Recently, Beheim et al.  introduced the idea of using archived records of movement patterns in the game of Go to study the cultural evolutionary dynamics of gaming strategy. Until the introduction of this approach, evidence for cultural evolution was based primarily on mathematical models [2, 3], simulation , short-term experimental evidence [5, 6, 7], and verbal arguments [8, 9]. Quantitative analysis of cultural evolution through social and prestige-biased learning in natural environments was hampered by the lack of finely-resolved, individual-level data on human behavior over long courses of time. Interestingly, strategy games such as Go and Chess have risen to such a level of cultural importance, that the project of voluntarily recording massive amounts of longitudinal data has been completed already by game aficionados. These data permit the application of annually-resolved time-series analysis to the evolutionary dynamics of game playing behavior, and allow for the effects of individual learning, social learning, and prestige-biased learning to be analyzed in a uni ed framework. This paper will examine evidence regarding the use of these learning strategies in the selection of Chess opening moves, over the period of time from 1975 to 2013, in a data set containing over 1.1 million games played by more than 54,000 professional Chess players.
Paper coming soon.