- Blind mode tutorial
lichess.org
Donate

Science of Chess: Seeing the board "holistically."

Great blog, great explanations of how we perceive faces and the similarity in chess experts to perceiving faces. Very interesting to read and think about.

NDpatzer: I think that would have to be the possibility that strong players see the board differently than mere patzers like myself

NDPatzer: What does a master subjectively experience that's different from what I see? What (if anything) actually appears different to them?

One of the great philosophical paradoxes. That an identical visual stimulus can 'look different'.

Is what you see 'visual'? It can't be if it can look different. Visual consciousness can't be solely based on the visual cortex.

The difference between novices and experts is 'knowing what to do'.

But what is 'knowing?'. Some say it's information like memory in a computer, others say is just delayed motor responses (there is no distinction between perception and cognition). In this view, Intelligence is only shown by physical action/potential physical action.

What do you think @NDpatzer ?

Great blog, great explanations of how we perceive faces and the similarity in chess experts to perceiving faces. Very interesting to read and think about. >NDpatzer: I think that would have to be the possibility that strong players see the board differently than mere patzers like myself >NDPatzer: What does a master subjectively experience that's different from what I see? What (if anything) actually appears different to them? One of the great philosophical paradoxes. That an identical visual stimulus can 'look different'. Is what you see 'visual'? It can't be if it can look different. Visual consciousness can't be solely based on the visual cortex. The difference between novices and experts is 'knowing what to do'. But what is 'knowing?'. Some say it's information like memory in a computer, others say is just delayed motor responses (there is no distinction between perception and cognition). In this view, Intelligence is only shown by physical action/potential physical action. What do you think @NDpatzer ?

@RuyLopez1000 said in #2:

Great blog, great explanations of how we perceive faces and the similarity in chess experts to perceiving faces. Very interesting to read and think about.

NDpatzer: I think that would have to be the possibility that strong players see the board differently than mere patzers like myself

NDPatzer: What does a master subjectively experience that's different from what I see? What (if anything) actually appears different to them?

One of the great philosophical paradoxes. That an identical visual stimulus can 'look different'.

Thanks for reading! A few thoughts below:

Is what you see 'visual'? It can't be if it can look different.

I think I disagree with this, but it depends on what you mean. The same physical stimulus can absolutely look different to two people because of visual processes that also differ between those people. Cognitive processes absolutely contribute to our experience of seeing too, but some differences in appearance are definitely about visual processing.

The difference between novices and experts is 'knowing what to do'.

Sure, but there are other differences worth considering! Knowing what to do depends on one's memory, one's perception of the board, and lots of other processes, each of which I think is interesting to think about.

But what is 'knowing?'. Some say it's information like memory in a computer, others say is just delayed motor responses (there is no distinction between perception and cognition). In this view, Intelligence is only shown by physical action/potential physical action.

I don't think I follow what you're saying here about delayed motor responses and a perception/cognition distinction. Is the question whether or not it's meaningful to talk about a person knowing something in the absence of them making physical responses?

@RuyLopez1000 said in #2: > Great blog, great explanations of how we perceive faces and the similarity in chess experts to perceiving faces. Very interesting to read and think about. > > >NDpatzer: I think that would have to be the possibility that strong players see the board differently than mere patzers like myself > > >NDPatzer: What does a master subjectively experience that's different from what I see? What (if anything) actually appears different to them? > > One of the great philosophical paradoxes. That an identical visual stimulus can 'look different'. > Thanks for reading! A few thoughts below: > Is what you see 'visual'? It can't be if it can look different. I think I disagree with this, but it depends on what you mean. The same physical stimulus can absolutely look different to two people because of visual processes that also differ between those people. Cognitive processes absolutely contribute to our experience of seeing too, but some differences in appearance are definitely about visual processing. > The difference between novices and experts is 'knowing what to do'. Sure, but there are other differences worth considering! Knowing what to do depends on one's memory, one's perception of the board, and lots of other processes, each of which I think is interesting to think about. > But what is 'knowing?'. Some say it's information like memory in a computer, others say is just delayed motor responses (there is no distinction between perception and cognition). In this view, Intelligence is only shown by physical action/potential physical action. I don't think I follow what you're saying here about delayed motor responses and a perception/cognition distinction. Is the question whether or not it's meaningful to talk about a person knowing something in the absence of them making physical responses?

@NDpatzer

Is what you see 'visual'? It can't be if it can look different.

I think I disagree with this, but it depends on what you mean. The same physical stimulus can absolutely look different to two people because of visual processes that also differ between those people. Cognitive processes absolutely contribute to our experience of seeing too, but some differences in appearance are definitely about visual processing.

I meant that conscious vision is not just what is on the retina. Like when you see a cube, it appears as a 'cube', not a hexagonal shape.

The bistable cube is another example where it can appear from the bottom or top perspective (same visual input, but different visual experience).

The difference between novices and experts is 'knowing what to do'.

Sure, but there are other differences worth considering! Knowing what to do depends on one's memory, one's perception of the board, and lots of other processes, each of which I think is interesting to think about.

How do you conceive the process from seeing the board to playing the move? I'm interested to know about these things.

But what is 'knowing?'. Some say it's information like memory in a computer, others say is just delayed motor responses (there is no distinction between perception and cognition). In this view, Intelligence is only shown by physical action/potential physical action.

I don't think I follow what you're saying here about delayed motor responses and a perception/cognition distinction. Is the question whether or not it's meaningful to talk about a person knowing something in the absence of them making physical responses?

I mean when we play chess, is there parts of the brain that encode in a abstract form ideas like 'place the rook behind the passed pawn'.

Or is cognition just motor responses: 'When you see a position it induces mechanical responses, visual input of a rook endgame position causes a prepared motor response such as preparing to extend your arm to move the rook behind a passed pawn, or moving the mouse to accomplish this action'.

In this case there is no area of the brain that encodes representations, it's just behavioral responses to inputs caused by learning.

@NDpatzer > > Is what you see 'visual'? It can't be if it can look different. > I think I disagree with this, but it depends on what you mean. The same physical stimulus can absolutely look different to two people because of visual processes that also differ between those people. Cognitive processes absolutely contribute to our experience of seeing too, but some differences in appearance are definitely about visual processing. I meant that conscious vision is not just what is on the retina. Like when you see a cube, it appears as a 'cube', not a hexagonal shape. The bistable cube is another example where it can appear from the bottom or top perspective (same visual input, but different visual experience). > > The difference between novices and experts is 'knowing what to do'. > > Sure, but there are other differences worth considering! Knowing what to do depends on one's memory, one's perception of the board, and lots of other processes, each of which I think is interesting to think about. How do you conceive the process from seeing the board to playing the move? I'm interested to know about these things. > > But what is 'knowing?'. Some say it's information like memory in a computer, others say is just delayed motor responses (there is no distinction between perception and cognition). In this view, Intelligence is only shown by physical action/potential physical action. > > I don't think I follow what you're saying here about delayed motor responses and a perception/cognition distinction. Is the question whether or not it's meaningful to talk about a person knowing something in the absence of them making physical responses? I mean when we play chess, is there parts of the brain that encode in a abstract form ideas like 'place the rook behind the passed pawn'. Or is cognition just motor responses: 'When you see a position it induces mechanical responses, visual input of a rook endgame position causes a prepared motor response such as preparing to extend your arm to move the rook behind a passed pawn, or moving the mouse to accomplish this action'. In this case there is no area of the brain that encodes representations, it's just behavioral responses to inputs caused by learning.

@RuyLopez1000 said in #4:

I meant that conscious vision is not just what is on the retina. Like when you see a cube, it appears as a 'cube', not a hexagonal shape.

The bistable cube is another example where it can appear from the bottom or top perspective (same visual input, but different visual experience).

Certainly vision isn't limited to the retina - if we stop there we haven't even made it to the brain yet! The Necker Cube's bistability is something one can still think about in terms of more bottom-up visual processes, though. Binocular rivalry (when the two eyes receive very different inputs and your experience alternates between the two) is another phenomenon where a constant image leads to fluctuating experience, but again, this is something you can still think about in terms of processes happening i visual cortex.

How do you conceive the process from seeing the board to playing the move? I'm interested to know about these things.

You could get very granular with this if you wanted, but I think about perceptual processes first (recognizing pieces, moving one's eyes to see the entire board, integrative processes to combine the information across those eye movements, perhaps mid- to high-level recognition of piece configurations), visuospatial processes that support what we call calculation (considering how different sequences of moves would transform the board), decision-making to select a favorite option, and then motor processes to actually make the move.

But throughout these processes you're also using working memory, possibly trying to recall relevant information from long-term memory, etc.

I mean when we play chess, is there parts of the brain that encode in a abstract form ideas like 'place the rook behind the passed pawn'.

Or is cognition just motor responses: 'When you see a position it induces mechanical responses, visual input of a rook endgame position causes a prepared motor response such as preparing to extend your arm to move the rook behind a passed pawn, or moving the mouse to accomplish this action'.

In this case there is no area of the brain that encodes representations, it's just behavioral responses to inputs caused by learning.

I definitely don't think cognition is just motor responses.

@RuyLopez1000 said in #4: > I meant that conscious vision is not just what is on the retina. Like when you see a cube, it appears as a 'cube', not a hexagonal shape. > > The bistable cube is another example where it can appear from the bottom or top perspective (same visual input, but different visual experience). Certainly vision isn't limited to the retina - if we stop there we haven't even made it to the brain yet! The Necker Cube's bistability is something one can still think about in terms of more bottom-up visual processes, though. Binocular rivalry (when the two eyes receive very different inputs and your experience alternates between the two) is another phenomenon where a constant image leads to fluctuating experience, but again, this is something you can still think about in terms of processes happening i visual cortex. > How do you conceive the process from seeing the board to playing the move? I'm interested to know about these things. You could get very granular with this if you wanted, but I think about perceptual processes first (recognizing pieces, moving one's eyes to see the entire board, integrative processes to combine the information across those eye movements, perhaps mid- to high-level recognition of piece configurations), visuospatial processes that support what we call calculation (considering how different sequences of moves would transform the board), decision-making to select a favorite option, and then motor processes to actually make the move. But throughout these processes you're also using working memory, possibly trying to recall relevant information from long-term memory, etc. > I mean when we play chess, is there parts of the brain that encode in a abstract form ideas like 'place the rook behind the passed pawn'. > > Or is cognition just motor responses: 'When you see a position it induces mechanical responses, visual input of a rook endgame position causes a prepared motor response such as preparing to extend your arm to move the rook behind a passed pawn, or moving the mouse to accomplish this action'. > > In this case there is no area of the brain that encodes representations, it's just behavioral responses to inputs caused by learning. I definitely don't think cognition is just motor responses.

Firstly, let me say that I love the "Put an APB out for Linus" police sketch example! There's a cute mini-story just contained in that (initial scorn for the sketch artist, and her subsequent vindication).

Anyway, maybe part of @RuyLopez1000 's question is whether the distinction between the good and bad chess players is about "seeing", or if it's about some other layer of understanding.

I know nothing about building bridges, so excuse me if this example doesn't actually make sense. But I can imagine showing bridge designs (blueprints? architectural diagrams?) to some engineers. Then you try to do the same kind of experiment where you mix and match pieces of the bridge (maybe the top vs the bottom, or the supporting elements vs the road surface, etc.).

I can imagine some expert civil engineers always trying to understand the entire bridge at once, perhaps even compulsively, so that they would have trouble isolating just the changed bits. But that might just be their habits from long training, and may not imply that they really "see" the bridge plans differently.

Similarly, an expert chess player certainly understands the board more holistically, and when given a position, thinks about the "story" of the position in a way a novice doesn't. But that may not always be because they "see" the position differently.

(I know you have some other blog posts, which get into eye-tracking and which do show that experts do see the position differently.)

Firstly, let me say that I love the "Put an APB out for Linus" police sketch example! There's a cute mini-story just contained in that (initial scorn for the sketch artist, and her subsequent vindication). Anyway, maybe part of @RuyLopez1000 's question is whether the distinction between the good and bad chess players is about "seeing", or if it's about some other layer of understanding. I know nothing about building bridges, so excuse me if this example doesn't actually make sense. But I can imagine showing bridge designs (blueprints? architectural diagrams?) to some engineers. Then you try to do the same kind of experiment where you mix and match pieces of the bridge (maybe the top vs the bottom, or the supporting elements vs the road surface, etc.). I can imagine some expert civil engineers always trying to understand the entire bridge at once, perhaps even compulsively, so that they would have trouble isolating just the changed bits. But that might just be their habits from long training, and may not imply that they really "see" the bridge plans differently. Similarly, an expert chess player certainly understands the board more holistically, and when given a position, thinks about the "story" of the position in a way a novice doesn't. But that may not always be because they "see" the position differently. (I know you have some other blog posts, which get into eye-tracking and which do show that experts do see the position differently.)

Anyway, maybe part of @RuyLopez1000 's question is whether the distinction between the good and bad chess players is about "seeing", or if it's about some other layer of understanding.

I'd say there isn't going to be just one distinction between good and bad chess players, but many things that differ. Being a vision scientist, I'm most interested in the possibility that perceptual processes differ with expertise. That's not to minimize the other likely differences in memory, emotion regulation, etc. - I just like the vision-y stuff.

I know nothing about building bridges, so excuse me if this example doesn't actually make sense. But I can imagine showing bridge designs (blueprints? architectural diagrams?) to some engineers. Then you try to do the same kind of experiment where you mix and match pieces of the bridge (maybe the top vs the bottom, or the supporting elements vs the road surface, etc.).

I can imagine some expert civil engineers always trying to understand the entire bridge at once, perhaps even compulsively, so that they would have trouble isolating just the changed bits. But that might just be their habits from long training, and may not imply that they really "see" the bridge plans differently.

Part of how we try to determine if effects like the CFE are about perception vs. cognition has to do with the judgments we ask people to make. In the case of someone looking at blueprints like you described, If we limited presentation time and asked for a same/different judgment (are these images identical or not?), for example, we'd be more confident that a composite effect had more to do with perceptual processes than later cognitive processes.

It's a good question to ask, though: Behavioral manipulations are imperfect tools for isolating neural processes, so we always have to be careful. In a real game, all those processes are working together in parallel as well, so high-level understanding of the game may interact with low-level perceptual processing.

> Anyway, maybe part of @RuyLopez1000 's question is whether the distinction between the good and bad chess players is about "seeing", or if it's about some other layer of understanding. I'd say there isn't going to be just one distinction between good and bad chess players, but many things that differ. Being a vision scientist, I'm most interested in the possibility that perceptual processes differ with expertise. That's not to minimize the other likely differences in memory, emotion regulation, etc. - I just like the vision-y stuff. > > I know nothing about building bridges, so excuse me if this example doesn't actually make sense. But I can imagine showing bridge designs (blueprints? architectural diagrams?) to some engineers. Then you try to do the same kind of experiment where you mix and match pieces of the bridge (maybe the top vs the bottom, or the supporting elements vs the road surface, etc.). > > I can imagine some expert civil engineers always trying to understand the entire bridge at once, perhaps even compulsively, so that they would have trouble isolating just the changed bits. But that might just be their habits from long training, and may not imply that they really "see" the bridge plans differently. Part of how we try to determine if effects like the CFE are about perception vs. cognition has to do with the judgments we ask people to make. In the case of someone looking at blueprints like you described, If we limited presentation time and asked for a same/different judgment (are these images identical or not?), for example, we'd be more confident that a composite effect had more to do with perceptual processes than later cognitive processes. It's a good question to ask, though: Behavioral manipulations are imperfect tools for isolating neural processes, so we always have to be careful. In a real game, all those processes are working together in parallel as well, so high-level understanding of the game may interact with low-level perceptual processing.

This is certainly one of the highest quality blog posts on the whole site. As someone with a professional connection to STEM journalism, I salute you! This is excellent stuff!

This is certainly one of the highest quality blog posts on the whole site. As someone with a professional connection to STEM journalism, I salute you! This is excellent stuff!

Hi there, Dirk Gently.

Hi there, Dirk Gently.