Language Learning最新文献

Hu and Nation's (2000) study, which stipulated that second language (L2) readers need to be familiar with 98% of lexical items for adequate text comprehension, has become highly influential in L2 vocabulary research and pedagogy. However, the 98% critical threshold figure is based on findings from a research project in which a regression analysis was conducted with only 66 university students in New Zealand. The present study replicated Hu and Nation's research in a context different from a typical Western, educated, industrialized, rich, and democratic context with a sample of 104 Sri Lankan adult L2 learners in a nonacademic context. They each took a Vocabulary Levels Test and read one of five versions of two reading texts at different levels of density of unknown words before answering comprehension questions. The results of the original study could not be fully replicated.

The comprehension of signed language requires linguistic and visual–spatial processing, such as perspective-taking for correctly interpreting the layout of a spatial scene. However, little is known about how adult second-language (L2) learners process visual–spatial constructions in a signed language that they are studying, including which angles of viewing are most challenging to process and whether there are relationships between perspective-taking and the comprehension of non-spatial (i.e., non-scene based) constructions. We examine the performance of 95 intermediate signers of American Sign Language (ASL) on linguistic and non-linguistic perspective-taking tests. Half the participants completed a test of narrative comprehension that included visual–spatial scenes, and half took a test of signed phonological and morphophonological discrimination. Performance on linguistic perspective-taking correlated moderately with performance on the narrative, but not with the discrimination test. These findings support the claim that perspective-taking skills are yoked to some—but not all—aspects of signed language learning.

The social sciences have grappled with sampling biases, perhaps most notably the prevalent reliance on convenience samples drawn from university student populations. Researchers in second language acquisition (SLA) have likewise taken steps to assess and address the scope of these biases and their effects on theory construction. This special issue presents a collection of replications and registered replication reports comprising one such initiative, titled SLA for All? The replications aimed to evaluate the generalizability of findings from foundational SLA studies, which were obtained with university-educated participants, to nonuniversity participant groups. In this introduction to the special issue, we review and discuss the general failure to replicate the initial results, the complex notion of replication, and questions of research ethics. We offer an in-depth reflection on how our perspectives, practices, and future plans have evolved and conclude with a vision for a more inclusive, diverse, and informative approach to SLA research going forward.

The current study investigated gender (control) and number (target) agreement processing in Northern and non-Northern Norwegians living in Northern Norway. Participants varied in exposure to Northern Norwegian (NN) dialect(s), where number marking differs from most other Norwegian dialects. In a comprehension task involving reading NN dialect writing, P600 effects for number agreement were significantly affected by NN exposure. The more exposure the NN nonnatives had, the larger the P600 was, driven by the presence of number agreement (ungrammatical in NN). In contrast, less exposure correlated to the inverse: P600 driven by the absence of number agreement (ungrammatical in most other dialects). The NN natives showed P600 driven by the presence of number agreement regardless of exposure. These findings suggests that bilectalism entails the representation of distinct mental grammars for each dialect. However, like all instances of bilingualism, bilectalism exists on a continuum whereby linguistic processing is modulated by linguistic experience.

A large number of children are exposed to more than one language. One well-established method of assessing early vocabulary development in monolingual children is parent report; however, its use in bilingual/multilingual contexts is less established and brings unique challenges. In this methodological scoping review, we reviewed studies of early vocabulary development using parent report with bilingual/multilingual children (January 1980–March 2022). A total of 576 articles were screened, yielding 101 studies for analysis. The number of studies on bilingual/multilingual vocabulary has grown in the last two decades; yet representation of the world's languages remains sparse. The majority of studies assessed bilingual/multilingual children's vocabulary in each language and used instruments adapted for linguistic and cultural characteristics. However, the field could benefit from standardized reporting practices regarding definitions of bi/multilingualism, selection of reporters, and tool development and is in critical need of studies that develop, validate, and norm parent report instruments specifically for the bilingual/multilingual case.

The present study administered six test instruments to 13- to 14-year-old learners of English in Austria and Sweden (N = 213), countries offering settings with more explicit and implicit learning environments, respectively. Confirmatory Factor Analyses for Austria yielded a factor comprising timed grammaticality judgment tests, an oral narrative test, and elicited imitation, labelled in this study Automatized and/or Implicit Knowledge, and a factor including an untimed grammaticality judgment test and a metalinguistic knowledge test, named in this study Explicit Knowledge. In the Swedish context, goodness-of-fit indices provided some evidence that a single-factor model shows a better fit, although a comparison of this model with two-factor models did not reach statistical significance. The findings point to the potential importance of considering the specificities of a learning environment in interpreting learner achievement on measures of the implicit versus explicit knowledge spectrum.

This study presents a corpus-based investigation of self-repairs in hearing adult L2 (M2L2, second modality and second language) learners of Swedish Sign Language (Svenskt teckenspråk, STS). This study analyses M2L2 learners’ STS conversations with a deaf signer and examines the learners’ self-repair practices and whether there are differences among learners of different proficiency levels. This provides a description of characteristics of self-repair made by M2L2 learners as well as the frequency and distribution of self-repair categories. The results show that the frequency of self-repair decreases with increased proficiency, at least after the initial stage. Furthermore, the self-initiated repair categories of the beginners are often phonological repairs, while intermediate learners tend to carry out self-repairs at the lexical and syntactic level. The results also reveal a specific type of STS repair linked to fingerspelling repairs. We discuss the effects of second modality learning as well as the relationship between monitoring and language proficiency.

We thank our esteemed colleagues who provided insightful commentaries on our feature article “(Why) are open research practices the future for the study of Language Learning?” (Marsden & Morgan-Short). Their responses very usefully illustrated and amplified points in our review, provided nuance and extension to some of our ideas, and pushed us to make stronger statements and deeper considerations of some of the facets and consequences of open research practices.

Three common and prominent themes seemed to emerge from the responses, which we identify as: Community; Equity, Diversity, and Inclusion; and Changing Culture, and we organize our own response around these themes. We note that some of the issues raised by our generous commentators were addressed in arguments that had originally been included in our submitted manuscript (Marsden & Morgan-Short) but, due to length considerations, had to be moved to its Appendix. That Appendix can be found in the online Supporting Information for the Marsden & Morgan-Short article and is also held on the Open Science Framework (OSF) at https://osf.io/ru5n4. We refer to some of those arguments in our response here.

This study reports on strategies to indicate plural referents in hearing learners of Sign Language of the Netherlands. This is the first explorative study that focuses on L2 expressions of plurality in a sign language. Using data from two datasets, I examined when learners start to express plural and which strategies they apply, and I noted typical learner characteristics. The first study examined spontaneous conversations of three learners, during the first 18 months of their learning. The second study analyzed elicited data from 11 learners during their first year of learning. The data reveal that learners are able to express plural referents in early stages, using strategies that are familiar to them (quantifiers) as well as strategies that do not occur in their mother tongue (reduplication of the noun, use of spatial devices). The early emergence might be explained by the salient nature of the devices and the resemblance with gestural portrayals.

This study addresses the role of awareness in learning and the variables that may facilitate adult second language (L2) implicit learning. We replicated Williams's (2005) study with a similar group of academic learners enrolled at university as well as a group of non-college-educated adults in order to explore the generalizability of the findings to an underrepresented population in research on L2 acquisition. Our results revealed that academic learners implicitly acquired items encountered during training (trained items), but this learning disappeared when academic and nonacademic groups were combined. We also observed modest correlations between intelligence and implicit learning of trained items; however, this association disappeared when other variables were considered. Overall, our study highlights the limited potential of implicit form–meaning associations for L2 adults in more general populations and emphasizes the challenges associated with convenience sampling in L2 research (Andringa & Godfroid, 2020). Additionally, it underscores the independence of individual differences in reading exposure, years of education, and nonverbal intelligence for implicit learning of trained items.