CONFLICTING EVIDENCE OF EXPLICIT AND IMPLICIT KNOWLEDGE FROM OBJECTIVE AND SUBJECTIVE MEASURES

Abstract This study is the first to compare objective and subjective measures of explicit and implicit knowledge under learning from incidental exposure. An experiment was conducted, during which L1 English speakers were trained on a semiartificial language, Japlish. A measure of explicit knowledge and a recently proposed measure of implicit knowledge (i.e., an untimed auditory grammaticality judgment and a word-monitoring task) were applied to gauge the two types of knowledge at two testing sessions, and their results were compared with those of subjective measures of awareness. Results revealed clear discrepancies between the two measurement approaches in terms of their sensitivity. In particular, while the subjective measures varied in identifying explicit and implicit knowledge of various Japlish constructions, the objective measures indicated that most of the knowledge was explicit, and development of implicit knowledge (measured by the word-monitoring task) was minimal, only manifested in detecting a case-missing violation at the delayed posttest. The results are discussed with reference to the current literature on explicit and implicit learning and knowledge, and it is concluded that the criterion of (un)awareness might not be by itself sufficient to provide a full account of L2 knowledge developed under incidental conditions.