Nature Expands Registered Reports to All Fields: A Quiet Revolution in Scientific Publishing

On May 27, 2026, *Nature* published an editorial announcing a potentially underrated milestone: Registered Reports will be extended to every field in which Nature publishes.

This is not a small-scale pilot of a niche concept. It is a paradigm migration at the very top of the publishing hierarchy.

How Bad Was the Old System?

In 2015, the American Psychological Association estimated that for every paper published, about five completed studies went unpublished—left in drawers because their results were negative or non-significant.

This means the "scientific consensus" the public sees is heavily filtered survivorship bias. Drugs that "don't work," theories that "don't hold up," and interventions with "no difference" are never seen. To get published, researchers either run only "easy-win" studies or keep tweaking data analyses until P<0.05—the notorious "P-hacking."

Nature's Solution

The solution is simple, almost blunt:

1. Stage 1: Researchers submit a study plan—research question, methods, analysis protocol. Reviewers judge whether the question matters and whether the methods are rigorous. 2. Commitment: If approved, *Nature* promises publication—regardless of positive, negative, or null results. 3. Stage 2: Researchers execute the plan and submit results. Reviewers only check that the plan was followed; they no longer judge whether results are "interesting."

This process completely inverts the incentive structure:

• Researchers dare to test bold hypotheses—failed hypotheses can be published.
• Researchers invest in costly, high-risk studies—the publication commitment removes the fear of wasted effort.
• Reviewers shift from "judges of results" to "consultants on design," participating more deeply and adding more value.

A Real Case: Brady's Bluesky Experiment

William Brady, a social psychologist at Northwestern University, wanted to know whether social media algorithms amplify emotional, toxic, politicized, and moralized content—and whether this affects users' perceptions of their social environment.

During the 2024 US presidential election, his team ran an experiment on Bluesky: manipulating the feeds of 2,000 users with three different algorithms for eight weeks.

The cost: hiring a software engineer; operating the project for a full year like running a startup; no guarantee of any finding. Under the traditional model, such a study would be nearly impossible to justify. But because it was a Registered Report, *Nature* committed to publication regardless of outcome. Brady called it a "no-brainer."

Crucially, pre-experiment reviewer feedback improved the design: reviewers suggested using existing Bluesky users rather than recruiting new volunteers, yielding responses from users genuinely invested in their interactions.

The results:

• The hypothesis was confirmed: algorithms designed to boost engagement did amplify moralized, emotional, toxic, and politicized content.
• They also found ways to reduce users' exposure to divisive content without reducing the platform's fun.
• A null result: inflammatory posts did not change user behavior by significantly increasing engagement with such content. Brady said this negative finding would likely never have been published without the Registered Reports format.

Expanding from Behavioral Science to All Fields

Nature began accepting Registered Reports in 2023, but only for confirmatory research in cognitive neuroscience and behavioral/social sciences.

The 2026 expansion means:

• Registered Reports are welcomed from all disciplines—physics to biology, medicine to Earth science.
• Beyond hypothesis testing—large-scale data collection, method comparisons, and other exploratory studies now qualify.

The editorial states directly: "the importance of these questions, the quality of the data collection and the rigour of the planned analyses should not be confined to confirmatory research. Expanding Registered Reports to more disciplines and study types reflects our belief that these criteria can strengthen all forms of inquiry."

In short: research quality should depend on whether the question matters and the methods are rigorous—not on how exciting the results are.

Not a Utopia: The Costs

• Two review rounds, double workload. Traditional papers get one round; Registered Reports require reviewing the proposal (Stage 1) and the results (Stage 2), each possibly with multiple revision cycles.
• The suffocation of locked-in analyses. Analysis plans must be fixed before data collection. Unexpected data patterns—such as Brady's team finding far fewer "toxic posters" than expected—can make planned methods a poor fit. Nature's new guidelines allow exploratory analyses as a remedy under strict conditions: clearly labeled as exploratory, well justified, reported separately from main results, and not used as the basis of conclusions.
• Slower publication. The timeline is much longer than traditional papers—a real cost for tenure-track early-career researchers who need quick publications.

The Deeper Signal: Redefining "Good Research"

The core issue is not a publishing format—it's a shift in how the scientific community defines good research.

Traditional implied values: significant results = good research; novel and counterintuitive = high impact; negative results = failure.

Registered Reports promote: important questions = good research; rigorous methods = credible; all results = increments of knowledge.

If this succeeds, it will change researchers' daily behavior: replication studies—once academic suicide—become publishable; hypotheses face stricter testing because falsification can be published; transparency improves because protocols are public and locked.

What Chinese Researchers Should Watch

• Submission strategy: Registering a Report with *Nature* means accepting longer review cycles in exchange for a publication commitment—suitable for high-cost, high-risk experiments.
• Grant applications: The publication commitment can serve as evidence of feasibility in funding proposals.
• Training: Graduate students need to learn pre-registration methodology—writing analysis plans and handling unexpected data.
• Evaluation systems: If Registered Reports go mainstream, should tenure assessments value high-quality negative results over merely "significant" ones?

References

• Nature Editorial: *Nature is expanding Registered Reports to all the fields in which we publish* (2026-05-27) — https://www.nature.com/articles/d41586-026-01629-y
• Brady et al., *Nature*: *Redesigning algorithms to intervene on social norm misperceptions during a national election* (2026) — https://doi.org/10.1038/s41586-026-10536-1
• Gorman & Hubbard (2025): Quantitative Science Studies, 6, 611–622