Winning starts with what you know
The new version 18 offers completely new possibilities for chess training and analysis: playing style analysis, search for strategic themes, access to 6 billion Lichess games, player preparation by matching Lichess games, download Chess.com games with built-in API, built-in cloud engine and much more.
Emanuel Lasker once shared one of his biggest secrets in becoming a chess grandmaster. He is often quoted as saying: "When you see a good move – wait – and look for a better one!" Valuable advice, because players often make mistakes in positions where a familiar solution is present. This can happen when a position looks similar to previous ones solved. Believing you actually played the best move in the game, you later find out an even better one in the post mortem analysis. You then ask yourself: "Why didn't I find that move during the game?”
Can a familiar solution prevent us from discovering a better one? In chess, and many other fields, the presence of a familiar solution can block the discovery of a less familiar, better solution. A recent scientific study has looked at this issue, by comparing how expert and novice chess players solve chess problems. You can download the full article here: Sheridan & Reingold (2013). Or, you can read the summary below for a short description of the main highlights from the study.
When we solve problems, our prior knowledge usually helps us by efficiently guiding us towards solutions that worked for us in the past. However, if a problem requires a new solution, then sometimes our prior knowledge can make it surprisingly difficult to discover the new solution. This problem-solving effect was discovered 75 years ago by a psychologist named Abraham Luchins. In a famous problem-solving experiment, Luchins asked people to use jugs of water to measure specific quantities of water. He first gave them five introductory water-jug problems that could be easily solved using a simple solution. Next, he gave them a superficially similar problem that required a new solution. Many people said that this new problem was impossible, even though it was easily solved by another group of people who had not seen the introductory problems. In this experiment, prior knowledge prevented people from discovering a new solution. Luchins called this effect the Einstellung effect (Einstellung is a German word, which means “setting”).
Does the Einstellung effect also occur in chess? Several years ago, a group of researchers – Merim Bilalic, Peter McLeod, and Fernand Gobet (pdf) – created a chess version of Luchins’ water-jug experiment. They asked chess players to solve chess problems that contained both a familiar solution and a less familiar, but more advantageous, solution. Like the participants in the water-jug experiment, many chess players couldn’t find the better solution, and an analysis of the chess masters’ eye movements revealed that the chess players were spending a large percentage of time looking at chess squares associated with the familiar solution, even when they said they were searching for a new solution. Based on these results, the researchers argued that the Einstellung effect operates by biasing our attention towards the familiar solution, thereby preventing us from discovering a new solution.
More recently, a new study by Heather Sheridan and Eyal Reingold (in collaboration with Rick Lahaye) examined the question: “What conditions are most likely to produce the Einstellung effect in chess?” These researchers compared the performance of 17 novice chess players (unrated club players), and 17 chess expert players (average Elo rating = 2223), while they solved four chess problems that were designed to induce Einstellung-like effects.
All of these problems were designed to falsely give the impression that a familiar checkmate solution is possible (i.e., the Einstellung solution). This familiar Einstellung solution was always located in one corner of the board (the region is indicated by dotted lines as shown in the figure above), and the optimal moves on the board were always located outside of this region. The chess players’ task was to choose White's best move.
If you want to participate in this experiment you should spend some time looking at the positions before you look at the solutions lower down on this page. Make a note of the ideas that popped into your mind, the areas of the chessboard that kept your attention of most of the time. Then compare this to the results given below.
Incidentally, the researchers used eye movements to study the percentage of time that the chess players spent looking at the familiar solution (i.e., the dotted region of the board).
For an example of the eye movements of one of the chess experts, see the above video (the purple dot indicates where the chess player was looking), and image below (the red dots indicate the location and duration, in milliseconds (1 second = 1000 milliseconds) of each of the chess players’ eye fixations). The video and the following diagram were created using the Data Viewer software created by SR Research.
As can be seen from this example, this chess player spent a large amount of time looking at the familiar solution in the corner of the board. Since eye movements are a good index of where we are directing our attention, this suggests that the chess player spent a lot of time focusing on the familiar solution. However, over time, the expert began to look at the optimal move on the board, and at the end of the trial, the chess player successfully chose the optimal move on the board (rook to b3).
Well, here are the solutions (best moves, in green) to the four positions and the Einstellung moves (red) that captured most people's attention:
To study which conditions produce the strongest Einstellung effect, the researchers contrasted two types of problems. In one type of problem, the familiar Einstellung solution was a reasonable move (i.e., bishop to a7 in Problem 1), although it wasn’t one of the best moves on the board. In the remaining three problems, the Einstellung solution was a blunder.
To study the chess players' bias in attention towards the familiar solution, the researchers looked at the average percentage of time that the chess players spent looking at the familiar solution. This percentage of looking time measure revealed that the chess players found it easier to disengage their attention from the familiar “Einstellung” solution when it was a blunder (Problems 2, 3, and 4), rather than a suboptimal move (Problem 1). Consistent with this finding, most of the chess players were able to avoid choosing the familiar solution when it was a blunder, but almost half of the players chose the familiar solution when it was a suboptimal move.
The above table is a summary of the chess players’ move choices for problem 1. The experts showed superior performance, but the experts and novices were equally likely to choose the Einstellung move. "Quality rating" was derived by five expert chess players who did not participate in the study. They rated each move on a scale from 1 to 10 (1 = a blunder, 10 = a very strong move).
Move Choices for Problems 2, 3, and 4 (the blunder problems) – for the blunder move problems, all of the experts and the majority of the novices avoided choosing the Einstellung move. These findings suggest that it was easier for the chess players to resist the Einstellung effect when the familiar solution was a blunder, rather than a good (but suboptimal) solution.
The researchers also discovered an interesting difference between the pattern of eye movements for the experts and novices. Compared to the novices, the experts were more likely to alternate between looking at the familiar solution, and looking at the other parts of the board. In other words, their eyes would leave the familiar solution and then come back to it. This pattern of “alternating” eye movements was especially common near the end of the trial. After the experiment, some of the chess experts commented that they had spent time strategizing about how the optimal move would impact the pieces in the familiar solution. Based on these comments, we speculate that the chess experts sometimes looked at the familiar solution even though they had ruled out checkmate, because they were considering the implications of the optimal move for the pieces associated with the familiar solution. The novices were less likely to show this pattern of alternating eye movements, which suggests that the novices were not engaging in the same type of strategizing as the experts.
In conclusion, the Einstellung effect can cause us to spend a lot of time examining a familiar solution. By narrowly focusing on a familiar solution, we can sometimes miss a less familiar, better solution. Chess players find it easier to avoid the Einstellung effect when the familiar solution is a blunder (rather than a suboptimal solution), possibly because a blunder provides clear feedback that the familiar solution is no longer appropriate.
Psychological warfare and Einstellung effect – part 2
After reading the summary, you might ask yourself what you can do to prevent such errors in your own games. The one-liner of Lasker to control yourself is valuable, but there's much more to it. In the second part of the article we will have a look at what you can do to improve your play and use this knowledge about Einstellung for your own advantage. If you have any questions or feedback concerning the research, please let us know. You can contact us at H.Sheridan (at) soton.ac.uk and rick (at) kennisstroom.nl.
Heather Sheridan is a Postdoctoral Researcher at the Centre for Vision and Cognition (CVC), at the University of Southampton. Her current work uses eye movements and computational modeling to study visual expertise in reading and chess. More broadly, she is also interested in understanding expert/novice differences in problem-solving and human memory. She collaborates with Prof. Erik Reichle at the University of Southampton, and with Prof. Eyal Reingold at the University of Toronto, who is one of the founders of the eye-tracking company SR Research.
Rick Lahaye is the founder of Kennisstroom, a Dutch based company focused on knowledge flow management. To stay connected with science, he uses his knowledge as a chess player (Elo 2380) and coach to consult with scientific researchers. At the same time, he investigates the strategies and beliefs Olympic Gold Medalists use(d) to win, and moreover, to deal with extreme fatigue near the end of a race (including pacing strategies, personality, beliefs, coping strategies, culture, etc.).
Bilalic, M., McLeod, P., & Gobet, F. (2008a). Inflexibility of experts – reality or myth? Quantifying the Einstellung effect in chess masters. Cognitive Psychology, 56(2), 73–102. doi:10.1016/j.cogpsych.2007.02.001
Bilalic, M., McLeod, P., & Gobet, F. (2008b). Why good thoughts block better ones: The mechanism of the pernicious Einstellung (set) effect. Cognition, 108(3), 652–661. doi:10.1016/j.cognition.2008.05.005
Bilalic, M., McLeod, P., & Gobet, F. (2010). The mechanism of the Einstellung (set) effect: A pervasive source of cognitive bias. Current Directions in Psychological Science, 19(2), 111–115. doi:10.1177/0963721410363571
Luchins, A. S. (1942). Mechanization in problem solving – The effect of Einstellung. Psychological Monographs, 54(6), i–95.
Sheridan, H., & Reingold, E. M. (2013). The mechanisms and boundary conditions of the Einstellung effect in chess: Evidence from eye movements. PloS One, 8(10), e75796. doi:10.1371/journal.pone.007579.
