The building blocks of expertise : examining extreme experts in the video game “Tetris”

Authors
Lindstedt, John K.
ORCID
Loading...
Thumbnail Image
Other Contributors
Gray, Wayne D., 1950-
Schoelles, Michael J.
Fajen, Brett R.
Ericsson, K. Anders (Karl Anders), 1947-
Issue Date
2017-08
Keywords
Cognitive science
Degree
PhD
Terms of Use
Attribution-NonCommercial-NoDerivs 3.0 United States
This electronic version is a licensed copy owned by Rensselaer Polytechnic Institute, Troy, NY. Copyright of original work retained by author.
Full Citation
Abstract
I present three related studies which examine the features and theoretical structures associated with extreme expertise in the complex, real-time cognitive task of Tetris. To facilitate this research, I developed a platform for experimental research in Tetris, Meta-T, which allows for careful ex- perimenter manipulation and offers high-fidelity behavioral performance logging. My approach “zooms in on and unfolds” high-fidelity, rich human performance datasets to improve the reso- lution of the portrait of human expertise in complex, real-time tasks in 3 studies. Study 1 used principal component analysis (PCA) to explore the behavioral features of undergraduate Tetris players in the laboratory and identify a set of features associated with a player skill: latent skill component (I named the “decide-move-placed” component), composed primarily of measures of speed and efficiency. This component reliably distinguishes player skill across the spectrum of expertise, under different conditions of time pressure, and even succeeds when examining only a small portion of player data without knowledge of game outcomes. Study 2 validates the model of player skill constructed in Study 1 in two ways: first, by predicting the game scores of Tetris tournament players; and second, by predicting the outcome of individual tournament matches at a rate well above chance. Study 3 examines world champion Tetris player performance in conjunc- tion with local tournament players to compare the novice and expert player strategies, highlight their consequences for performance, and illuminate the cognitive underpinnings of such strategies.
I find that even world champion Tetris players remain impacted by Hick’s law, indicating there may be some lower bound, beyond which the process of expertise acquisition can no longer adapt to and reduce a task’s decision space. These studies together lead me to four conclusions: (1) player skill can be reliably predicted by examining the efficiency of execution; (2) measures of process appear to more reliably predict real-world expertise than measures of individual out- comes; (3) “Hick’s law bends with practice, except when it cannot”; and (4) “both task and strategy together define the mold into which human expertise may shape itself.”
Description
August 2017
School of Humanities, Arts, and Social Sciences
Department
Dept. of Cognitive Science
Publisher
Rensselaer Polytechnic Institute, Troy, NY
Relationships
Rensselaer Theses and Dissertations Online Collection
Access
CC BY-NC-ND. Users may download and share copies with attribution in accordance with a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License. No commercial use or derivatives are permitted without the explicit approval of the author.