Subscribe to our weekly newsletter

π Science and technology π Health and biotech π Neuroscience

Behind the scenes of the scientific method: when research corrects itself

3 episodes

Épisode 1/3

The reproducibility crisis: is science deceiving us?

On June 23rd, 2026

6 min reading time

Florian Naudet

Psychiatrist, Professor of Clinical Medicine at Université de Rennes, and Senior Member of the Institut Universitaire de France

Thomas Rhys Evans

Psychologist and Lecturer in Occupational Psychology and Open Research at the University of Greenwich

Larry Vernon Hedges

Professor of Statistics and Data Science, Education and Social Policy, Psychology, and Medical Social Sciences at Northwestern University

In 2005, John Ioannidis demonstrated that the probability of an effect reported in a scientific article being real was significantly reduced under certain conditions such as small sample sizes.
Numerous studies have shown that researchers’ cognitive biases, particularly preconceptions, can influence their results.
Studies aimed at reproducing results (replication) can contribute to confirmation of scientific knowledge.
Open science and transparency initiatives remain limited and coexist with the dominant culture of scientific evaluation rather than replacing it.

Épisode 2/3

On May 18th, 2026

3 min reading time

Alice Dreger

PhD in Philosophy and Former Professor at Northwestern University

With several million scientific articles published every year, cleaning up the scientific literature is like looking for a needle in a haystack.
Focusing solely on high-impact publications is the approach adopted by The Medical Evidence Project, an American initiative funded by a $900,000 start-up grant.
Analysts primarily use statistical methods to identify inconsistencies in the data and conclusions presented in the articles.
GRIM-U is a tool capable of detecting a specific type of mathematical inconsistency in certain statistical tests.

Épisode 3/3

On November 12th, 2025

5 min reading time

Cyril Labbé

Professor at Université Grenoble-Alpes

A fraudulent article is a deliberately erroneous scientific publication that contains one or more breaches of scientific integrity.
Currently, analysis tools are far from covering all cases of fraud, as these are very diverse in nature and each requires specific detectors.
The tools used to conceal falsifications leave specific signatures, such as groups of words with specific frequencies.
These breaches of trust are not always carried out by isolated individuals but may be linked to cooperation within networks of publishers and authors.
One solution to tackling scientific fraud would be to strengthen the evaluation system through more preventive measures, or even to rethink its current indicators.