Eye Tracking for Studying Chess Strategies

Background image

SMI Eye Tracking Glasses and multi-modal data are used to analyze mental modelling techniques of expert and novice chess players within the CEEGE project.

Challenge

The Chess Expertise from Eye Gaze and Emotion (CEEGE) project was commissioned to explore the understanding of players abilities while problem solving based upon multi-modal data (such as eye patterns and facial expressions). It set out to establish more effective deep learning techniques for mental modeling and action prediction, with the result of improved modeling approaches by integrating principles of cognitive control, learning and expertise.

Solution

SMI Eye Tracking Glasses are used to record visual perception data during dyadic chess game situations – in other words, high-pressure periods during the match. Based on eye tracking data and other multi-modal feedback such as facial expressions, other? the researchers could compare different computational models to specify and measure chess expertise. Scan-path data and other information are correlated to established chessboard configurations and known solutions to chess problems.

SMI Eye Tracking

SMI Eye Tracking integrates with other data multi-modal data streams and, the first findings from that data revealed that chess experts concentrate most of the time on the main chess pieces that can make or break the game and control their attention more efficiently than novices do. Amateurs, by way of contrast, jump very frequently with their gaze from one figure to the next and look at nearly all the pieces on the board, regardless of whether they play an important role in the particular game situation.

Benefit

Multimodal user feedback provides a foundation for artificial systems that can infer a player’s emotional state, expertise, attentional focus and situational understanding. By comparing and implementing different models, CEEGE helps to specify and measure chess expertise, to detect individual impairments and offer rigorous and detailed training methods

Background

“Cognitive Interaction Technology” (CITEC) at Germany’s Bielefeld University in cooperation with the Inria Public Research Institute in Grenoble, France. The three year research project is funded by the German Research Foundation (DFG) and the Agence nationale de la recherché (ANR).

The project is set out to experimentally evaluate, refine and compare current theories of mental modeling for problem solving, as well as techniques for observing the physiological reactions of humans in various chess situations. From the data, the researchers seek to determine the degree to which a player understands the game situation based on abstract chess configurations in a dyadic match.

Benefit

Multimodal user feedback provides a foundation for artificial systems that can infer a player’s emotional state, expertise, attentional focus and situational understanding. By comparing and implementing different models, CEEGE helps to specify and measure chess expertise, to detect individual impairments and offer rigorous and detailed training methods

Customer institution Cognitive Interaction Technology” (CITEC) at Germany’s Bielefeld University in cooperation with the Inria Public Research Institute in Grenoble, France
Customer website CITEC
Institute website Bielefeld University, “Neurocognition and Action – Biomechanics”- Research Group
Partner website Inria