Gorilla's At The Chess Board

Sort:
Kingpatzer

There's an interesting piece on NPR today about radiologists who don't see a gorilla in an image when they are searching for cancer.

 

In it they reference the now famous gorilla study. You can see the illusion the study was based on in this video:

 

[youtube http://www.youtube.com/watch?v=IGQmdoK_ZfY&w=640&h=360]

 

if you haven't tried the video, you really should. There's a great follow up video from a Ted talk by one of the study members:

 

[youtube http://www.youtube.com/watch?v=9Il_D3Xt9W0&w=640&h=360]

 

There is also a lengthy presentation from Authors@Google:

 

[youtube http://www.youtube.com/watch?v=4rdUk52h-MY&w=640&h=360]

 

But the basic idea comes down to this: When we are focused intently on looking for one type of information, we are often amazingly blind to other types of information.

 

This is directly applicable to chess, I believe.

 

In recent games I've been plagued by blunders that happen in very specific situations. The situations type is observable from the following diagram:

 

 

I am black and I've already looked for threats from my opponent. I considered Ne5-c4, but rejected it as a threat I need to meet because I have a response to that attack. The response is that I'll move my queen to some available square.

 

Now, I've moved on to considering my own moves. And I'm looking at the idea of Ba3. What makes Ba3 so attractive? Well, it puts a lot of pressure on the black king position, and after moves such as Nb4, it will be hard for white to last long unless he has a good response. That is, presuming he doesn't see the obvious one-move refutation to the whole idea.

 

So I start looking for how black can respond. I consider Ne5-d3, defending the bishop and the b4 square. I consider Ne5-g4 attacking the f6 knight. I cnosider other options as well. But what's important is that I calculated lines involving the knight on e5 but I didn't consider Ne5-c4 as a possibility worth considering.

 

So why didn't I? It's not like I don't know about forks. I was legitimately blind to the possibility. I didn't see it. It wasn't there. 

 

I think I am falling victim to the gorilla!

 

I've looked at that knight once, it's not a threat. Now, I'm calculating my own attack and I'm looking at responses, but because I want to see my attack as successful, I'm ignoring one-move responses that would make my attack fail. Because that's not what I want to see!

 

I think the gorilla study has a great deal to tell us about chess blindness. And it is why having a regimented thinking process that makes us pause and consider what type of information we are looking for is so important to being successful.

Shivsky

I think the blindness is specific to what's happening "outside" the context of where our focus is. So I don't think the gorilla (as the article alludes to) can be something on-the-board that is "chess-based".

A more exact analogy would be chess would be a guy in a gorilla suit walking past your board when you both have seconds left on the clock. 

Kingpatzer

Shivsky, if you watch the Authors@Google video, you'd see that is not the case. 

It's about how our perception and memory relate to reality that is at issue. 

AndyClifton

I would think that having a "regimented thinking process" is the sort of thing that would lead to blindness...not the opposite.

Kingpatzer

Perhaps "regimented" is the wrong word. I'm thinking about specifically doing whatever we can to avoid blind spots in our analysis through making ourselves look at the different aspects of the board in order to attempt to elicit recognition of all the salient features. 

If you watch the video looking for the Gorillas, you'll see them. But you'll likely not manage to count all the passes, or you'll miss something else.

I think the same thing applies to the chess board, if we look for forks, we'll find them. If we look for 1-move refutations of an idea, if it exists it will be seen. 

plutonia

I understand what you mean and I think this is an interesting thread.

However I don't experience what you describe, probably because I have a different way of thinking.

I first analyze my possible moves, then for each I try to guess my opponent's responses individually. I mean, I try to picture the board after my move and look for the best move for my opponent, being careful that one analysis doesn't spill onto the other analysis.

 

I wouldn't fall into your example for an additional reason: I make a mental note when I have a tactical weakness, such as an unprotected piece or having a checkable king. In this case I would have made a mental note that my Queen had to prepare for a hit from that knight. These mental notes are all checked before moving.

fissionfowl

Seems like you're over-thinking this. It's a common mistake to think a move isn't worth looking at because it was no problem to deal with in another only slightly different position you've just been analyzing. This is probably just human nature, so just try to be vigilant if the same type of mistake happens often.