Capturing public attention has become essential for communicating science, but making scientific content engaging can come at a cost. Information designed to attract clicks, shares or views is often simplified, leaving out details that help readers fully understand what a study actually found. As a result, people may walk away with inaccurate impressions even when the information they read is technically correct. The challenge is not necessarily the spread of falsehoods, but the loss of context and nuance that can shape public understanding in misleading ways.

That tension is at the centre of new research led by Marta Serra-Garcia from the University of California, San Diego Rady School of Management. Published in the American Economic Review, the study explored how incentives in today’s online attention economy influence the communication of scientific findings and what readers ultimately learn from them. Serra-Garcia’s research suggests that misinformation does not always stem from deliberate deception. In many cases, incomplete information alone can lead readers to draw inaccurate conclusions.

Importantly, the study found that summaries designed to attract attention were not more likely to contain factual errors. Instead, they tended to provide less information overall, particularly about research methods, sample sizes and study limitations. According to Serra-Garcia, the issue is more complicated than simply dismissing attention-grabbing content as “clickbait.” Engaging headlines and summaries can encourage curiosity and motivate people to learn about science. However, the same techniques that increase engagement can also unintentionally contribute to misunderstandings when essential details are omitted.

One example highlighted in the study involved research showing that a compound in broccoli reduced cancer cell growth in mice. Removing the phrase “in mice” dramatically changes how readers may interpret the finding, making it appear more directly applicable to human health than the evidence supports. Serra-Garcia noted that adding those two words is simple. Yet, even small additions that reduce excitement or certainty may lower the likelihood that people will click on or share the content. The research suggests that these seemingly minor omissions can significantly influence public interpretation of scientific results.

To examine these effects, Serra-Garcia conducted a large, multi-stage experimental study involving freelance writers and thousands of participants. In the first stage, 149 freelance writers produced nearly 600 summaries of real scientific studies on topics including cancer, vaccines, sleep and climate change. Writers received different instructions depending on the experimental condition. Some were asked to prioritise accurate communication, while others were encouraged to write summaries that would attract attention and generate clicks or shares online.

In the second stage, more than 3,700 participants read the summaries under varying conditions, including whether they had the opportunity to click through for additional information. The findings were consistent across experiments. Attention-driven summaries were generally shorter, easier to read and more engaging, but they also contained fewer important details. Readers who relied only on the summaries demonstrated lower levels of understanding, with knowledge scores dropping by approximately six to seven percentage points. They were also more likely to make incorrect assumptions, such as believing findings applied directly to humans or represented firm medical recommendations. Even when participants had access to additional information, most chose not to seek it out, reflecting broader patterns observed on social media where users frequently share articles without reading beyond headlines or summaries.

The same pattern emerged when the writing process involved artificial intelligence rather than humans. When a large language model was instructed to maximise attention and engagement, it also produced summaries with less detail and context. This suggests the problem is driven less by individual writers and more by the incentives built into digital communication systems that reward attention above completeness. For Serra-Garcia, the findings raise an important question for scientists, journalists and institutions alike: how can scientific information remain engaging and accessible without sacrificing the details necessary for readers to understand the full picture accurately?

More information: Marta Serra-Garcia, The Attention-Information Trade-Off, American Economic Review. DOI: 10.1257/aer.20240850

Journal information: American Economic Review Provided by University of California – San Diego