Chess Survey - Mental Speed Study


Hello dear chess players!

My name is Arman and I am an undergraduate student majoring in Psychology, Physics, and Neurobiology at the University of California, Berkeley. I am conducting a study on mental speed as it relates to chess skill and experience, specifically by looking at chess ratings, time played, and Puzzle Rush scores. Studying mental speed in the context of chess can shed some light on how knowledge becomes internalized in the human mind. 

If you have time, I would really appreciate it if you could take the survey below, and help me by being a part of this research and study. The survey should not take more than 10 minutes of your time and your responses are completely anonymous.

Thank you in advance and here’s the link:


There are some flaws in your study design you should probably be aware of. For instance, I've played ~500+ games in rapid, but only 10 blitz games; my blitz rating is low because I just don't ever play that time format. And I've been doing a ton of puzzles lately, but never in the 5 minute challenge mode, so my personal best there is also quite low relative to my true ability. Your dataset will be ruined by data that is not truly representative of study participants' abilities. For blitz rating particularly, you should try to control for this error by also collecting data on the total # of blitz games played by a study participant, and excluding participants that don't have a sufficient number of games to have a reliable rating. Same applies to puzzles. Glicko RD might also be useful since a high Glicko is associated with greater uncertainty in ELO, and as it narrows ELO of any player becomes more reliable measure of ability. I did not complete the survey because I had no way to go back a page to correct a wrong entry on a previous page. Lastly, you should differentiate between hours of casual play vs spent actually studying theory/opening prep/tactics etc., as a person with elo of 800 who spends 20 hours a week mindlessly messing around probably is also going to skew your data compared to an 800 who is actively studying the principles of chess and not just playing casually for entertainment.

Other information that might be interesting to collect - SAT/ACT/MCAT/GPA, other metrics of intellectual pursuit, and other activities a person might engage in that have crossover to chess (poker, sudoku, other problem-solving things, hours of video game play, or years of video game play).

Also, handedness is something you might want to collect data about - as a neurobio/psych major you should know that left-handed people rep ~10% of the population and have less lateralized brains than right-handed people, which gives them some interesting performance advantage with mental tasks, spatial reasoning, etc. that likely have carryover into mental speed in chess.

Good luck with your study, try to reduce the confounding and ensure your data is representative of what you're trying to observe.


 JA Smith, MD, BA MCB c/o 2015


First of all, thank you for taking the time to write up such a thoughtful response, I really appreciate it. I just want to mention that this survey is essentially the first phase of this study.
To address some of the points you raised: I was aware of the Blitz vs. Rapid ratings issue, and decided to use only Blitz ratings for two reasons: firstly, that Blitz is more popular than Rapid, and therefore any chess player who takes the survey is likely to have more Blitz games played; secondly, because Blitz games (usually 5 min. or 3+2 min) are closer to the 5-min Puzzle Rush aspect. Furthermore, I designed the survey to be as relevant as possible to any given chess player, and not particular cases like the ones you mentioned.

Nonetheless, I will keep all of your suggestions and advice in mind as I work on the next stages. Once again, thank you so much for your participation!