CogWorks Laboratory

11/28/2020

Rensselaer research team studying Tetris players to better understand expertise

05/14/2020

Hey all! Just a short update. We are still alive and still researching even though we are not still in our lab!!!!

Thanks to efforts by the entire crew we are within striking distance of collecting data online and plan to use our new powers to expand data collection into ``cognitive task battery'' perceptual-motor decision-making tasks as well as more Tetris data!

10/11/2019

See what they wrote about us on the Science Friday page!

What observing champion Tetris players can teach researchers about how the brain makes snap decisions.

10/11/2019

I did a live segment on Ira Flatow's NPR show, Science Friday today (Friday 2019.10.11). Ira asks great questions and gave me some great opportunities to talk about ``why we collect data on Tetris players when we say we are really interested in the extremes of human expertise''!!!!

05/03/2019

Come see our symposium at this year's Cognitive Science Conference in Montreal -- July 25-27th. A 2-page symposium PDF can be downloaded from here: http://homepages.rpi.edu/~grayw/pubs/.

: The documents below are provided here as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronic...

12/26/2018

Latest paper from the CogWorks Lab-- available to all until Jan 29th via this link from Elsevier

https://authors.elsevier.com/c/1YCRD2HxoHLnK

Lindstedt and Gray are pleased to announce the latest paper from the CogWorks Lab's project on Extreme Expertise in Dynamic Decision Making. The paper is on early view for the Cognitive Psychology journal and can be downloaded NOW via the link above.

Tetris is a complex task notable for the increasingly substantial demands it makes on perception, decision-making, and action as the game is played. T…

11/16/2018

Just published!

We are very pleased to inform you that our latest research paper has just been published in ``Computational Brain & Behavior’’ (the newest journal of the Society for Mathematical Psychology) as:

Sibert, C. and Gray, W. D. (2018). The Tortoise and the Hare: Understanding the influence of sequence length and variability on decision making in skilled performance. Computational Brain & Behavior, pages 1–13. DOI: 10.1007/s42113-018-0014-4

As you may know, our recent work has focused on extreme expertise and data collection at the annual Classic Tetris World Championships. Compared with the 450 hrs of data we have from 450 Rensselaer students who have played an hour each of Tetris in our lab, these champions have a larger repertoire of “movements” an expanded sense of “dynamic decision making” that allows the best of them to make placement decisions based on (a) their mid-term goals (i.e., what they hope to achieve in the next, say 10 zoid placements), (b) the info they have available concerning the “current” and the “next” zoid, and (c) the current state of the Tetris board.

A specific focus of this work is looking for “blunders” made by these experts as well as “blunder recovery”. To remind you, at level 0 of Tetris it takes a zoid 16s to drop 20 lines from top to the bottom of display. At level 16 that drip takes 1 (one!!) second, at level 19 2/3’s of a sec, and at level 29 1/3 of a sec!! At these speeded rates during tournament play the best players execute moves that novices are incapable of making, make blunders, and often but not always recover from these blunders.

Many thanks for your following this project!

11/16/2018

We are very pleased to inform you that our latest ONR sponsored research paper has just been published in ``Computational Brain & Behavior’’ (the newest journal of the Society for Mathematical Psychology) as:

Sibert, C. and Gray, W. D. (2018). The Tortoise and the Hare: Understanding the influence of sequence length and variability on decision making in skilled performance. Computational Brain & Behavior, pages 1–13. DOI: 10.1007/s42113-018-0014-4

This paper is available on my Faculty Website --

http://homepages.rpi.edu/~grayw/pubs/

This work is Catherine’s second 1st author publication on “machine learning approaches” for understanding a complex, cognitive/perceptual/decision-making task. Her first was published last year as:

Sibert, C., Gray, W. D., and Lindstedt, J. K. (2017). Interrogating feature learning models to discover insights into the development of human expertise in a real-time, dynamic decision- making task. Topics in Cognitive Science, 9(2):374–394. DOI: 10.1111/tops.12225

Both papers are part of her PhD dissertation which takes a cognitive modeling/machine learning approach towards understanding, what is for humans, a complex, dynamic, real-time decision making task.

As you may know, our recent work has focused on extreme expertise and data collection at the annual Classic Tetris World Championships. Compared with the 450 hrs of data we have from 450 Rensselaer students who have played an hour each of Tetris in our lab, these champions have a larger repertoire of “movements” an expanded sense of “dynamic decision making” that allows the best of them to make placement decisions based on (a) their mid-term goals (i.e., what they hope to achieve in the next, say 10 zoid placements), (b) the info they have available concerning the “current” and the “next” zoid, and (c) the current state of the Tetris board.

A specific focus of this work is looking for “blunders” made by these experts as well as “blunder recovery”. To remind you, at level 0 of Tetris it takes a zoid 16s to drop 20 lines from top to the bottom of display. At level 16 that drip takes 1 (one!!) second, at level 19 2/3’s of a sec, and at level 29 1/3 of a sec!! At these speeded rates during tournament play the best players execute moves that novices are incapable of making, make blunders, and often but not always recover from these blunders.

Many thanks for your following this project! We hope you enjoy the paper.

Wayne & Catherine

: The documents below are provided here as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronic...

11/06/2018

CWL at the Classic Tetris World Championships. Photo 1: Cogworker Catherine calibration the eye tracker for one of the finalists. Photo 2: Drs. Gray and Berry discuss the schedule.

10/24/2018

Hey all - check out Jackie's post -- which somehow was added to the Visitor Posts rather than to here!

10/07/2018

Now in press: Sibert, C. and Gray, W. D. (2018, in press). The Tortoise and the Hare: Understanding the influence of sequence length and variability on decision making in skilled performance. Computational Brain & Behavior.

Computational Brain & Behavior is the newest journal of the Society for Mathematical Society. We are pleased to have this paper accepted by that journal.

ABSTRACT below -- we will post links to the publisher's website once the paper is published:

Tetris is a complex task which taps into several human skills; among them perceptual learning, planning, motor skills, and sequential decision-making. Following a divide-and-conquer strategy, we adopt a machine modeling approach to isolate the contribution of sequential decision-making from the other three skills. In two studies, we test three sets of 1,771,561 feature-based machine players (MPs) (11^6, 11 weights for each of 6 features) of Tetris for both long-running (Tortoise) and short-running (Hare) MPs. Tortoise models run until they die. Hare models are stopped after 506 episodes. For both studies we select the longest running Tortoise model and compare its score and behavior with that of the best scoring Hare model. The best Tortoise models adopt an Endurance Strategy which emphasizes single-line over multi-line clears. The best Hare models adopt an Escalation Strategy which stresses multi-line clears. In contrast, our human players tend to adopt the Escalation Strategy early in their game but switch to the Endurance Strategy as speed demands increase. Unexpectedly, across three model runs, each with a different random seed, we obtain three different sets of “best fitting” models; that is, the MPs overfit the data even though that data is generated by an essentially infinite random sequence. However, in each model run, the best Tortoise adopted the endurance strategy and the best Hare adopted escalation.

09/25/2018

CogWorks returns to the Classic Tetris World Championships (CTWC) in Portland, Oregon this October 20-21st. We are looking forward to seeing old friends and meeting new ones. Be sure to stop by our booth, say hello, and play some Tetris with us!