A multilevel model analysis of expertise in chess across the life span.

Abstract

The authors examined longitudinal change in chess skill using a multilevel model analysis of a large database of active, elite chess players (N = 5,011). Parameters estimated from quadratic growth curves indicated that the age of peak performance occurs later in life than originally proposed and that this peak is independent of initial skill level. The findings are also consistent with the hypothesis that aging is slightly kinder to the initially more able, who show milder decline past their peak. Higher tournament activity levels predicted higher ratings overall and interacted with age in the initially more able sample, suggesting that activity had smaller effects on rating for older adults. The authors discuss implications of these findings for lifetime changes in skilled performance.