What Should Happen in Your Head When You Visualize?
Guest post by Aiden Rayner
Welcome to Say Chess! This newsletter goes out to 4,042 chess players. If you haven't joined yet, sign up now and get the ebook '100 Headachingly Hard Mate In Two Puzzles Composed By Sam Loyd' for free.
This week’s newsletter is a guest post by Aiden Rayner. Aiden is the founder of Don’t Move Until You See It, and he is most probably the only dedicated chess visualization/conceptualization coach in the world.
After picking up chess in 2020 at age 27, Aiden developed his own method for improvement through isolated, intensified conceptualization training, which saw him raise 500 ELO online in under 6 weeks.
Since then he has been focused on helping adult improvers improve their skills and communicating the psychology, history, and neuroscience of chess conceptualization.
His goal is to bring this powerful and deeply neglected part of the game into the mainstream for adult improvers everywhere.
Aiden reached out to me after reading my blog “Insights into Chess Visualization Skills”, which he wanted to comment on and expand upon. I thought it was a brilliant idea to give space to commentary on an earlier experiment that some of you have participated in.
What Should Happen in Your Head When You Visualize?
“Theoretically, it is possible to play blindfold chess without visualization, merely by remembering all the moves. This requires such a prodigious feat of memory that I have met only one person who has played several games simultaneously in that way.”
- Reuben Fine
"I do not proceed on a visual basis, but on a mathematically reasoned strategy. These men who play visually lack certainty in their games - and they lose most of them."
- Samuel Rosenthal, when asked what he sees in his head when he plays blindfolded.
We're all familiar with the classic image of Chess visualization in action. It was perhaps best presented in The Queen's Gambit, with Beth Harmon peering up at a detailed, colorful, and fully realized chessboard on the ceiling. There's something about this image that captivates chess players and non-players alike. The level of mastery it must require to have a fully realized image of the board in your mind, that you can perceive as well as any board in front of you.
But do clear mental pictures equate to Chess skill?
It’s a question that’s fascinated Chess players for centuries. And various psychologists, researchers, and instructors have weighed in over the years. In 2021, Say Chess’ Martin Justesen surveyed 165 players to try and find the answers himself, and his results were fascinating.
I’ve spent the past three years deep in research into chess visualization and cognition. Through my training platform, Don’t Move Until You See It, I’ve trained hundreds of adult improvers to supercharge their visualization skills. Along the way, I have collaborated with some of the best Chess educators around, and become the only dedicated Chess visualization coach in the world.
In this article, I offer my perspective on the data collected in Martin’s survey.
The key takeaway? We have visualization all wrong.
What we get wrong about visualization
The key to building Chess strength, calculating accurately, and playing blindfold Chess, most players think, is mental pictures. The more detailed the pictures, the clearer the image, the better the player must be. Crisp, detailed pictures are the sign of a good player, and you can't be a good player without these pictures in your head.
So goes one of the biggest myths in Chess. And it is a myth.
In the 1930s, Belgian-American chessmaster George Koltanowski set a world record. Not just any record either, but one of the most impressive feats in all of Chess. He successfully completed a blindfold simultaneous exhibition of 34 boards at once!
His skills in blindfold play were so strong that the great Alexander Alekhine once said "Koltanowski is the second best blindfold player in the world". (He put himself as number one, obviously; false modesty was not one of Alekhine's vices.)
Clearly, he's one of the best blindfold players we've ever seen. His visualization skill must have been truly monumental. Surely, it had to be. Right?
It wasn't.
Koltanowski never saw pictures in his head.
In an article for the San Francisco Chronicle in 1960, he wrote: "My mind is a gramophone record. When I want to know what moves have been made, I start the record in my mind. Then I listen."
In his 1990 book1, he clarified further: "I do not see the board or pieces in my mind; I just remember the moves and 'feel' the position."
He completed a 34-board blindfold simul, securing his spot as one of the greatest "visualizers" in history, and he did it with no mental pictures at all.
When we dig a little deeper, we find that plenty of the great players (for example Alekhine, Rosenthal, and Najdorf), either didn't see pictures in their heads, or relied far less on them than what we would expect. The ability to see mental pictures has nothing to do with strength.
We find clear evidence of this in Martin's data on the percentage of players who see nothing in their mind’s eye when “visualizing” the starting position.
If mental pictures were the key, we would see a clear downward trend in the percentage of players who see nothing with each rating tier. Instead, the percentages are practically even across the board.
Visualization does not have to be visual. And for roughly 40% of adult improvers, trying to force it to be visual is a waste of time and energy.
That's why I use a new term instead: conceptualization.
What is conceptualization?
In 1958, George Miller introduced a concept that has had wide-reaching effects on our understanding of psychology.
He proposed that our working memory, the part of our memory that holds temporary information pertaining to our current tasks, has a limit to its capacity. He concluded we have 7 “slots” for information in our working memory.2 (That’s an average, some of us get 5, and others up to 9.)
So we have around 7 slots in our working memory. There are 32 pieces on the chessboard. As I like to say, one of those numbers is bigger than the other.
To help us juggle all the information, our brains create a representation of the information. For some people, it takes the form of a mental picture. For others, it doesn’t. The method our brain uses to represent Chess information to us is called our conceptualization model.
Your conceptualization model actually has more to do with your cognition than it really does with Chess. Our brains don't see much of a difference between Chess information and any other kind of complex information we have to process. And we've been dealing with all sorts of complex information since the day we were born.
What we think of as visualization, the crystal clear picture of a board in our mind, is simply one conceptualization model. Koltanowski's "mental gramophone" is another. Each of us will have our own unique conceptualization model, reflecting how our brain naturally likes to process and represent information.
Your natural thinking style
"The world can be roughly divided into two kinds of thinkers: people who think in pictures and patterns, and people who think in words." - Temple Grandin, Visual Thinking (2022)
Since the 1930s, psychology studies have played with the idea that people are naturally either visual or verbal thinkers. They are either guided by images their mind creates or their inner monologue.
To get a quick sense of which you may be, try what Temple Grandin, world famous educator and advocate, describes as the "IKEA test":3
You buy a piece of flat-pack furniture. You get home, get your tools, open the boxes. You're ready to build. Would you prefer to read clear instructions, or look at clearly illustrated pictures?
If you prefer the pictures, you're likely a visual thinker. If you prefer the written instructions, you're likely a verbal thinker. It's not a 100% correlation, but it can be a good indicator.
These visual and verbal thinking styles aren't just preferences either; they're observable, measurable, biological differences in our brains. A 2015 study by Kazuo Nishimura had subjects complete complex mental tasks while hooked up to various devices to measure neurological activity.4 They found some subjects activated the sections of the brain known to produce internal visual imagery while others activated the parts of the brain connected to self-talk.
When presented with the same complex mental tasks, some of our brains use visuals, and others use verbal methods. Same tasks, different approaches.
In a 2002 study, Dr Linda Silverman at Rutgers University showed that these visual and verbal alignments are not an either/or argument, but instead occur across a spectrum.5
Each of us exist on this spectrum somewhere. Some of us are strongly visual, some strongly verbal, and most of us are in the middle somewhere, learning one way or the other.
Our location on the spectrum aligns with how our brains naturally want to handle chess information as well. There are as many different ways to conceptualize as there are points on this spectrum. And the Chess world's obsession with mental pictures is a huge problem for everyone on the verbal side.
We see evidence of this spectrum in Martin’s survey. The first question asked the participants to “close your eyes and try to visualize the starting position,” then select the option that best describes the resulting mental image.
The results were as follows:
48% saw a 2d board
8% saw a 3d board
29% saw a blurry image of the board
5% saw something else
10% couldn’t visualize anything and saw nothingness/darkness
56% of the participants reported seeing a clear image in their mind. These people are likely on the visual side of the thinking spectrum. We can also see that 39% of players either couldn’t see anything in their minds or their mental image was blurry and indistinct. Verbal thinkers often can visualize when they need to (or when they’re asked to) but it’s not their natural way of thinking, and tends to be limited in its fidelity. These 39% of players are likely on the verbal side of the spectrum.
Depending on our spot on the spectrum, our brain will naturally want to represent chess information to us in different ways.
The great news is there seems to be no correlation between a player’s strength and their location on this spectrum. We’ve had plenty of top players from each location along it. The bad news is that the myth that we need mental pictures leaves a staggering amount of verbal-thinking chess players feeling they’ll never improve or that their brain is somehow broken. You only need to scroll through Chess visualization threads on Reddit to find it over and over again. It hurts every time I see it.
If you don’t naturally visualize, you are not broken.
The key for each of us is to identify how our unique brain wants to conceptualize and let it do that. Visual, verbal, whatever. It doesn’t matter.
Let’s break the spectrum down in more detail.
Note: it’s difficult from a distance to categorize players’ models with certainty, especially players from the past. Please use the lists of notable players as an indication of where they may sit in their Chess, and not as an assertion of each individual’s thinking styles in wider life.
A special thank you to John Knott and Eliot Hearst for their 2008 book, “Blindfold Chess: History, Psychology, Techniques, World Records, and Important Games”, from which I have gathered much of the evidence regarding individuals’ models.
Visualizers (33% of us)
Visualizers use mental images and movies as their primary way of thinking. Some see photorealistic 3D images that they can turn and manipulate in their mind. Others think in abstract visual representations and patterns.
Strong visualizers tend to prefer video and image-based content for learning, and their workspace is often chaotic. They know where everything is in the chaos, but if someone were to come and "clean it up", they'd never be able to find anything again. They tend to learn things holistically all-at-once, and see the patterns that connect ideas from across domains. They are the people who instantly spot the one tiny thing that’s different in an otherwise familiar room. These are our Leonardo Da Vinci, Steve Jobs, and Mozart types.
In Chess, visualizers tend to:
Create either a photorealistic image or an abstract visual representation of patterns and "lines of force" in their mind
Love the geometry of Chess, and get excited by the shapes and patterns of the game
Reconstruct chess positions by remembering the shapes and patterns of the pieces.
Work out details like square colors by picturing the pattern on their mental board
Struggle with opening theory, as they don’t naturally think in sequences.
Notable visualizers include:
GM Reuben Fine. - once asserted that playing blindfold Chess without visual imagery would be exceedingly difficult for anyone. It’s safe to assume then, that Fine heavily relied on visual imagery with his own model.
GM Fritz Samisch. - After Samisch’s death, his close friend Val Zemitis contested an assertion made by GM Andrew Soltis that “nobody sees the whole 64 squares in their head at once.” Zemitis firmly believed Samisch did in fact see the whole board at one time in his mind, and that the two of them regularly discussed it.
FM Lefong Hua. - established in discussions with Lefong. He sees a complete 64-square, full color board in his mind. When he was younger and playing more OTB tournaments, the board was in 3D. Now that he’s primarily playing online, it’s in 2D. He can flip back and forth with ease.
Verbalizers (25% of us)
A strong verbalizer relies on their internal monologue to interact with and make sense of the world. When sitting and thinking, their mind is not seeing pictures, but instead talking to itself. They tend to be organized, have neat handwriting, spot typos a mile away, and think aloud to whoever will listen. They remember what they read and hear far better than what they see.
They tend to remember and learn things in sequences, step-by-step piece-by-piece, and make logical connections between ideas, often through analogy. They tend to be articulate, and tricky to argue with as their grasp of language and logic is strong. These are our Stephen Fry, Socrates, and Benjamin Franklin types.
In Chess, verbalizers tend to:
Be excellent at remembering opening theory (due to their comfort with sequences and sequential thinking) - this is a superpower!
Use logic-based techniques to work out details like square colors. Eventually they no longer need these techniques because they have memorized the squares.
Struggle far more conceptualizing an in-progress middlegame position than following a game from the beginning (again, because of sequences)
Regularly make mistakes to do with diagonals (Diagonals are extra tricky for verbalizers as there's no verbal cues that any two squares share a diagonal.)
Struggle with tracking the distance between squares in their mind, often over- or under-estimating them.
Notable verbalizers include:
GM George Koltanowski. - wrote: “I do not see the board and piece in my mind.” His model relied on his “mental gramophone”, his verbal recollection of the moves as they had been played.
IM David Pruess. - has talked publicly about his aphantasia and the non-visual methods he uses when conceptualizing.
Samuel Rosenthal. - responded to a survey on blindfold Chess: “I do not proceed on a visual basis, but on a mathematically reasoned strategy. These men who play visually lack certainty in their games - and they lose most of them."
Mid Spectrum (42% of us)
The bulk of us exist somewhere between the extremes of fully visual and fully verbal thinkers. We visualize and we listen to our internal monologue. But we'll each have a preference one way or the other; a main mode that takes most of the load. We'll have a mix of the strengths and challenges of the strong visualizers and strong verbalizers.
There is huge variation in conceptualization models among mid-spectrum players. Some, like IM Eric Rosen, describe their mental board as “shining a flashlight in a dark room”. They’ll see sections of the board one-by-one, and remind themselves how they go together using verbal and logical connections. Others are primarily verbal, but use flashes of mental pictures here and there to double check details.
Notable mid-spectrum players include:
GM Alexander Alekhine. - Alekhine relied on his “logical memory”, his sense of the story of the position, when playing blindfold Chess. He used flashes of visuals here and there when he needed to double-check something, but mostly went without them. I place him as a mid-spectrum verbalizer.
GM Richard Reti. - Reti offered routines and techniques for training blindfold skill. His recommendations indicate a mix of the visual and verbal, leaning further towards the visual side but with structure provided by non-visual techniques.
IM Eric Rosen. - In a video on his YouTube channel, Rosen describes his model as “shining a flashlight in a dark room”. He’ll visually recall chunks of the board at a time, then remember logically how those chunks go together. This is a common mid-spectrum visual model.
Aphantasia
Aphantasia is the name given to a cognitive phenomenon where someone is physically unable to produce pictures in their mind. As an aphant once described it to me, “the computer is operating, the monitor’s just turned off.” The research into aphantasia is still in its infancy, but it’s estimated that 2-3% of the population have it. Some researchers think this number may be much higher, as many people with aphantasia are not actually aware they have it.6
When aphants discover and fall in love with Chess, they quickly run into this idea of "visualization" and start to doubt themselves. As Dr Kevin Scull on the Chess Journeys podcast told me, "I’ve been waiting to hit my limit; the point where my aphantasia stops me from going any further."7
If you've had the same doubts, you don't need to despair. You don't need to "visualize" to play well. Whatever your brain does will work fine. Aphantasia is not a weakness.
The term aphantasia only appeared in 2015, so it's difficult to diagnose the great Chess players from the past with any certainty. But, whether George Koltanowski had aphantasia or not, it’s clear his conceptualization model was completely non-visual.
And if playing 34 games of blindfold Chess at once doesn’t indicate he was good at the game, I don’t know what does.
Does our conceptualization change as we get stronger?
One of the most famous breakthroughs in Chess psychology research was made by Herbert Simon and William Chase where they identified that much of a Chess player's strength comes from "chunks" they have stored in their long-term memories. While later researchers have argued and updated some of the specifics of Simon and Chase's research, the core of it holds strong: Chess masters have thousands of "chunks" of chess positions stored up in their long-term memory.
We have these chunks too, just fewer of them. I'll show you what I mean.
Take a moment and think of a kingside castle for White. You probably don't need to try too hard to remember where all the pieces are. You're aware the king's on g1, the rook's on f1, and the pawns are in a line above them on the second rank. Depending on where you are on the spectrum, your mental representation of this pattern will be different - but regardless it didn't take you much effort to conjure it in your mind.
This is a chunk in our long-term memory. It takes up very little space in those 7 slots of working memory and comes with a bunch of extra details (like we know without thinking that a castled king is vulnerable to back-rank mates). Once a chunk is in our long-term memory, it's much easier to hold, reference, and manipulate when we're conceptualizing.
It's estimated that Chess masters have over 50,000 chunks stored in their long-term memory. The way the kingside castle chunk is easy for most of us, 49,999 other chunks are easy for them as well. The ease with which they can recall and hold these chunks relieves a lot of pressure on their working memory and conceptualization skills.
It’s worth noting that most verbalizers, when asked to visualize a chunk they are familiar with, can force themselves to conjure some image of it. Sometimes even with reasonable clarity.
That may help explain the trend Martin found when he asked how clear the survey participants’ mental images were after a 10-move line that ended with this position:
We can see a clear upward trend in image clarity as rating increases.
I believe there are two possible reasons for this.
First, a 10-move sequence is a LOT for anyone who hasn’t trained their blindfold skills. Strong players will have a far clearer idea of what’s going on by the end of the sequence than will a weaker player, regardless of their conceptualization model.
Second, the question, “how clear is your mental picture of the position?” assumes each of the players has a mental picture, and many of the higher-rated verbalizers would have been able to conjure a mental picture of the familiar position in order to answer the question they were asked.
If this survey were to be repeated, I would reframe this question to focus on how things were being represented in participants’ heads (like the earlier question about the starting position). I expect the spread of data we would get from that would more closely resemble the visual-verbal spectrum percentages.
Martin’s survey ends by reminding us again that a fully detailed visual picture of a board is not the marker of strength. The final question asked participants whether they see the color of the squares in their mind. The results across the rating ranges was as so:
The graph trends upwards, with a big spike at the end. Stronger players are more likely to see the colors of the squares in their minds.
But the most interesting bit is not the spike but the percentage: only 50% of 2200+ players see the square colors. That’s less even than the 63% of people who lean towards the visual side of the visual-verbal spectrum.
Stronger players have a clearer understanding of the board in their head than the rest of us- that just makes sense. For visual thinkers, this means they may have a fully realized photographic image of the board in their head (including square colors). For verbal thinkers like Koltanowski, it might mean they’ve memorized things or used logical systems to work out various details when they become relevant. Either way, the result is the same.
What matters for strength is not pictures, but clarity.
What should we do in our heads?
As you've seen, there have been plenty of great players from across the conceptualization spectrum. It doesn't matter how you represent chess information in your head, only that you do.
The key is to find your natural conceptualization model, one that aligns with how your brain works. If it’s visual, great! If it’s verbal, fantastic! If it’s somewhere in the middle, whatever works!
It really doesn’t matter, as long as it works for you.
The best method to start exploring your model is with blindfold training. The intensity of blindfold training forces us to build a model that works to our strengths. It will take some trial and error, and it will be difficult. But it’s also really fun.
I created the Don’t Move Blindfold Trainer as a place for people to simulate true blindfold chess games. I intend for it to be free forever, so use it as often as you like. You can find it here.
Martin here at Say Chess has been an advocate for Blindfold training for years, and has released two books full of blindfold exercises: Blindfold Opening Visualization and Blindfold Endgame Visualization.
Regardless of the resources you use to train, focus on building the clarity of your mental board.
Don’t try to force pictures if they don’t come naturally, don’t try to copy your favorite streamer, don’t listen to anyone who tells you how you should be conceptualizing.
Find what your brain wants to do, then trust it. I promise it will have a huge impact on your Chess.
Brains are awesome, yours included.
Learn more
If you want to go deeper and learn more about Don't Move Until You See It, the next step is my 5-day Conceptualizing Chess Series.
I designed this free course to give you all the key insights you need to improve your conceptualization skills. The ideas within lay the foundation for everything I do at Don’t Move Until You See it. There’s a ton of unique blindfold exercises too.
Get the 5-day Conceptualizing Chess Series here.
Thanks to Aiden for writing this interesting and thoughtful review of chess visualization. It has given me something to think about, and maybe it will give hope to some readers who might have felt that they had no ability to see a clear mental picture of the board.
Later this week or at the beginning of next week I will share a newsletter about a tilt-o-meter I have built.
/Martin
Koltanowski, G. George Koltanowski: Blindfold Chess Genius (1990)
Miller, G. A. “The magical number seven, plus or minus two: Some limits on our capacity for processing information.” Psychological Review 101 (1958) p 343-352
Grandin, T. Visual Thinking (2022) p 17.
Nishimura, K., et al. “Brain Activities of Visual and Verbal Thinkers: A MEG Study.” Neuroscience Letters 594 (2015): 155-160
Silverman, L. K. Upside Down Brilliance: The Visual-Spatial Learner. Denver: Deleon, 2002.
Kendle, A. Aphantasia: Experiences, Perceptions, and Insights. (2017)
Scull, K. “Episode 82: Aiden” Chess Journeys. Podcast. (2023)
Interesting read! I do both, for some tasks I visualise a picture or 3D model I can move through, this usually requires sitting still, for others I talk to myself, often in the voices of people I've listened a little to much to on youtube, this I can easily do whilst walking but requires me to look down so I can see where I'm walking.
Here something else for you to think about: Does the way you think affect your chess style? Visual vs verbal, positional vs tactical (or other styles).
Fantastic! Thanks, Martin and Aiden!