Scientific institutions and mechanisms

1 Institution design

• Accelerating science through evolvable institutions

  The biggest red flag is with our institutions of science. Institutions affect all the other factors, especially the management of money and talent. And today, many in the metascience community have concerns about our institutions. Common criticisms include:

  • Speed. It can easily take 12–18 months to get a grant (if you’re lucky)
  • Overhead. Researchers typically spend 30–50% of their time on grants
  • Patience. Researchers feel they need to show results regularly and can’t pursue a path that might take many years to get to an outcome
  • Risk tolerance. Grant funding favors conservative, incremental proposals rather than bold, “high-risk, high-reward” programs (despite efforts to the contrary)
  • Consensus. A field can converge on a hypothesis and prune alternate branches of study too quickly
  • Researcher age. The trend over time is for grant money to go to older, more established researchers
  • Freedom. Scientists lack the freedom to direct their research fully autonomously; grant funding has too many strings attached

  Now, as a former tech founder, I can’t help but notice that most of these problems seem much alleviated in the world of for-profit VC funding. Raising VC money is relatively quick (typically a round comes together in a few months rather than a year or more). As a founder/CEO, I spent about 10–15% of my time fundraising, not 30–50%. VCs make bold bets, actively seek contrarian positions, and back young upstarts. They mostly give founders autonomy, perhaps taking a board seat for governance, and only firing the CEO for very bad performance. (The only concern listed above that startup founders might also complain about is patience: if your money runs out, you’d better have progress to show for it, or you’re going to have a bad time raising the next round.)

  I don’t think the VC world does better on these points because VCs are smarter, wiser, or better people than science funders—they’re not. Rather, VCs:

  • Compete for deals (and really don’t want to miss good deals)
  • Succeed or fail in the long run based on the performance of their portfolio
  • See those outcomes within a matter of ~5–10 years

  In short, VCs are subject to evolutionary pressure. They can’t get stuck in obviously bad equilibria because if they do they will get out-competed and lose market power.

• Focused Research Organizations: A New Model for Scientific Research

• Focused Research Organizations to Accelerate Science, Technology, and Medicine

• A Guide to DeSci, the Latest Web3 Movement

• The Overedge Catalog: The Future of Research Organizations

  Research organizations and institutions often are shoehorned into a set of well-established categories: universities, public companies, tech startups, and certain types of non-profits, such as think tanks. But there is the need for innovation here, particularly when it comes to encouraging the development of new ideas and the ability to operate on long timescales. We need new types of research organizations.

  In cartography, most maps are bound by the straight lines at their borders. But occasionally, there are parts of the map that don’t quite fit. They bleed over the edge and yet still cry out for being included in a map. These are the overedges. The Overedge Catalog is devoted to collecting the intriguing new types of organizations and institutions that lie at the intersection of the worlds of research and academia, non-profits, and tech startups

• Labs · Patrick Collison

• Organizations with structures worth learning from

2 Contrarianism and diversity

• Maxwell Tabarrok, Overfitting In Academia:

  But our selection criteria do not match with what we want academics to do. The out-of-sample prediction ability of academics is worse than random and laymen. They know lots about research they’ve already been tested on, but are rarely better than random at guessing the results of experiments in their field that they haven’t seen before. Citations and academic rank do not increase the ability of economists to predict the outcome of economic experiments or political events.

Hanania on Tetlock and the Taliban makes a point about the illusory nature of some expertise.

[Tetlock’s results] show that “expertise” as we understand it is largely fake. Should you listen to epidemiologists or economists when it comes to COVID-19? Conventional wisdom says “trust the experts.” The lesson of Tetlock (and the Afghanistan War), is that while you certainly shouldn’t be getting all your information from your uncle’s Facebook Wall, there is no reason to start with a strong prior that people with medical degrees know more than any intelligent person who honestly looks at the available data.

His examples about science community are neat. Then he draws a longer bow and makes some IMO less considered swipes at a weak-manned diversity argument which somewhat ruins the effect for me. Zeynep Tufekci gets at the problem that I think both people who talk about contrarianism and diversity would like to get at: Do the incentives, and especially the incentives in social structures, actually encourage the researchers towards truths, or towards collective fictions?

Sometimes, going against consensus is conflated with contrarianism. Contrarianism is juvenile, and misleads people. It’s not a good habit.

The opposite of contrarianism isn’t accepting elite consensus or being gullible.

Groupthink, especially when big interests are involved, is common. The job is to resist groupthink with facts, logic, work and a sense of duty to the public. History rewards that, not contrarianism.

To get the right lessons from why we fail—be it masks or airborne transmission or failing to regulate tech when we could or Iraq war—it’s key to study how that groupthink occurred. It’s a sociological process: vested interests arguing themselves into positions that benefit them.

Scott Alexander, Contrarians, Crackpots, and Consensus tries to break this idea open with an ontology.

I think a lot of things are getting obscured by the term “scientific establishment” or “scientific consensus”. Imagine a pyramid with the following levels from top to bottom:

FIRST, specialist researchers in a field…

SECOND, non-specialist researchers in a broader field…

THIRD, the organs and administrators of a field who help set guidelines…

FOURTH, science journalism, meaning everyone from the science reporters at the New York Times to the guys writing books with titles like The Antidepressant Wars to random bloggers…

ALSO FOURTH IN A DIFFERENT COLUMN OF THE PYRAMID BECAUSE THIS IS A HYBRID GREEK PYRAMID THAT HAS COLUMNS, “fieldworkers”, aka the professionals we charge with putting the research into practice. … FIFTH, the general public.

A lot of these issues make a lot more sense in terms of different theories going on at the same time on different levels of the pyramid. I get the impression that in the 1990s, the specialist researchers, the non-specialist researchers, and the organs and administrators were all pretty responsible about saying that the serotonin theory was just a theory and only represented one facet of the multifaceted disease of depression. Science journalists and prescribing psychiatrists were less responsible about this, and so the general public may well have ended up with an inaccurate picture.

Bright (2023):

Du Bois took quite the opposite route from trying to introduce lotteries, with their embrace of chance randomization. In fact, to a very considerable degree he centrally planned the sort of research his group would carry out so as to form an interlinking whole. Where the status quo system allows for competition between scientists to give funding out piecemeal to whoever seems best at a given moment, Du Bois’ work embodies the attitude that as far as possible our research activities should be coordinated, and not aimed at rewarding individual greatness but rather producing the best overall project. While ideas along these lines have not been totally without support in the history of philosophy of science (see e.g. Neurath 1946, Bernal 1949, Kummerfeld & Zollman 2015), it is safe to say the epistemic merits of this are relatively under-explored. Our brief examination of Du Bois’ plan will thus hopefully form a spur to generate more consideration of this sort of holistic line of action

3 Incoming

Smaldino and O’Connor (2020):

Why do bad methods persist in some academic disciplines, even when they have been clearly rejected in others? What factors allow good methodological advances to spread across disciplines? In this paper, we investigate some key features determining the success and failure of methodological spread between the sciences. We introduce a formal model that considers factors like methodological competence and reviewer bias towards one’s own methods. We show how self-preferential biases can protect poor methodology within scientific communities, and lack of reviewer competence can contribute to failures to adopt better methods. We then use a second model to further argue that input from outside disciplines, especially in the form of peer review and other credit assignment mechanisms, can help break down barriers to methodological improvement. This work therefore presents an underappreciated benefit of interdisciplinarity.

On the origin of psychological research practices, with special regard to self-reported nostril width:

when does a certain practice–e.g., a study design, a way to collect data, a particular statistical approach–”succeed” and start to dominate journals?

It must be capable of surviving a multi-stage selection procedure:

1. Implementation must be sufficiently affordable so that researchers can actually give it a shot
2. Once the authors have added it to a manuscript, it must be retained until submission
3. The resulting manuscript must enter the peer-review process and survive it (without the implementation of the practice getting dropped on the way)
4. The resulting publication needs to attract enough attention post-publication so that readers will feel inspired to implement it themselves, fueling the eternally turning wheel of Samsara publication-oriented science

• The Leisure of the Theory Class, Top 5 Tyranny and its Consequences
• Escaping science’s paradox discusses red-teaming science and incentivising productive failures
• Every Bug is Shallow if One of Your Readers is an Entomologist
• How much should theories guide learning through experiments?
• Jonathan Rauch, The Constitution of Knowledge: A Defense of Truth
• Samir Unni, Molecular Missionaries
• Progress Studies: A Discipline is a Set of Institutional Norms (connection to innovation)
• Scott And Scurvy (Idle Words)
• Stuart Buck, The Good Science Project
• Royal Society, The online information environment
• Samuel Moore, Why open science is primarily a labour issue
• Open call: Consequences of the Scientific Reform Movement · Journal of Trial & Error
• Eric Gilliam, FreakTakes

The next few come from The Overedge Catalog: The Future of Research Organizations.

• We need to create and foster new types of scientific organizations (by Samuel Arbesman)

• A Vision of Metascience: An Engine of Improvement for the Social Processes of Science (by Michael Nielsen and Kanjun Qiu)

• Reading material about applied research labs (by Patrick Collison)

• When should an idea that smells like research be a startup? (by Ben Reinhardt)

• Why DARPA has worked so well (by Ben Reinhardt)

• White paper on how to best structure and fund research organizations (by Alan Kay)

• On funding scientific research (by Alan Kay)

• Notes on Starting a Scientific Organization (by Sabrina Singh)

• Tools, Computing, and Research Labs (by James Hill-Khurana)

• Erika’s quick-start guide to research nonprofits (by Erika Alden DeBenedictis)

• White paper on “fractional scholarship” (by Samuel Arbesman and Jon F. Wilkins)

• The Half-Life of Facts (by Samuel Arbesman)

• Structure for a Private ARPA (by Ben Reinhardt)

• Focused research organizations (by Adam Marblestone and Sam Rodriques) (see also “What is (and is not) an FRO?” by AJ Kourabi)

• Startup nonprofits (by Mark Lutter)

• An orthogonal technology lab (by Matt Webb)

• Research Studios (by Tom Critchlow)

• DIY research studio (by Patrick Tanguay)

• The Need for Long-term Research(by Kevin Kelly)

• Retroactive Public Goods Funding

• Alt VC (by Mark McGranaghan)

• Crowdfunded research (by Andy Matuschak)

• Interesting organizational structures (by Ben Reinhardt)

• A list of research labs (by Matt Webb)

• Gatekeeping is Good and Everyone Already Agrees

• The Balto/Togo theory of scientific development

• Bringing Greater Balance to the Force (the Scientific Workforce, that is)

• Rachel Helps, Wikipedia’s Citations Are Influencing Scholars and Publishers

  In 2016, Wikipedia was the 6th-largest referrer for DOIs, with half of referrals successfully authenticating to access the article. External links on Wikipedia produce an estimated 7 million dollars of revenue per month. Given that Wikipedia is such a popular website, it’s unsurprising that academic publishers are actively pursuing ways to promote their work on Wikipedia.

  Scholarly publishers have reported increased traffic as a result of giving access to their publications to Wikipedia editors, and a controlled experiment on Wikipedia shows that they are right to value Wikipedia citations. Works cited on Wikipedia have an outsized influence on scholarly work—specifically in its literature reviews. Additionally, one research article found that open-access (OA) articles were cited more frequently than non-OA articles on Wikipedia in 2014, an idea supported by the generally increased readership of OA articles compared to paid-access articles

• When Does Science Self-Correct? Lessons from a Replication Crisis in Early 20th Century Chemistry

  What would it take for fields like economics and social psychology to self-correct as chemistry and physics (sometimes) do?

  A crucial feature that leads a field to become self-correcting is the potential for important technological applications. As long as scientists are only writing articles, allegations of error may be difficult to resolve. But once a field has advanced to the point where it can claim to make bread from air, or draw gold from the sea, it more quickly becomes clear if the claim can hold water.

• Introducing Massively Open Online Papers (MOOPs)

4 References