Tuesday, September 8, 2015

Physics of Chess?

This week's reading focused on two roughly analogous experiments. In Chase Simon (1973), the authors sought to assess the methods of a master, an amateur, and a beginner in recreating a chess board both from memory and while side by side. In Che et al (1981), the authors examined the ways experts and novices sorted various physics problems. Both studies offer valuable insights into how a subject's thought process relates to their level of expertise in the given field. In both studies, the depth of understanding the subject had of the topic correlated with an apparently stronger ability to recognize core patterns while novices seemed to focus on the more superficial surface image.

In Chase, the authors studied in particular the idea of chunking, essentially attempting to determine the subject's method of sorting the pieces either in their memory (as in the memory task) or physically on the board (as in the perception task) into different groups. They found that experts were able to remember larger chunks, especially when the board the pieces were in positions characteristic of an average chess game. Furthermore, they found experts tended to remember, or chunk, pieces of importance, such as in attacking or defending positions, while novices tended to chunk based more on more superficial characteristics, such as color, type, or relative location of the piece. Beyond simply showing that experts can remember more of the game board than novices, this experiment suggests that experts view the game on a much deeper level than a novice.

A similar trend is suggested in the Che paper. When experts and novices in the field of physics were given the task of sorting various problems into different subtopics, experts tended to group problems based on type while the novices tended to sort problems based more on what they called surface characteristics, such as objects and physical terms used. For example a physics professor might group together a ball falling straight down with a car at constant acceleration (kinematics), while an undergrad might group the falling ball with a ball being swung around on a string, even though the second problem involves a different skill set than the first. This study shows what I think is a very crucial part of not just physics but problem solving in general: Seeing a problem at its core is far more important than superficial details. Understanding what a problem is really asking is half the battle. After that it's usually plug and chug or looking up a table of values.

Perhaps an important conclusion with regards to teaching is to focus importance not on discrete problems, but more on the concrete, consistent principles that unite them and how to recognize these principles in an applied problem.


  1. I didn't understand the bit in the chess paper regarding the attacking and defending positions and so effectively missed the comparison between the types of chunks the experts see versus the novices. It makes sense, since I am very novice at chess, but it also provides another tie-in between the papers I missed. So then, with regards to teaching, how would we go about focusing importance on recognizing the principles behind a problem rather than simply solving the problem itself? How can we lead students so that they start doing this independently?

    1. I don't know a whole lot about chess either, but as far as I understand attacking piece might be a bishop moving in to threaten the king (or some other piece), where a defending piece might be a pawn moving in to block the bishop. The connection lies in recognizing the significance of the pieces themselves as they affect the game as a whole. While a novice might only recognize what and where the piece is, an expert might recognize why the piece is there, picking out familiar layouts of the board. This is similar to the physics experiments where experts sorted problems based on the core concepts of the problem (analogous to the significance of the piece's type location in the grand scheme of the game), while the novice focused on only the superficial (only the type and location of the piece, not the significance).

      I honestly do not know how I would try and teach principles more than problem solving. One thing that has helped me in physics has been to take a couple minutes to study the problem before diving in. That way I can sift out irrelevant details, recognize possible core concepts, and hopefully form a plan of attack to solve the problem. Hopefully we can discuss this more during class.

  2. This comment has been removed by the author.

  3. I definitely had to comment on your post since I thought I was original coining the term "Physics of Chess" in my own blog post. The first time around reading the paper I think I actually missed the bit about chess experts better perceiving the attacking/defending positions of the pieces. I think that is telling in that this is a call to maybe what the pieces are ontologically--the position or color of a piece is not intrinsic to the nature or 'principle' of the piece--really, what is the purpose of a chess piece? Its sole intent is to capture other pieces and move towards the objective of the gamer (capturing the King). Our experts focus much closer in on the principle and 'essence' of the pieces rather than the surface appearances. Wow. I guess these articles were much more similar than I first thought.