Distinct contributions of expertise and automatization to higher dual-task performance: A cognitive–motor dual-task study in pianists

Abstract Dual-task studies consistently find that experts are better at performing two tasks at the same time than non-experts. Usually, this effect is explained by the high degree of skill automatization in experts, given that skills become increasingly automatized during later stages of learning. However, it is not yet clear whether expertise and task automatization following extensive practice exhibit identical or independent mechanisms for reducing dual-task costs. In this study, we investigated how expertise, the learning stage, and higher cognitive abilities contribute to motor–cognitive dual-task interference. Thirty-eight conservatory-level pianists practiced two novel piano sequences. Each sequence involved the same practice time for all participants (i.e., 20 trials) for which individual learning rates—proxies of the learning stages by the 20 th trial—were computed. Subsequently, they performed the sequences either at maximal speed or to a fixed metronome tempo while concurrently solving a mental arithmetic task (counting in steps of 7). Using Bayesian hierarchical modeling, dual-task costs were estimated as the change in piano performance between single- and dual-task performances while statistically controlling for cognitive dual-task costs. Results showed that higher piano expertise and later learning stages robustly reduced dual-task costs. Further, higher piano single-task performance, successful memorization of the piano sequence and better cognitive abilities, such as processing speed and switching ability, were also found to reduce dual-task costs. Finally, mediation analyses indicated that expertise and task automatization had largely independent effects on dual-task costs, suggesting distinct mechanisms. Therefore, future studies should avoid equating task automatization with expertise when interpreting dual-task costs.