Open AccessDifferent levels of statistical learning - Hidden potentials of sequence learning tasks
Author(s) -
Emese Szegedi-Hallgató,
Karolina Janacsek,
Dezső Németh
Publication year - 2019
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0221966
Subject(s) - task (project management) , sequence learning , sequence (biology) , computer science , artificial intelligence , statistical learning , machine learning , cognitive psychology , stability (learning theory) , psychology , biology , management , economics , genetics
In this paper, we reexamined the typical analysis methods of a visuomotor sequence learning task, namely the ASRT task (J. H. Howard & Howard, 1997). We pointed out that the current analysis of data could be improved by paying more attention to pre-existing biases (i.e. by eliminating artifacts by using new filters) and by introducing a new data grouping that is more in line with the task’s inherent statistical structure. These suggestions result in more types of learning scores that can be quantified and also in purer measures. Importantly, the filtering method proposed in this paper also results in higher individual variability, possibly indicating that it had been masked previously with the usual methods. The implications of our findings relate to other sequence learning tasks as well, and opens up opportunities to study different types of implicit learning phenomena.