Abstract
Individual differences in memory performance in a domain of expertise have traditionally been accounted for by previously acquired chunks of knowledge and patterns. These accounts have been examined experimentally mainly in chess. The role of chunks (clusters of chess pieces recalled in rapid succession during recall of chess positions) and their relations to chess skill are, however, under debate. By introducing an independent chunk-identification technique, namely repeated-recall technique, this study identified individual chunks for particular chess players. The study not only tested chess players with increasing chess expertise, but also tested non-chess players who should not have previously acquired any chess related chunks in memory. For recall of game positions significant differences between players and non-players were found in virtually all the characteristics of chunks recalled. Size of the largest chunks also correlates with chess skill within the group of rated chess players. Further research will help us understand how these memory encodings can explain large differences in chess skill.
Highlights
Miller [1] proposed that limited storage capacity of human short term memory (STM) [2], could be expressed as seven plus/minus two chunks, where each chunk corresponded to a familiar pattern previously stored in long-term memory (LTM)
The first section compares chunks identified with the traditional RT technique with those identified with our new repeated-recall technique in terms of their reliability, validity and ability to account for overall memory performance
To assess the generalizability of the measures across different chess positions we capitalized on the fact that we presented two positions (Board 1 and Board 2) for each experimental condition
Summary
Miller [1] proposed that limited storage capacity of human short term memory (STM) [2], could be expressed as seven plus/minus two chunks, where each chunk corresponded to a familiar pattern previously stored in long-term memory (LTM). The experts’ large recall advantage did not transfer to randomly rearranged pieces in positions because the meaningful structure of the chess position was destroyed by the randomization, essentially eliminating opportunities for more skilled players to perceive familiar chunks. Consistent with their theoretical predictions experts in many other domains showed a recall advantage over less accomplished performers in the same domains, but only for representative structured stimuli (see [5, 6] for review).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
