Comment by Steve44 | Hacker News Reader

Comment by Steve44 | Hacker News Reader

This clicked with me when I watched The Brain with David Eagleman [1] on BBC a little while ago. I've always know there is a disconnect between reality and what you interpret but the way he described it stuck. Note this is my layman's way of describing how I understand it.

Firstly your eyes are not connected directly to your "brain". You don't see what is in front of you.

Your eyes write a rough scan of elements of what is in front of you to a "frame buffer". Your brain looks at this "frame buffer" and interprets what it thinks you are seeing. At this stage it can fill in blanks and do all sorts of manipulations to make it look smooth and complete.

What can also happen is your brain can write to this "frame buffer", this is indistinguishable from the information the eyes have put there. This is pretty much how psychosis works - those monsters/spiders/people/objects seem as real as if they are there.

[1] https://www.bbc.co.uk/programmes/b06yjrdp

It's also multiple frame buffers. The nerves before the brain already do some interpretation, and the brain an optimized version.

And the brain can rewrite the history it saved if it finds logical errors that came up because of it filling in gaps wrong.

Brains are awesome and scary.

That's the ultimate eventual consistency.

When you realize your previous approximation leads to a discontinuous function after new data arrives, just roll back your instruction pointer as well as your perception of time.

This ability to retroactively error-correct without affecting perceived continuity could also be affording our brains a greater margin of error.

Maybe, but then it turns out that your brain's cache isn't rolled back when you mispredict, and you have to close your eyes for ten seconds after looking at anything secret until another few hundred generations of evolution fix the bug.

It seems our biological neural networks, probably, aren't all that different from the ones we write. In their 'scope' they basically do just enough, for us to function. Beyond that, evolution probably started to optimise other things.

"Just enough" would mean being able to agree upon whether the object being seen was a danger (say a wild animal, an enemy tribe) or not (a friendly tribe). This goes deep enough for us to agree upon facial expressions and the like.

Interestingly, the 'lack of agreement' would theoretically be impossible to detect, because for consensus, we would achieve "eventual consistency" of conclusions - thereby reforming the neural pathways of all individuals.

Reply on news.ycombinator.com