Dan Goodman
Computational neuroscientist at Imperial College. I like spikes and making science better (Neuromatch, Brian spiking neural network simulator, SNUFA annual workshop on spiking neurons).
🧪 neural-reckoning.org
📷 adobe.ly/3On5B29
- How do neuroscientists define computation? I've seen it used to mean specifically Turing-equivalent computation, or discrete computation. Are these are common interpretations? I see how in physical reality dynamical systems might be at odds w/ Turing-equivalent computation (Moore's conjecture).
- I don't think neuroscientists have a common definition of computation and this might be a good thing. For me, such concepts are abstractions and can be useful in constructing models, but inevitably break down somewhere in application and so identifying the model with reality creates confusion.
- Better in this case to have multiple abstractions with different domains of application and a clear understanding that they are different. The puzzles/confusion only arises when we take the link between the abstraction and reality to be too tight.
