CONP/MRM Panel on reproducibility

The CONP continued its monthly webinar series with an online panel discussion on reproducibility, openness and intellectual property (IP). The discussion centred around the 2019 Magnetic Resonance in Medicine editorial written by Nikola Stikov, Joshua Trzasko and Matt Bernstein and Dylan Roskams-Edris’s Twitter thread commentary. Four panelists took part in the webinar: Peter Jezzard (MRM editor-in-chief, Professor at Oxford University), Nikola Stikov (Co-Chair of CONP Communications Committee, Professor at École Polytechnique, University of Montreal), Dylan Roskams-Edris (Open Science Alliance Officer, Tanenbaum Open Science Institute) and Elizabeth DuPre (PhD candidate, Montreal Neurological Institute, McGill University). More than 90 attendees joined in to listen and ask questions, including Joshua Trzasko and Matt Bernstein (editorial co-authors).

The webinar has been recorded and can be accessed at the following link:

Nikola Stikov summarized the editorial’s main arguments in favour of moving away from papers (and PDFs) toward sharing data and code to promote reproducibility and transparency in the research process. Dylan Roskams-Edris was in general agreement with the editorial’s points but added that the pace of pursuing intellectual property (IP) has not kept up with the pace of scientific progress, suggesting that innovative solutions are needed to address this mismatch. Elizabeth DuPre contributed insights from the viewpoint of a software developer and young researcher. She pointed out that open source is often conflated with “free software” but the reality is often more nuanced than that, with many complex issues to be considered, such as community gate-keeping. Peter Jezzard added his insights into the paper review process, including the possibility of sharing parts of the review process post-publication.

A lively discussion and Q&A followed, touching on issues of openness and reproducibility, the need to involve all stakeholders (academic institutions, journals, funding agencies, researchers, vendors, etc.), creating a community that can cause a cultural shift in thinking, and changing the incentive structure to promote and recognize reproducibility. Examples of current initiatives were shared, such as publishing negative results, curating and documenting shared data, and finding ways to give credit for the behind-the-scenes work that enables reproducibility.

The final consensus was that the community needs to work collectively to encourage and support open science and reproducibility. Dylan Roskams-Edris urged us to be active in building this community, rather than waiting for it to come into existence. Peter Jezzard asked for help from younger researchers in steering the ‘dinosaurs’ in the right direction. Elizabeth DuPre pointed out that different fields (such as astronomy and neuroscience) could learn from one another on best practices and practices to avoid.

Read more from our partner Magnetic Resonance in Medicine (MRM), whose Highlights initiative is run by Dr. Mathieu Boudreau, a CONP developer: