Date Published: June 13, 2019
Publisher: Public Library of Science
Author(s): Joseph J. Thompson, Caitlyn M. McColeman, Mark R. Blair, Andrew J. Henrey, Bryan C. Daniels.
In tasks that demand rapid performance, actions must be executed as efficiently as possible. Theories of expert motor performance such as the motor chunking framework suggest that efficiency is supported by automatization, where many serial actions are automatized into smaller chunks, or groups of commonly co-occuring actions. We use the fast-paced, professional eSport StarCraft 2 as a test case of the explanatory power of the motor chunking framework and assess the importance of chunks in explaining expert performance. To do so, we test three predictions motivated by a simple motor chunking framework. (1) StarCraft 2 players should exhibit an increasing number of chunks with expertise. (2) The proportion of actions falling within a chunk should increase with skill. (3) Chunks should be faster than non-chunks containing the same atomic behaviours. Although our findings support the existence of chunks, they also highlight two problems for existing accounts of rapid motor execution and expert performance. First, while better players do use more chunks, the proportion of actions within a chunks is stable across expertise and expert sequences are generally more varied (the diversity problem). Secondly, chunks, which are supposed to enjoy the most extreme automatization, appear to save little or no time overall (the time savings problem). Instead, the most parsimonious description of our latency analysis is that players become faster overall regardless of chunking.
Performance timings of motor behaviour are suggestive of higher level processes that control entire sequences (i.e., ‘chunks’, e.g., [1–3]). These chunks have often be used as an explanation for performance improvements during learning. For example learning curves have been explained in terms of the acquisition of these chunks by several researchers [4–7]. Critically, chunked sequences are advantageous for performance because chunks are executed quickly , and because automatization frees up cognitive resources for higher-level processing [9, 10].
We began by ensuring that our data contain expertise related changes in performance speeds. Fig 2 shows decreases in average time taken to complete an action for each of the 8 skill levels. The differences between skill levels that are clear in the figure were confirmed by statistical analysis. The nuisance factor ‘player species’ is added to our ANOVA models as StarCraft 2 players must select their species prior to play, and this choice impacts some game mechanics and, ultimately, a player’s average action latency (F(5,3178) = 126.65; p<2e-16; ηp2 = 0.17; for additional details, see S1 Text). Most importantly, there was a significant main effect of league (F(7,3178) = 421.49; p< 2.2e-16; ηp2 = 0.481) on action latency, confirming that speed does indeed change with expertise (also see Fig 2). The Motor Chunking Framework claims that chunks save time: actions are performed in a learned sequence and so do not require individual planning, thus, actions after the initial action are executed faster. As illustrated in Fig 1, the MCF explanation for expert performance involves the conversion of an increasing proportion of individual actions into timesaving chunks. We sought to test these predictions about the proportion, diversity, and speed of chunks using replay data from the real-time strategy game StarCraft 2, collected from players across eight skill levels. Source: http://doi.org/10.1371/journal.pone.0218251