Original title: Quantifying the Engagement Trap: Impact of Short-form Video Recommender Systems on Users with ADHD
Vedad Misirlic, Gregor Mayr, Elisabeth Lex · September 2026
Short-form video apps use recommender systems to keep you watching as long as possible. Yet, little is known about how these algorithms affect people with ADHD compared to people without it. Researchers studied this issue to map out what they call the Engagement Trap. This happens when an algorithm successfully keeps people hooked, but leaves certain users worse off. The researchers compared how both groups experience these personalized video feeds. They discovered that while both groups find the suggested clips relevant, participants with ADHD pay a much higher price. Specifically, viewers with ADHD experience much more time blindness, feel more regret after watching, and suffer greater emotional distress. Finally, the team gathered user feedback on theoretical designs meant to make video platforms safer and fairer.
Imagine you are at a buffet where the plates keep refilling with your favorite foods as soon as you take a bite. You lose track of how much you ate and feel sick afterward. For most people, the buffet is fine, but for some, it triggers a cycle of overeating and regret. In this paper, the video recommendations are like that buffet – they keep serving new clips and for users with ADHD, this makes them lose track of time and feel distressed, even though the recommendations themselves are good.
For the millions of users with ADHD, short-form video platforms like TikTok are not just entertaining – they can be harmful. Our study shows that even when recommendations feel relevant, these users experience more time blindness, regret, and distress after using the app. This calls for design changes that account for neurodiversity, rather than optimizing solely for engagement.
The study used self-reported ADHD status and a sample of 302 participants from Prolific, which may not be representative. The design interventions were only theoretical and tested via feedback, not implemented. The paper does not establish causation between recommendations and harm, only correlation. It also does not propose specific, tested solutions to mitigate the trap.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Quantifying the Engagement Trap: Impact of Short-form Video Recommender Systems on Users with ADHD — Vedad Misirlic, Gregor Mayr, Elisabeth Lex
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.