Barth, Marius 
ORCID: 0000-0002-3421-6665
(2018).
Measuring Implicit and Explicit Sequence Learning.
PhD thesis, Universität zu Köln.
Abstract
Sequence learning in the serial reaction time task (SRTT) is considered a key demonstration of learning that may proceed both implicitly or explicitly. However, claims about the implicit or explicit nature of the acquired knowledge were frequently based on identifying tasks with one of these two types of learning. The present studies assume that probably no task is a pure measure of implicit or explicit learning; instead, within each task, it is necessary to disentangle the contributions of both types of learning by means of measurement models of task performance.
In the first two studies, we scrutinized a measurement model that has already been used to dissociate implicit and explicit sequence knowledge: The process-dissociation (PD) approach as applied to the generation task yields separate estimates of implicit and explicit sequence knowledge that are derived from two variants of the same task. It therefore avoids many problems of previous measurement approaches, and studies utilizing this approach have yet provided the most convincing evidence in favor of dissociable types of sequence knowledge. However, the PD approach comes with its own set of critical assumptions. In two studies, we investigated the assumptions underlying the PD approach and found that they are violated in applications to sequence learning. Taking these limitations into account, studies that utilized the PD approach do not provide firm evidence for two dissociable types of learning.
In a third study, we investigated the processes that are involved in the expression of implicit and explicit sequence knowledge in the SRTT. We found that stimulus encoding, response execution, and response selection mediated the expression of implicit sequence knowledge. An involvement of response selection indicates that implicit sequence learning is mediated by representations containing both stimulus and response features. In contrast, the acquisition of explicit sequence knowledge resulted in a decision bias towards regular responses, and other effects of sequence learning disappeared in the course of training, indicating that participants switched from stimulus-based to plan-based action control.
Downloads per month over past year
Export
Actions (login required)
 |
View Item |
