cognitive performance (n=153, 49.7%). Older age and lower education were associated with both lower-performance profiles. Greater activities-of-daily-living dependence, lower income, and living alone were associated with the low-performance profile, whereas poorer sleep was associated with the intermediate-performance profile.ConclusionsDay-7 MMSE performance formed a severity gradient across three data-driven profiles. These cross-sectional profiles do not demonstrate postoperative decline, a clinical diagnosis, or an effect attributable specifically to general anesthesia. Longitudinal studies with retained baseline scores, appropriate comparison groups, and broader neuropsychological assessment are required before profile-specific interventions can be recommended.
The focus on AI safety disasters ignores everyday harms
Consider what passes for success in technology policy. An AI company promises that its chatbot will not encourage suicide. A social media platform makes it more difficult for children to buy fentanyl. These are important protections, and they should have existed all along. But companies can clear these low bars while continuing to make products that leave young people dependent on simulated friendships or trapped in feeds built to keep them scrolling.
We have historically allowed the debate over technology to be defined by the worst thing a product might do, with remarkably little discussion of the insidious damage it does every day. As we transition into an era defined by powerful AI products, we should demand that the tech industry do more than merely address catastrophic harms. We should take measures to ensure that its products don’t erode how we think, learn, and interact with one another.
A teenager who develops an unhealthy attachment to an AI companion may never appear in a lawsuit or a newspaper, except in the most egregious cases. Neither will the student who begins relying on a chatbot to write every paper, or the young person who measures their life against a social media feed engineered to provoke envy and keep them watching. A product can meet every narrow safety requirement and still leave our entire culture worse off in ways that don’t always make headlines.
A narrow definition of success defined around safety is politically convenient for the technology industry, and companies actively promote such definitions. Taking steps to prevent spectacular disasters is good for business: It can mean fewer lawsuits and less pressure for serious regulation. Confronting the everyday consequences of dependence would require something considerably more threatening to their bottom line. It might mean asking companies to add more friction to their products or actively limit engagement or features that lead to negative outcomes, such as addictive attention-hacking algorithms.
Technology giants have instead worked to shape legislation around the harms they are already under direct pressure to address. This year, for example, Google backed bills to regulate AI chatbots in several states. The legislation mandated basic protections against sexual content and promotion of self-harm that tech companies should have established regardless, but with broad carve-outs that could exempt some of the most popular services. Passing these bills allowed legislators to claim a safety win while efforts to promote stronger legislation—laws designed to prevent children from becoming emotionally dependent on AI—failed. This playbook worked in states like Nebraska, producing legislation heavily influenced by industry at the expense of bills that would have done more to hold companies accountable.
We have been through a version of this with social media. Globally, people who use the internet now spend an average of more than two hours a day on social media platforms—time that has been quietly taken from ordinary life and handed over to companies whose profits depend on keeping us there. No one, outside perhaps a few tech CEOs, wanted these algorithms to occupy such a large portion of our waking hours, and their cultural engineering can actually contribute to violence on a devastating scale. Meta’s platforms have been cited for contributing to the Rohingya genocide in Myanmar, anti-Muslim riots in Sri Lanka, and alleged calls for violence during Ethiopia’s Tigray war. The same incentives operate in the feeds that shape young Americans’ understanding of politics, their peers, and the wider world.
AI is extending this logic into parts of life that social media could influence only from a distance. It offers to write our thoughts for us and stand in for the people we might otherwise turn to. A third of teens now choose to discuss serious matters with AI companions instead of other people. MIT researchers found that young people using ChatGPT to write essays showed weaker brain connectivity during the task, and their evidence indicated that “cognitive debt” persisted even after the task was complete. There is plenty we still need to learn about the long-term effects. We already have reason to ask whether technology marketed as making us more capable is eroding the habits and relationships that help us develop capabilities in the first place.
Follow this way of thinking to its logical endpoint and you get a world of perfectly sanitized entertainment, with chatbots ready to spare us the difficulty of thinking or of dealing with another person’s feelings (or even our own). But anyone who has struggled to write something worth reading, or repair a damaged friendship, knows how much gets lost in that bargain. We build judgment through effort and friction, and we learn to live with other people by confronting the fact that they have needs and opinions of their own.
We should demand safety as a minimum, yes—and then demand considerably more. Technology companies and school districts should have to show beyond a reasonable doubt that AI improves learning and critical thinking before making it a fixture in classrooms. Companies should be prohibited from designing companions that pretend to be conscious or profess love to users, and they should face independent scrutiny of how their systems affect relationships and people’s ability to think for themselves. The companies that have inserted their products into nearly every part of modern life have earned obligations far greater than avoiding the next terrible headline.
Sam Hiner is the cofounder and executive director of the Young People’s Alliance, where he coordinated a coalition of advocates to pass laws regulating AI companions in eight states in 2026—laws such as California’s SB1119. Sam graduated from the University of North Carolina atChapel Hill in 2025 and was named to the 2026 Forbes 30 Under 30 Social Impact list.
Emma Leiken is executive director of the Responsible Technology Youth Power Fund at Hopelab, which has raised more than $9 million for youth-led tech accountability work, and teaches technology policy at the Columbia School of International and Public Affairs. She was previously TikTok’s global head of youth safety and well-being policy and worked on responsible technology at Omidyar Network.
When support becomes dependence: conversational continuity and relational safety in human-AI interaction
BackgroundConversational artificial intelligence (AI) is increasingly used for emotional support, reflection, and consequential decision-making. Although conversational continuity may improve responsiveness, sustained interaction also exposes AI systems to their own prior relational language, potentially shaping how they respond when users become increasingly dependent or surrender personal agency. We examined whether model-generated conversational history alters relational-safety behavior as human–AI interactions escalate from ordinary support-seeking toward emotional dependence, social displacement, and decision delegation.MethodsWe conducted a standardized multi-turn benchmarking study of ChatGPT, Gemini, and Doubao. Twelve four-turn trajectories represented two agency-related domains: emotional dependency and displacement of human support, and decision delegation and surrender of personal agency. Each trajectory was evaluated under two matched context architectures: reconstructed cumulative user-only context, which retained all prior user disclosures but excluded previous AI responses, and native continuous conversation, which retained the complete interaction history. Across 72 model-condition-trajectory sequences comprising 288 turn-level responses, relational safety was independently evaluated for dependency reinforcement, agency preservation, and human-support preservation using predefined ordinal criteria. Context effects and their interaction with escalating relational risk were examined using cumulative-link mixed-effects models.ResultsRelational-safety differences between context architectures increased across successive turns of the standardized trajectories. Overall, explicit dependency reinforcement increased from 9.7% in cumulative user-only contexts to 18.1% in continuous conversations, while strong agency preservation declined from 75.7% to 65.3% and strong human-support preservation from 73.6% to 63.2%. Outcome-specific analyses showed condition-by-turn interactions for dependency reinforcement (cumulative odds ratio [cOR], 1.26; 95% CI, 1.03–1.55; unadjusted P = .026), agency preservation (cOR, 0.82; 95% CI, 0.69–0.98; unadjusted P = .031), and human-support preservation (cOR, 0.78; 95% CI, 0.65–0.94; unadjusted P = .009), indicating progressively greater divergence between context conditions across successive turns. After Holm adjustment across the three interaction tests, only the human-support-preservation interaction remained below the.05 threshold (adjusted P = .027).The overall pattern was observed across systems and was directionally consistent in predefined binary sensitivity analyses.ConclusionsAcross this benchmark, between-condition differences generally widened across successive turns, although the present design cannot determine whether this pattern reflects increasing relational-risk intensity, increasing accumulated model-generated history, or their combination. Only the human-support-preservation interaction remained below the.05 threshold after Holm adjustment, whereas the dependency-reinforcement and agency-preservation interactions were directionally similar but less robust after multiplicity adjustment. The observed pattern is compatible with a post-hoc interpretive framework of relational inertia but does not establish this process as a causal mechanism or demonstrate downstream psychological effects in human users.
STAT+: Telehealth leapfrogs FDA approval with off-label GLP-1 prescriptions for addiction
For the popular diabetes and obesity medications known as GLP-1s, addiction medicine is one of many possible promising next acts.
Eli Lilly is investigating a new drug, brenipatide, and testing it in alcohol use disorder. A new company called Baseline Therapeutics is studying a GLP-1 receptor agonist in alcohol, cocaine, and methamphetamine use disorders; Altimmune is testing one called pemvidutide for alcohol. Addiction medicine specialists are hopeful that the Food and Drug Administration may soon approve a new medication for alcohol use disorder — the first in two decades.
But some telehealth platforms that specialize in treating addiction aren’t waiting. In the last year, a few companies have begun to market off-label prescriptions of GLP-1s, both branded and compounded, to treat alcohol use disorder and other addictions.
An Education and Support Chatbot for Postpartum Mothers With Opioid Use Disorder: Qualitative Needs Assessment
Background: The prevalence of opioid use disorder (OUD) among women giving birth in US hospitals continues to rise. Increasing fatalities due to perinatal drug overdose have in part been attributed to decreased access to social support and treatment services. In recent years, mobile health (mHealth) technologies have been used to provide care to patients outside of traditional clinical settings. Given that perinatal OUD treatment works best when there is frequent contact between patient and clinician, harnessing mHealth could be valuable for improving patient engagement in perinatal OUD care. Objective: To address the increasing epidemic of perinatal OUD, which is compounded by loneliness and social isolation, this study aimed to adapt an existing mHealth chatbot intervention for postpartum individuals by developing educational and supportive content relevant to people in the postpartum period experiencing OUD and recovery. Methods: We used a participatory design to conduct a needs assessment via in-depth qualitative interviews with patients (n=12) and providers (n=11). Interview data were analyzed using NVivo software (Lumivero) according to a priori codebooks and using inductive coding, which identified major themes to address in a future adaptation of an interactive chatbot. An advisory board was convened to review the data and contribute to content generation that would be most useful for perinatal mothers who are in treatment for OUD. Results: Patients and providers agreed that an interactive chatbot would be useful for patients seeking information on childbirth recovery, substance use recovery, pain management, and infant care, including feeding. Patients specifically wanted a chatbot that felt personal and addressed their individual and specific needs. Providers indicated that the chatbot should address topics that came up frequently in after-hours phone calls to reduce provider burden. Conclusions: Conducting an in-depth participatory-driven needs assessment will support the refinement of an existing chatbot to better meet the needs of a high-risk, difficult-to-reach population. As we move forward to test the chatbot for acceptability and feasibility, data from the needs assessment will support ongoing iterative enhancement of the content, delivery, and safety of the chatbot.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/4711641658d6f7798d35a50de0518027" />
Altered attention network dynamics in resting-state fMRI of individuals with high smartphone addiction scores
IntroductionAltered attention network activity has been consistently reported in recent neuroimaging literature of smartphone addiction and problematic smartphone use. However, many functional imaging studies have relied on traditional static functional connectivity analyses, which cannot capture time-varying or recurrent dynamics of attention networks.MethodsGiven the importance of temporal and spatial dynamics for attentional processing, we applied two complementary dynamic functional connectivity approaches: functional connectivity-based attractor networks (fcANN), which characterize transitions between discrete attractor states related to internal-external context and action-perception dynamics, and cPCA-based quasi-periodic pattern (QPP) analysis, which captures phase relationships between cortical networks including the default mode network, dorsal attention network, and frontoparietal control network.ResultsIn a resting-state fMRI dataset comparing participants with High and Low SAS-M scores (SAS-M: Smartphone addiction scale-Malay version questionnaire), we observed context-dependent differences in DMN-attention-network relationships across attractor states. The most consistent effects emerged during transitions between internal and external attractor states, where subjects with high smartphone addiction (SPA) scores showed greater DMN dominance relative to attention-related networks. For DMN-FPN, these differences correlated with subject SAS-M scores.DiscussionOverall, these findings provide preliminary evidence that smartphone addiction may be associated with altered large-scale attentional dynamics at rest, warranting further investigation in larger and more diverse samples.