.png){kind=icon}
.png){kind=icon}
Return
Exploring implicit, automatized, and explicit knowledge: Testing competing CFA models
.png){kind=icon}
PRE
AI
DOI：10.1016/j.rmal.2026.100303.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The construct validity of implicit and explicit measures remains debated. Grammaticality judgment tests (GJTs) and elicited imitation (EI), assumed to tap into implicit knowledge, have been criticized for involving explicit processes; additionally, the notion of automatized explicit knowledge has further blurred construct boundaries. This study examined the validity of word monitoring task (WMT), self-paced task (SPT), EI, timed and untimed GJTs, and metalinguistic knowledge test (MKT) in advanced EFL learners (N=48) using confirmatory factor-analytic models. The two-factor model, distinguishing psycholinguistic tasks as implicit measures and form-focused tasks as explicit measures, provided the best fit. The second-order model also fit well, suggesting that EI and timed GJT reflect automatized explicit knowledge distinct from metalinguistic knowledge (untimed GJT, MKT), though both align along a single explicit knowledge continuum. However, the three-factor model failed to establish automatized knowledge as an independent construct due to near-perfect correlations with explicit measures. Overall, the findings challenge the classification of EI and timed GJT as implicit measures, highlight the value of second-order modeling, and underscore the importance of real-time psycholinguistic tasks for capturing implicit knowledge.
Keywords:
Implicit knowledge
Explicit knowledge
Automatized knowledge
Construct validity
CFA
Journal
R
IF:
0
Papers:
58
Citations:
0