castleofargh
Sound Science Forum Moderator
- Joined
- Jul 2, 2011
- Posts
- 10,450
- Likes
- 6,070
We need pattern recognition to learn stuff, be functional, and probably just to think. But our memory is a fluid mess of inaccuracy. So we need something to explain how we can still function with both a flawed memory and the need to identify stuff even when they're somehow different to the original, or to what we remember of the original(which is clearly needed for survival).
In some ironic evolutionary move, our "make do" system turned out to be a great tool. Our heuristic approach is allowing us to identify things rapidly and react just as fast instead of making an exhaustive checklist before validating anything(so of course we're going to mess up more often than we want or think). Whatever alternative options we had, it makes complete sense to me that this one would prosper.
The process hypothesized in the paper allows to get around memory flaws(up to a point) and even to actually one up the heuristic. Identify some cues we consider enough to recognize some event, and actively dismiss the extra stuff related to that specific type of event pattern. Because once we have enough we should probably use the blob CPU for something else instead of working on cues considered confusing or just plain superfluous. Being able to close the door around some already identified stuff is a really effective method for that. Lucky us.
Of course as that system is organized mostly from habit(because that's how we get to find our useful patterns), our experience and expectations end up having even more impact on our way of perceiving the world. and that has always been obvious from a behavioral point of view.
The same talent that lets us see animals in the clouds, allows easy communication, allows to recognize someone with a new haircut, to the point of maybe not noticing the haircut at all . It makes us pass the "I'm not a robot" question online so easily. And the entire thing could just be a consequence of the brain deciding that it won't do anything beyond the bare minimum. I love the idea and implications.
Now in term of better getting to know how we work, I again don't see a big revelation from the paper in term of behavior. Just consider our ability to focus on something, and you have all the elements and implications you need. If we couldn't make abstraction, if we couldn't use patterns as a sort of pre-filter both to identify and to dismiss stuff. How could we decide to only focus on one thing and actually do it? how could we attenuate the perceived reverb in a room after being in it for a while?
I'm listening to music right now, but when I was typing, it was like my brain had turned down the gain. one second before I stopped and the music "came back", I couldn't have told you what track was playing. so ignoring stuff because we want to, yes we can!
The joyful short song that I was sadly missing a part of while typing:
In some ironic evolutionary move, our "make do" system turned out to be a great tool. Our heuristic approach is allowing us to identify things rapidly and react just as fast instead of making an exhaustive checklist before validating anything(so of course we're going to mess up more often than we want or think). Whatever alternative options we had, it makes complete sense to me that this one would prosper.
The process hypothesized in the paper allows to get around memory flaws(up to a point) and even to actually one up the heuristic. Identify some cues we consider enough to recognize some event, and actively dismiss the extra stuff related to that specific type of event pattern. Because once we have enough we should probably use the blob CPU for something else instead of working on cues considered confusing or just plain superfluous. Being able to close the door around some already identified stuff is a really effective method for that. Lucky us.
Of course as that system is organized mostly from habit(because that's how we get to find our useful patterns), our experience and expectations end up having even more impact on our way of perceiving the world. and that has always been obvious from a behavioral point of view.
The same talent that lets us see animals in the clouds, allows easy communication, allows to recognize someone with a new haircut, to the point of maybe not noticing the haircut at all . It makes us pass the "I'm not a robot" question online so easily. And the entire thing could just be a consequence of the brain deciding that it won't do anything beyond the bare minimum. I love the idea and implications.
Now in term of better getting to know how we work, I again don't see a big revelation from the paper in term of behavior. Just consider our ability to focus on something, and you have all the elements and implications you need. If we couldn't make abstraction, if we couldn't use patterns as a sort of pre-filter both to identify and to dismiss stuff. How could we decide to only focus on one thing and actually do it? how could we attenuate the perceived reverb in a room after being in it for a while?
I'm listening to music right now, but when I was typing, it was like my brain had turned down the gain. one second before I stopped and the music "came back", I couldn't have told you what track was playing. so ignoring stuff because we want to, yes we can!
The joyful short song that I was sadly missing a part of while typing: