5 ms·
\tangent In the spirit of understanding the mind in terms of the engineering of the day, the cerebellum's striated structure seems something of a DSP, for linea
by hyperpallium 8y ago
\tangent In the spirit of understanding the mind in terms of the engineering of the day, the cerebellum's striated structure seems something of a DSP, for linear processing, especially where timing is involved.
There's a dramatic contrast between the sophistication of optogenetics, and the primitive hypotheses being tested: is this connected to that?
Does there need to be a direct connection, or could other parts of the brain use the cerebellum, when limear/timing is needed, for particular stages of processing, like a math co-processor?
e.g. Is it possible that instead of direct "wired" connections, there's a packet radio or UDP-like system, where neurons (or columns of neurons) aren't directly signalling, but passing on a packet of information to be shunted elsewhere in the brain.