Pooling 71 studies of nearly 100,000 people, researchers found heavier short-video use was linked to poorer attention, weaker impulse control and more stress — a pattern that held across youths and adults

In September 2025, the journal Psychological Bulletin published a large review of the research on short-form video, the endless feeds on TikTok, Instagram Reels and YouTube Shorts.

Its title reads almost like a diagnosis of the phone in your hand: “Feeds, feelings, and focus.” Led by Lan Nguyen at Griffith University, the team combined 71 earlier studies covering 98,299 people to ask a question many of us have quietly asked ourselves: does all this scrolling actually do something to us?

A note before going further. We are writers and editors reading a research paper, not clinicians or psychologists. What follows is reflection on one study, not advice about your own mind or habits. The research it draws on can only show links, not causes, and a pattern across a crowd is not a verdict on any single reader.

The short answer is uncomfortable, and the authors state it plainly. Nguyen and colleagues write that more short-form video use “was associated with poorer cognition” in the pooled data. The same pattern showed up for mental health, in teenagers and adults alike. The key word is “associated,” and it carries the whole argument.

What the numbers actually show

The strongest links were in thinking and attention. The authors report that more short-video use “was associated with poorer cognition (moderate mean effect size, r = −.34), with attention (r = −.38) and inhibitory control (r = −.41) yielding the strongest associations.”

Those numbers are easy to over-read, so a plain translation helps. A correlation runs from 0, meaning no relationship at all, to 1, meaning two things move in perfect lockstep. An r of -.34 sits in the middle: a real link visible in the data, but far from fate. The minus sign means the two things move in opposite directions. More scrolling, lower scores. The strongest link, -.41, was with what researchers call inhibitory control, roughly the mental brake you use to stop yourself acting on an impulse.

The mood side, and why age didn’t rescue anyone

The mental-health links were weaker but pointed the same way. The review found a link of r = -.21 for poorer mental health overall, with stress (r = -.34) and anxiety (r = -.33) showing the strongest links in that group.

What gives the paper its weight is not any single number but its breadth. The pattern held whether the participants were teenagers or adults. There is a comfortable story we tell about screens: that the effects are really a young-brain problem the rest of us have outgrown. This review does not support that story. Anyone who has opened an app to check one thing and surfaced twenty minutes later, unsure where the time went, has felt the pull the numbers are trying to describe, and it does not appear to switch off with age.

Not everything moved together, and the authors are careful about it. They note that “interestingly, SFV use was not associated with body image or self-esteem, which may reflect the diverse content and creators featured on these platforms.” That hedge belongs there. Finding no link in one review is not proof there is no link, only that this pooled data did not show one.

The problem hiding in every number

This is the limit that shapes everything above. Most of the studies took a single snapshot of people rather than following them over time. A snapshot can show that heavy scrollers tend to score lower on attention, but not which came first.

That matters more than it sounds. The obvious reading is that scrolling wears down focus. The reverse fits the same data just as well: people who already struggle to focus, or who already feel low or worried, may be the ones who reach for an endless feed in the first place. As the PsyPost coverage of the study notes, existing difficulties could be driving the overuse rather than the other way around.

What the review asks for next

The authors do not pretend to have settled the question. Where the picture is murkiest, they write that “further research is therefore needed to clarify how different types of content exposure may shape these associations.” It is a measured close from a team sitting on a large, headline-friendly dataset. They could have overreached, and did not.

If any of this lands close to home, and worry or low mood has started to feel heavier than usual, a qualified counsellor or therapist is worth talking to. A study like this describes crowds, not the person reading it.