Junya FUKUTA

In recent years, insights from cognitive psychology on unconscious learning have been applied to second language acquisition (SLA) research, proposing that unconscious knowledge gradually becomes conscious through repetitive experience. This perspective offers a reinterpretation of the interface between implicit and explicit knowledge in SLA, contrasting with the traditional view that explicit knowledge gives rise to implicit knowledge. While the phenomenon of unconscious knowledge becoming conscious provides valuable insights into the dynamics of knowledge acquisition, it also suggests the need to reconceptualize the traditional notions of implicit and explicit knowledge. This article aims to propose a framework that integrates recent insights on unconscious learning with traditional claims in SLA. Specifically, we propose replacing the unidimensional linear continuum view – which positions implicit and explicit knowledge as two ends of a single axis and conceptualizes second language development as a back-and-forth transition between these poles – with a multilayered knowledge view: The Hierarchical-Layer Model of Conscious–Unconscious Knowledge (H-L Model). This alternative model serves as a working hypothesis to better capture the multilayered and differentiated nature of linguistic cognition.

Abstract This study examines when and how second language (L2) learners begin to exhibit sensitivity to key factors influencing the choice between the English double object and prepositional object constructions. While previous research has shown that such choices in native speakers are influenced by such factors as animacy, pronominality and verb bias, little is known about the developmental timing of these effects in L2 production. Using 5,785 dative constructions from a large-scale learner corpus, we analyzed how these variables interact with learners’ proficiency levels across 23 verbs. We found that learners showed systematic sensitivity to all of these factors, including statistical verb bias derived from a native speaker corpus (Corpus of Contemporary American English), at much earlier stages than previously suggested. These results suggest that learners may possess a cognitive bias that maps preexisting conceptual structures onto linguistic constructions, reflecting more than mere statistical learning.

Abstract This study, based on Jiang’s (2000. Lexical representation and development in a second language. Applied Linguistics 21(1). 47–77) bilingual lexicon model, investigates the learnability of collocations among 34 Japanese EFL learners and examines the influence of their L1 on such learning. An acceptability judgment task assessed the knowledge of three different types of collocations: English-only collocations that cannot be directly translated into Japanese (e.g., flat rate); congruent collocations that can be translated into Japanese without changing their meaning (e.g., cold tea); and Japanized collocations that are infelicitous in English, but felicitous in Japanese (e.g., yellow voice). After the task, participants translated the collocations and rated the difficulty on a four-point Likert scale. The relationship between the accuracy of these collocations and the translation difficulty rating scores was analyzed using mixed-effects logistic regression modeling to assess L1 influence. The results showed that with increasing L2 proficiency, learners tend to regard congruent and English-only collocations as acceptable, but even highly proficient learners did not fully reject the Japanized collocations. This suggests that as L2 proficiency increased, participants learned to accept felicitous collocations but did not learn to reject Japanized infelicitous ones. In addition, the influence of L1 was evident in English-only and Japanized collocations and could not be avoided by those with increasing proficiency.

Abstract This article addresses the pitfalls of performance analysis in investigating cognitive processing during second language (L2) learning. The problems that we discuss in this paper are twofold: (1) Assuming psychological variables to be ontological entities without meeting the criteria for ontological reality and (2) Inappropriateness of assessing abilities based on learner’s speaking or writing performance to investigate cognitive processes. By addressing these problems, we argue that some latent variables postulated by observing L2 performance do not exist in reality and emphasize the difficulty of interpreting cognitive mechanisms through performance analysis. We also enumerate some problems that arise from the epistemological perspectives of previous research practice (e.g., the bifurcation of contradictory hypotheses and their indeterminacy). Finally, two alternative approaches treating L2 performance are proposed. The implications of this line of discussion for future research are also discussed.

This study investigates how implicit and explicit learning and knowledge are associated, by focusing on the salience of target form–meaning connections. The participants were engaged in incidental learning of artificial determiner systems that included grammatical rules of [± plural] (a taught rule), [± actor] (a more salient hidden rule), and [± animate] (a less salient hidden rule). They completed immediate and delayed post-tests by means of a two-alternative forced-choice task with subjective judgments of source attributions. Awareness during the learning phase was identified through analysis of thinking aloud protocols. The results did not support a one-to-one relation between either explicit learning and conscious knowledge, or implicit learning and unconscious knowledge; rather, they indicated that implicit and explicit learning are intricately linked to conscious and unconscious knowledge mediated by the salience of form–meaning connections in target items. This result also suggests the possibility of the later emergence of knowledge without any conscious awareness of it.

Abstract Although native speakers (NSs) of English make plural agreement in preverbal-subject sentences (e.g., A pen and eraser *is/are…), previous studies have demonstrated that they prefer singular – not plural – agreement between verbs and conjoined noun phrases (NPs) in expletive there constructions (e.g., there is/are a pen and an eraser…), showing efficiency-driven processing prioritization of agreement between nearest constituents. This paper assesses whether Japanese L2 learners of English (JLE) show this tendency. The results of two self-paced reading experiments together indicated that even though efficiency-driven processing was available to L2 learners, their use was unstable due to the repeated exposure to there are NP_pl- and NP_pl-type sentences during the task. It seems possible that repeated exposure triggered learners’ knowledge that conjoined NPs are always plural. Hence, it could conceivably be hypothesized that a learner’s specific knowledge intervenes the efficiency-driven processing strategy.

ABSTRACT This study aimed to investigate how Japanese learners of English as a foreign language, whose first language does not have obligatory morphological number marking, process conceptual plurality. The targeted structure was reciprocal verbs, which require conceptual plurality to interpret their meanings correctly. The results of a sentence completion task confirmed that participants could use reciprocal verbs reciprocally in English. In a self-paced reading experiment, participants read sentences with reciprocal verbs and those with optionally transitive verbs (e.g., while the king and the queen kissed/left the baby read the book in the bed). There was no reading time delay for reciprocal verbs but a delay for optionally transitive verbs. Therefore, the participants succeeded in processing second language conceptual plurality in the online sentence comprehension task.

WritingMaertiX (WMX), which we have developed and will report about in this paper, is a versatile and easy-to-use multi-function software package for recording, visualizing, and analyzing learners’ writing process. WMX applies keystroke logging to gather all information concerning to real-time writing processes of writers, and can literally “replay” the writing process by reusing the recorded keystroke information. Additionally, WMX is endowed with a sophisticated function for analyzing the writing process by automatically calculating the number of words over time and the frequencies of revisions. These time series data can also be visualized using WMX. We have also provided alternative tools for analyzing the writing process data such as (a) a semi-automatic tool for calculation of the number of words over time, (b) an original function for R (a statistical tool), which also automatically calculates the number of words over time. This paper introduces the background and the rationale for developing WMX, and its specifications. We also report the results of some surveys using WMX, demonstrating how it works in reality. We believe that our new tool, WMX, which is freely available to every researcher, teaching practitioner, and even every autonomous learner who is interested in real-time writing processes, will evoke improvements in foreign language writing research, teaching, and learning.

The priming method was recently introduced into the field of L2 studies as a new psycholinguistic tool to capture how and to what extent L2 learners possess their implicit knowledge or proceduralize their knowledge of specific grammatical structures. However, this line of studies exclusively explores the productive dimension, but not the receptive dimension, which is said to be better to obtain a deeper insight into the implicit knowledge (Jiang, 2007). In consideration of this, the current study investigated whether L2 learners can utilize syntactically driven algorithmic processing with implicit knowledge through a self-paced reading task with syntactic priming. Twenty highly proficient Japanese learners and 18 native speakers of English engaged in a phrase-by-phrase self-paced reading task after reading prime sentences. The results indicate that: (1) Japanese EFL learners seem to be able to utilize syntactic processing with implicit knowledge in terms of dative alternation; and (2) learners process sentences more "deeply" than natives under certain conditions. The results are discussed in terms of semantic/heuristic processing (good-enough processing) and syntactic/algorithm processing.

in press