Novice construction of chess memory.

Area of Science:

• Cognitive Psychology
• Expertise Research
• Human Learning

Background:

• Novice chess players' ability to recall chess positions is crucial for developing expertise.
• Understanding the cognitive mechanisms underlying skill acquisition in complex domains like chess is essential.

Purpose of the Study:

• To investigate the learning curve of novice chess players acquiring skilled recall of chess positions.
• To model and analyze the learning process using computer simulations and compare different cognitive theories.

Main Methods:

• An experiment involving two novices studying 500 chess positions over several months.
• Development and testing of two computer simulation models: neighborhood-based chunk construction and frequency-based associative processes.
• Analysis of learning curves derived from both human subjects and computational models.

Main Results:

• Novice chess players' recall improved from 16% to 40-50% after self-study.
• The learning curve showed rapid initial gains followed by a substantial decrease in learning speed after 100-150 positions.
• The frequency-based associative model provided a better fit to the observed learning data than the neighborhood-based model.

Conclusions:

• Chess learning, specifically the acquisition of chess positions, appears to be driven by associative processes.
• Common co-occurrence of pieces, alongside chess-specific features, likely guides the formation of memory chunks in chess players.
• The findings support a frequency-based associative mechanism for learning complex visual patterns in games.