Interesting that this exhibits learning without the 6 layers, thalamus or hippocampus. That suggests that all these structures are optimisation, and the raw process of learning is intrinsic to neurons themselves. Perhaps that’s obvious since flies can learn flight control (for example) without the structures of the mammalian brain, but very interesting nonetheless imo. Here’s the article :
4 Likes
Neurons in a dish learn to play Pong — what’s next?
https://www.nature.com/articles/d41586-022-03229-y
3 Likes
And another:
https://itwire.com/science-news/biology/brain-cells-in-a-dish-can-play-‘pong’.html
I really do think this is the way to go. Live brains are just too hard.
[BTW this is my home city. 
]
1 Like
Would explain the overnight posts 
1 Like
I honestly hope that mass of neurons isn’t truly sentient, as the idea of creating a living (in the mental sense) being who’s only purpose in life is to play pong is… disquieting
2 Likes
Has anyone checked the original paper in detail? Specifically, how confident are we that the results are not just random noise + p-hacking? For example, at a glance, I am not super-impressed by these plots:
Video S1 kinda looks real, and they say it’s “representative” (as opposed to cherry-picked), but they don’t say anything specific to pin down what “representative” means, which makes me suspicious that it was in fact cherry-picked.
2 Likes
What is your definition of intelligence?
1 Like