The Reliability and Validity of Procedural Memory Assessments Used in Second Language Learning

Previous research provides evidence for a role of domain-general cognitive constructs, such as attention and working memory, in second language (L2) acquisition. Evidence is also emerging regarding the role of different types of long-term memory, such as declarative and procedural memory, in L2 acquisition. However, knowledge on the role of procedural memory has been limited by the absence of research on the reliability and validity of procedural memory assessments. As such, this thesis examined the reliability and validity of procedural memory assessments used to study L2 acquisition. Participants completed three procedural memory tasks used in prior research on the role of procedural memory in L2 acquisition, along with assessments of declarative memory and IQ. Measures from each task were first analyzed to construct a final set of measures that was conceptually clear and psychometrically viable. Results from a correlation matrix and exploratory factor analysis indicated that all three assessments failed to demonstrate convergent validity, and one procedural memory assessment (the Weather Prediction Task) patterned with assessments of declarative memory, but not procedural memory. By presenting a systematic test of the measurement validity of procedural memory assessments, the present study shows that more work is needed to develop valid tasks before strong conclusions can be made on the role of procedural memory in L2 acquisition.