Analogy Generation in Chess

Study Overview: Chess expertise effects on analogical reasoning through analogy generation tasks Participants: 67 total (32 advanced chess players, 35 novices) Core Data Structure: 10 chess scenarios per participant (670 total scenario responses) Participant-level data: Chess expertise classification, demographic information, exit survey data Response-level data: Generated analogical chess positions, source attribution, subjective ratings, judge rating Key Variables Collected: Analogy Generation Performance Success rate in generating analogies (binary: successful/unsuccessful) Total analogies generated per participant Analogy Source Classification Memory-based vs. fantasy-created analogies (categorical) Participant self-reported source attribution Perceptual Similarity Measures Pixel-by-pixel overlap scores (continuous, 0-1 scale) Piece count in generated positions (integer) Non-empty pixel-by-pixel overlap (continuous, 0-1 scale) Subjective Assessments Confidence ratings (likely Likert scale) Satisfaction ratings (likely Likert scale) AI-Derived Measures Structural similarity scores from multiple AI models Second-order analogical mapping assessments