Scientific papers are designed to make discovery look orderly. A question appears, methods follow, results arrive, and a conclusion closes the argument. Real research is rarely so clean. Experiments fail, promising ideas collapse, conversations redirect entire projects, and crucial decisions often survive only in the memories of the scientists who made them. A new format called the “inverse seminar” is designed to recover that hidden history by turning the traditional research talk upside down.
Writing in Nature Physics, a team led by University of Pittsburgh physicists Jeremy Levy and Chandralekha Singh describes how the format allows junior researchers to present a senior scientist’s published work back to the scientist who produced it. The senior researcher is invited without being asked to prepare a lecture. Instead, they listen as someone else reconstructs the paper from the published record, then respond to questions and interpretations in real time. The result is part seminar, part scientific investigation and part oral history.
The idea emerged from the transformation of academic life during the COVID-19 pandemic. When seminars moved online in 2020, many talks became passive experiences: cameras went dark, audiences became grids of names, and participants could quietly answer email while a speaker continued in the background. Remote communication did not create disengagement, the researchers argue, but it removed much of the social pressure that had previously kept audiences attentive. The inverse seminar attacks that problem by making the speaker’s own scientific memory central to the event.
In a conventional seminar, a researcher presents familiar material, often using a carefully rehearsed narrative developed through dozens of previous talks. In an inverse seminar, the senior scientist does not control the initial story. A junior scientist must study the original paper closely, explain its experiments and theoretical logic, and identify points that may be unclear or incomplete. The senior author then reacts to the reconstruction. Because the presentation is not rehearsed by the person who carried out the work, it can trigger memories of uncertainty, disagreement and unexpected discoveries that rarely appear in the final publication.
That distinction is scientifically important. Published papers usually describe the successful path from hypothesis to evidence, but the route actually taken may include discarded experiments, failed measurements and interpretations that changed several times. These abandoned paths contain information about experimental design, troubleshooting and scientific judgment. They can reveal why one measurement was trusted over another, how an anomaly was recognized as meaningful, or why a particular collaborator became essential. Without documentation, such knowledge can disappear when researchers retire or move away from active work.
One example involved Michael Crommie of the University of California, Berkeley, whose work helped establish important directions in graphene research. His group published a 2010 Science paper on how graphene behaves under strain, presenting the result as a logical sequence of observations and explanations. During the inverse seminar, however, Crommie said the discovery happened “totally by accident.” His team had been studying graphene for unrelated reasons when it encountered an observation nobody could initially explain. The mystery continued for months until theorist Antonio Castro Neto happened to visit Crommie’s office and recognized the significance of what the group was seeing.
That detail changes how the discovery is understood. Castro Neto later became a coauthor of the paper, but the publication does not explain how his involvement began or how long the researchers struggled before identifying the underlying phenomenon. Crommie also described two other projects that each lasted approximately two years and failed to produce publishable results. Those efforts are absent from the scientific literature, even though they may contain lessons about what did not work and why. The inverse seminar created a record of those otherwise invisible experiments.
The format is not limited to physics. Any discipline with a published literature can use it, provided that a junior researcher is willing to analyze the work in detail and a senior scientist is prepared to listen without knowing exactly what will be presented. The Pittsburgh-led team conducted seven inverse seminars over 27 months and reported a consistent pattern: senior scientists were generally generous, and the reconstructed presentation encouraged them to supply missing context rather than defend their work. Hearing one’s research described by someone who had studied it from the outside appeared to make the conversation more open, not more adversarial.
The researchers have now launched inverseminars.org, where sessions can be viewed and cited. Each event receives its own DOI, giving a recorded conversation a formal scholarly identity similar to that of a paper. The Crommie session lasts 92 minutes and includes discussion of the graphene research as well as the decisions and accidents that shaped it. Levy says that small changes to an ineffective seminar format produce only slightly different versions of the same problem; reversing the roles of presenter and subject creates a more fundamental change. The approach is deliberately uncomfortable, but that discomfort may be precisely what allows science to recover the stories that polished papers leave behind.
Subject of Research: Not applicable
Article Title: What Physics Papers Don’t Tell You
News Publication Date: 31-Jul-2026
Web References: https://inverseminars.org/; https://www.nature.com/articles/s41567-026-03380-7
References: Nature Physics, DOI: 10.1038/s41567-026-03380-7
Image Credits: Michael Crommie and the authors
Keywords: Inverse seminar, scientific communication, academic publishing, research history, physics, graphene, science education, scientific collaboration, oral history, research methodology
