HBP mediation report and reaction
The HBP mediation report was released today. Overall the report echoes many of the criticisms that we have raised over the last few months. In particular, the report concludes:
(p 17) In conclusion, the [mediation committee] MC largely supports and emphasizes the critique voiced by parts of the scientific community regarding objectives, scientific approach, governance and management practices (see Section 2). Most of the members of the MC agree that the way the HBP was presented to the public and to the scientific community lacked self-reflection, and thus contributed to a loss of credibility of the HBP in the scientific community. … Most of the members of the MC agreed that major changes are necessary to create value for the scientific community as well as for society… These issues could put the entire project at risk and pose a serious obstacle to the success of HBP.
The report contains a set of recommendations and a set of observations. The main recommendations are
1. a new governance structure.
2. the reintegration of cognitive neuroscience in the core project.
3. a refocus of the project on technology development in close concertation with the neuroscience community.
4. better communication with the general public, based on realistic goals.
The changes in the governance have already started. The executive board, composed of Markram, Frackowiak and Meier, has recently resigned.
The observations in the appendix are too numerous to review here but they contain extremely important recommendations which, in our opinion, will determine whether the HBP is a viable project. With regards to the bottom-up large scale simulations of the brain, we note that the mediation committee (MC), which includes several members of the HBP, deem the simulations of the human brain as 'not credible' and recommend to redirect the research for this sub project:
(p 44) Members of the MC agree with many of the criticisms voiced by substantial parts of the neuroscience community and recommend that the research in this part should be redirected. With regard to many aspects of brain function, knowledge is still much too limited to permit a credible bottom-up simulation.
The report also emphasizes several times that the HBP must establish a dialogue with the rest of the neuroscience community if it is to be successful. However, the mechanisms by which such a dialogue will be established remain very much unclear.
On Wednesday, the board of directors of the HBP approved the mediation report . This is obviously an important development and a very positive step. However, at least two outstanding questions remain:
1. Who will check that the recommendations are being properly implemented? We recommend that the European Commission nominates a committee to oversee this process. This committee should be composed of scientists external to the HBP whose names are publicly known. Obviously, the non-HBP scientists who served on the mediation committee would be ideal candidates for this new committee.
2. What will happen to the 'observations' in the report? If the HBP only implements the main recommendations, as far as the science is concerned, the project will simply be back to where it was in May 2014. In our opinion, the project was already deeply flawed at the time, it is therefore essential that the observations be addressed through changes in the project. Not doing so would greatly increase the chances that the project will fail.
Implication for partnering projects
While the publication of report is a step forward, it is too early to tell whether it will achieve actual progress, particularly given the ambiguity that surrounds the 'observations'. The HBP must now rewrite its Framework Partnership Agreement, which is supposed to reflect the recommendations of the mediation report. To our knowledge, this FPA will be due this summer. Until then, it is impossible to assess whether the mediation process was successful. In fact, there are reasons to be concerned.
Interviews from Philippe Gilet since the release of the mediation report, indicate that the HBP does not intend to implement the observations, particularly with respect to the simulations. Gilet confirmed in the newspaper 'Liberation ' that the main goal of the HBP is to simulate the brain, even though the report specifically asked the HBP to refocus on technology development. When asked by the newspaper Le Temps , about the fact that the bottom-up simulations are perceived as a waste of time and resources by most neuroscientists, he responded that this opinion was shared only by a few people, thus ignoring the fact that this is expressed in the Observations section of the mediation report (p44).
Given these facts, it might be wise not to apply for HBP partnering projects until the new FPA is released. The future of the partnering projects is still unclear, as very little money has been committed by national funding organizations.
In summary, the mediation report represents a very positive step in the resolution of the HBP crisis. We are particularly pleased that the mediator chose to publish the entire report along with the name of the authors of the report. It is also important to realize that the report was co-authored by many members of the HBP. The fact that a significant fraction of people within the project share our concerns should considerably facilitate future discussions between the HBP and European neuroscientists. However, we remain deeply concerned by the recent declarations from Philippe Gilet that the HBP is primarily about bottom-up simulations of the brain and the fact that the observations in the report might be ignored.
Thanks for your support,
Alexandre Pouget and Zachary Mainen