Psychological Bulletin · 2025

Feeds, feelings, and focus.

The paper that keeps circulating on X is a systematic review and meta-analysis: 71 studies, 98,299 people, short-form video use mapped onto cognition and mental health. Here is what it actually found — and what it cannot say.

Feeds, feelings, and focus: A systematic review and meta-analysis examining the cognitive and mental health correlates of short-form video use

Lan Nguyen, Jared Walters, Siddharth Paul, Shay Monreal Ijurco, Georgia E. Rainey, Nupur Parekh, Gabriel Blair, and Miranda Darrah

DOI 10.1037/bul0000498

The headline numbers

Higher short-form video engagement was associated with poorer cognition (moderate, r = −.34) and poorer mental health (small, r = −.21). Attention (r = −.38) and inhibitory control (r = −.41) were the strongest cognitive links. Stress (r = −.34) and anxiety (r = −.33) were the strongest mental-health links.[Source]

Who was in the data

Youth and adults. Several platforms, not only TikTok. A large share of studies came from Asia, where Douyin and similar apps are everyday infrastructure. The authors report the pattern was consistent across age groups and platform types.

How large are the associations?

In this literature, r around .10 is small, .30 moderate, .50 large. Negative means “more use, worse score.” Grey bars did not reach statistical significance in the review.

1. Inhibitory control-0.41
2. Attention-0.38
3. Cognition (overall)-0.34
4. Stress-0.34
5. Anxiety-0.33
6. Mental health · addiction measures-0.32
7. Depression-0.23
8. Loneliness-0.23
9. Sleep-0.22
10. Working memory-0.21
11. Mental health (overall)-0.21
12. Well-being-0.14
13. Memory-0.14
14. Self-esteem (n.s.)-0.08
15. Body image (n.s.)-0.10

Addiction measures beat stopwatch measures

When studies measured “I can’t stop” rather than “how many minutes,” the mental-health association jumped (r = −.32 vs −.10 for duration). Frequency was not significant. That is why this site talks about control, not only screen time.

What they did not find

No reliable association with body image or self-esteem. Relatively few studies tested memory, language or reasoning — so those domains are still thin. Reasoning was not significant in the one available estimate.

The mechanism they sketch

Fast, highly stimulating clips may habituate the mind to novelty and make slower work — reading, problem-solving, a conversation — feel underpowered. On the mood side, infinite, algorithm-ranked rewards look a lot like a variable-ratio slot machine: you stay because the next pull might be the good one.

Other evidence on this site

• Neuroimaging correlates of short-video addiction: orbitofrontal structure and executive-control function
• Short-video addiction, prefrontal theta power, and self-control during executive tasks
• Fragmented learning from short videos modulates neural activity and connectivity during memory retrieval
• Oxford Word of the Year 2024: brain rot

Read this twice

• Correlation is not causation. People who are already anxious or distractible may also scroll more.
• Most studies are cross-sectional and self-report.
• Imaging and EEG papers have small samples.
• “Brain rot” is culture. The papers are psychology and neuroscience. We keep them distinct.