To search for possible cultural biases in the dataset of chess moves and games, the researchers used mathematical models to describe patterns that correspond to each kind of bias. "The population of games in the following year is produced by players picking moves from the previous year to play in their own games." "We used a population genetics model that treats all chess games played in a year as a population," Lappo said. The researchers considered chess matches of master-level players between 19, millions of which have been digitized and are publicly available for analysis by enthusiasts. The opening lines of master and grandmaster (top-level) players are often memorized by other players for use in their own games. There are few move options in the opening (beginning) of a chess game, and players often stick to tried-and-true sequences of moves, called lines, which are frequently given names like Ruy Lopez and the Frankenstein-Dracula Variation. The player with the white pieces makes the first move, each piece type (e.g., knight, pawn) moves a specific way, and (except for a special move called castling) each player moves one piece each turn. In chess, two players take turns moving white (player 1) and black (player 2) pieces on a board checkered with 64 positions. Games are won by visualizing the future positions of pieces, and players develop this skill by studying the moves made by top chess players in different situations.įortunately for chess players (and researchers), the moves and game outcomes of top-level chess matches are recorded in books and, more recently, online chess databases. Yet simply knowing the present location of all pieces won't win a chess game. In a large database of chess games by master-level players, the players' biases can change over time, and that makes chess an ideal subject to use to explore cultural evolution."Ĭhess is often called a game of perfect information because all pieces and their positions are clearly visible to both players. "There didn't seem to be anything rational about this choice. "In the mid-century players eschewed the Queen's Gambit," Feldman said. Any deviations from this random choice are known in the field of cultural evolution as cultural biases." "The baseline is to choose a move randomly among the moves played recently by other expert players. "The thesis of the paper is that when an expert player makes a move, many factors could influence move choice," Rosenberg said. "Even if a move obviously led to a win, if it could be interpreted as cowardly, the player would reject it. "In the 18th century, players subscribed to a knightly sort of behavior," said Egor Lappo, lead author and a graduate student in Rosenberg's lab. "Over the last several hundred years, paintings of chess playing show a change from crowded disorganized scenes to the quiet concentration we associate with the game today," said Noah Rosenberg, the Stanford Professor in Population Genetics and Society in H&S. The way chess is played has evolved over time too. In the past, many studies of cultural evolution were theoretical because large datasets of cultural behavior didn't exist. "Most biases are acquired from our parents or learned from our teachers, peers, or relatives."įeldman, a professor of biology, co-founded the field of cultural evolution 50 years ago with the late Luca Cavalli-Sforza, professor of genetics at Stanford School of Medicine, as a framework for studying changes in human behavior that can be learned and transmitted between people. and Mildred Finley Wohlford Professor in the Stanford School of Humanities and Sciences and senior author. "We are all subject to biases," said Marcus Feldman, the Burnet C. 15 in the Proceedings of the Royal Society B: Biological Sciences. The study summarizing their results was published Nov. Specifically, they found evidence of players copying winning moves (success bias), choosing atypical moves (anti-conformity bias), and copying moves by celebrity players (prestige bias). The researchers' analysis of chess games revealed three types of biases described by the field of cultural evolution, which uses ideas from biology to explain how behaviors are passed from person to person. Now, a new Stanford study that used population and statistical models to analyze the frequency of specific moves in 3.45 million chess games helps reveal the factors that influence chess players' decisions.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |