A new systematic review of 42 studies and around 47,000 young people links short-video platform design to inattention, anxiety and weaker self-regulation in under-25s. Published in the journal European Child & Adolescent Psychiatry and led by researchers at the University of Bayreuth, the work focuses on the apps’ algorithmic curation, infinite scroll and rapid pace as the exposure variables. First author Marlene Ebster said the team set out to move past blanket calls for “less screen time” and pinpoint the specific mechanics behind the documented harms.
The Study at a Glance
The paper, titled “Taming the endless scroll? Short-form videos, digital routines and neurocognitive outcomes in youth,” was published with DOI 10.1007/s00787-026-03083-7 and is available at the published systematic review on short-video platform design. Ebster, a graduate of the Master’s programme in Health Economics at the University of Bayreuth’s Institute of Medical Management and Health Sciences, is the first author, with Michael Lauerer, Eckhard Nagel and Sebastian Schmidt as co-authors. The University of Bayreuth’s official press release on the short-video review findings describes the work as one of the first to focus specifically on platform design in this age group.
The review followed PRISMA and Cochrane guidelines, with two reviewers searching PubMed, PsycINFO, Web of Science and Cochrane Reviews in April 2025 for studies published between January 2015 and April 2025 in English or German. Of around 1,500 datasets initially screened, the final synthesis covered an average participant age of 16.8 years.
The paper grew out of Ebster’s master’s thesis, which passed peer review and was published in a leading journal. Her supervisor Sebastian Schmidt, a research associate at the IMG responsible for the Prevention and Health Promotion division, called it “remarkable and particularly gratifying when a scientific paper emerges from a master’s thesis that withstands the stringent requirements of a peer-review process and ultimately makes a substantial contribution to the advancement of knowledge and becomes accessible to a broad audience.”
How the Review Defined Harmful Use
Before mapping outcomes, the authors had to define exposure. They settled on two working definitions the rest of the analysis hangs on.
- Intensive use: four or more hours per day.
- Unstructured use: continuous scrolling without a fixed purpose, particularly when it pushes back bedtime or study time.
- Structured use (the contrast case): watching during a commute or as a shared social activity, where the session has a natural endpoint.
These thresholds matter because the review deliberately separates use patterns from design mechanics. Two adolescents can scroll for the same number of minutes and end up in different exposure categories depending on whether they set out to look for something specific, were nudged there by a recommendation, or were simply killing time before sleep. Structured use, the authors note, can co-exist with non-trivial daily use without carrying the same associations with negative outcomes.
What the Review Linked Short Videos To
Across the included studies, intensive and unstructured use showed consistent statistical associations with several negative cognitive and emotional outcomes. The effects were strongest for attention and self-control, with weaker but consistent signals for mood.
- Effect range for inattention and impulsivity: mild to moderate
- Cognitive signals: reduced working memory capacity, weaker self-regulation
- Mood signals: higher anxiety, depression and stress
- Imaging signals: changes in grey matter, altered neural signal synchronisation
- Evidence rating: all included studies rated “low” certainty under GRADE
The cognitive findings cluster around attention and impulse control. Increased inattention and impulsivity appeared within a mild to moderate range. Reduced working memory capacity and weaker self-regulation also showed up across multiple studies. The emotional findings tracked higher anxiety, depression and stress in heavy, unstructured users.
The authors repeatedly flag that the underlying studies are observational and cannot, on their own, prove that short-video use causes these outcomes. The review applies the GRADE system, an internationally recognised framework used in evidence-based medicine, and all included studies were rated “low” certainty. None met the criteria for upgrading to moderate or high certainty.
The Three Design Mechanics the Study Isolates
What sets this review apart from earlier screen-time syntheses is its effort to isolate the specific features of short-video apps doing the work. The authors break the user interface into three observable mechanics and analyse each one on its own terms.
- Rapidity: the fast pace at which videos succeed one another, linked in prior research to impaired prospective memory and constant context-switching.
- Infinite scroll: the removal of natural stopping cues, a design choice that lets sessions stretch indefinitely.
- Personalised algorithmic curation: the sequencing of content for the user based on prior behaviour, tied to flow states, time distortion and problematic short-video use.
The model is heuristic rather than causal. The authors describe their framework as non-causal and use Engel’s biopsychosocial model as a scaffold, treating the three mechanics as exposures that may shape outcomes through interacting biological, psychological and social pathways. The point is to give clinicians and designers a vocabulary for the parts of the experience that are, in principle, changeable.
What Imaging Studies Suggest
A subset of the included research used EEG and MRI to look directly at the brain. These imaging studies suggest possible biological correlations, including changes in grey matter and altered neural signal synchronisation, alongside the behavioural findings.
The signals are early, the samples tend to be small, and the authors underline that neuroimaging findings in this area remain correlational. That caveat sits at the centre of the paper.
The mechanisms are biologically plausible because attention control, reward prediction and self-regulation are still maturing through adolescence and into the early twenties. But plausibility is not proof, and the review repeatedly flags the need for longitudinal and experimental designs that can move the evidence from association toward causation.
All included studies were observational and graded “low” certainty under GRADE, with no study meeting the criteria for upgrading. That collective rating is itself a finding: the field has many signals pointing the same way but lacks the trial designs that could tell clinicians and policymakers exactly which design changes would help most.
From “Screen Time” to “Screen Design”
The Bayreuth review lands inside a debate that has been quietly shifting for several years. Public health guidance, including the American Academy of Pediatrics’ guidance on helping kids thrive in a digital world, has moved away from clock-watching toward what the AAP calls the “digital ecosystem,” with attention to engagement-based design, autoplay, infinite scrolling and targeted ads.
The shift mirrors the Bayreuth framework. Earlier syntheses aggregated screen time across heterogeneous platforms and treated the exposure as essentially one thing. The new review argues that delivery mechanics, not duration alone, are the variable that distinguishes short-video apps from television, gaming or older social networks.
For the under-25 cohort the review focuses on, the demographic stakes are large. In 2023, more than a quarter of TikTok’s users were aged 13-17, and, despite age restrictions, millions of U.S. children under 13 accessed these apps in 2022. The platform design reaches users whose attention control and self-regulation are still developing, which is the developmental window in which the underlying biological mechanisms appear most responsive to highly salient digital environments.
Protective Factors and Calls for Regulation
The review is not a verdict against short-video apps. It identifies protective mechanisms that can offset the documented risks, including a supportive social environment, clear digital routines and well-developed media literacy.
It is particularly important not to leave young people alone in navigating digital environments, but to empower them to engage with them consciously.
That quote comes from first author Marlene Ebster in the University of Bayreuth release. The authors also call for regulatory attention, noting that “already today, it is necessary to develop framework conditions for short-video platforms that particularly protect vulnerable groups.” Practical recommendations include fostering structured digital routines, improving understanding of algorithmic systems and strengthening digital literacy, with related evidence on attention gains from blocking smartphone internet covered in the phone friction trial that improved attention in two weeks. The audience the authors name is broad: parents, educators, healthcare professionals, therapists, app developers and policymakers, the group with the leverage to change the design rather than the user.
Frequently Asked Questions
What did the 2026 University of Bayreuth study find about short videos?
The systematic review, published in European Child & Adolescent Psychiatry, linked intensive and unstructured short-video use to increased inattention and impulsivity, reduced working memory capacity, weaker self-regulation and higher anxiety, depression and stress in under-25s. The included studies were observational and rated “low” certainty under GRADE, so the findings describe statistical associations rather than proven causes.
How much short-video use counts as “intensive”?
The review defines intensive use as four or more hours per day. Unstructured use, a separate category, refers to scrolling without a fixed purpose, particularly when it delays bedtime or study time.
Are the brain changes from short videos proven to be causal?
No. The authors state that the neuroimaging findings, including changes in grey matter and altered neural signal synchronisation, remain correlational and require further validation through longitudinal and experimental studies.
Does this mean parents should ban TikTok and similar apps?
The review does not call for bans. It identifies structured digital routines, supportive social environments and media literacy as protective factors that can offset risks, and recommends that framework conditions be developed to protect vulnerable user groups rather than that individual use be eliminated.
What is different about this study compared with earlier screen-time research?
Earlier syntheses typically aggregated screen time across heterogeneous platforms. This review separates platform design into three mechanics, personalised algorithmic curation, infinite scroll and rapidity, and treats the user interface itself as the exposure of interest rather than the hours logged.
What age range does the review cover?
The included studies range from toddlers to young adults up to 25 years. The average participant age across the included cohort was 16.8 years, with the upper age limit treated as a pragmatic developmental boundary rather than a fixed maturational endpoint.
Disclaimer: This article reports on a peer-reviewed systematic review and is for informational purposes only. The findings describe statistical associations, not established causes, and digital media use in adolescents and young adults can affect health in ways that vary by individual. Consult a qualified health professional for personal guidance. Figures are accurate as of publication.
