There's a movement in neuroscience suggesting we should be pursuing bigger bets with larger teams. I think there's a case for doing a bit of this, but I think it's a bad idea to prioritise it for two reasons, and a good case for saying we should be moving in the exact opposite direction. 🧠🧪
Comments
Neurons whisper ancient truths—
Maps of minds take shape.
(dad helped me write this as a team)
The weather be what it may.
Where "weather" means cash.
Japanese poem
Lines of five, seven, and five.
Evoking ~N~nature.
Natural Wonder evoked.
Acquiesce haiku.
Hence, I cannot compute, dude.
Five seven five, jive?
https://youtu.be/doqeXBoFiGU?si=MQimk_uZysvncgUB
But NeuroAI? No way. There is no universal consensus what that even means; much less a single top-down effort.
That is the status quo and it's questionable how well it's worked. Moving towards bigger team science would still leave plenty of room.
But the fact that we don't fund more exploratory research seems like a slightly different problem that should also be fixed independently.
Big science and distributed exploration should be complementary.
The "space" would be something like the space of all possible hypotheses / experiments / results.
Then the question is how to navigate it?
And, here would roughly equate to something like:
- Many small steps = experiments within our current paradigm.
- The occasional big jump = a paradigm shift.
Though, whether these map neatly to small vs large-scale projects doesn't seem obvious?
Then once a big, seemingly solid foraging spot/cluster emerges, *then* big team effort to put it to the test, whereas smaller teams are free to search for new spots?😃
Funding is easiest to get for specific projects and locations, and we are continuously building our support network.
There's also a donation page and merch if you want to support @manybabies.org
For all this to happen we would also need a culture shift where new ideas are more acceptable. No point in dishing out money to lots of eg clever junior people if they all (either strategically or implicitly) follow the leads of the existing big dogs... as often case today
Until then, we need to do a better job of exploring
But in methods, good practice, & learning from painful experience, large groups feel like the only way to lift everyone's game.
(some fields are so full of junk data, it will take real team effort to make progress)