Scientific papers should have two Discussion sections – one written by the authors, and the other by an independent researcher.

According to a new paper from Michael S. Avidan, John P. A. Ioannidis and George A. Mashour, this “second discussant” system could help ensure more balanced and objective inference in science.

The authors begin by noting that while the reproducibility crisis has focussed attention on the Methods and Results sections of papers, Discussion sections are not free