Earticle

This study examined processing of L2 English relative clauses by testing the Filler-Gap Hypothesis (Hawkins, 1999; O"Grady 1997). For that goal, we partially replicated Diessel and Tomasello (2004), who examined L1 acquisition of relative clauses by English-and German-speaking children. The Filler-Gap Hypothesis states that the structural distance between head and gap determines the processing difficulty pertained within relative clauses. Taking this hypothesis as a theoretical starting-point, we used an elicited imitation task to tap 48 L2 learners" knowledge of English relative clauses. The results of the study demonstrated that subject relatives retained greater accuracy scores than object relatives, which in turn retained greater accuracy scores than indirect object and oblique relatives. These results were largely consistent with the predictions made by the FGH, but only partially consistent with Diessel and Tomasello"s L1 data.

This study examined how L2 learners of English process morphologically complex words (e.g., bareness) and psuedowords (e.g.,*boilness) to investigate if the L2 processing of derivational morphology is subject to critical period effects. It has been claimed that native speakers rely on procedural memory to process morphologically complex words, while L2 learners tend to utilize declarative memory to process them (Ullman, 2001, 2004, 2005). According to Ullman, L2 learners’ reliance on declarative memory for processing morphologically complex words is due to maturational changes in procedural memory. Two related experiments with a masked priming lexical decision task were conducted with two groups of Korean learners: early and late L2 learners. With regard to the results, early L2 learners showed stem priming effects for morphologically related prime-target pairs, which is indicative of morphological decomposition. On the other hand, late L2 learners did not provide any evidence of stem priming, indicating that they might have processed the morphologically complex word/pseudoword as a whole unit. These data suggest that the L2 processing of late learners is different from that of early learners.

본 연구는 일본인 한국어 학습자의 한국어 피동문에 나타나는 어순 교체에 대한 습득을 살펴보는 데 목적이 있다. 한국어 피동문 어순 실현은 구조적 기준(어순 전형성)과 의미역 기준(논항 배분 전형성)에 따라 유표성이 다르게 나타나며, 어떠한 기준을 우선적 단서로 삼아 문장을 처리하는지에 따라 습득의 어려움을 예상할 수 있다. 본 연구에서는 한국어 모어화자 30명과 일본인 학습자 60명을 대상으로 하여 피동문 실시간(online) 처리를 관찰하기 위해 자기 조절 읽기 과제(self-paced reading)를 실시하였다. 그 결과, 한국어 모어화자는 의미적 단서보다 구조적 단서에 민감성을 보여 ‘피동작주-동작주’ 순서로 나타난 문장을 ‘동작주-피동작주’보다 빠르게 처리하였다. 그러나 이와 달리 일본인 학습자는 숙달도와 상관없이 두 유형의 문장의 읽기 속도에 차이가 나타나지 않아 구조적 단서와 의미적 단서에 모두 민감성을 보이지 않았다. 이러한 연구 결과는 일본인 한국어 학습자가 L1과 L2의 형태적 유사성에도 불구하고 피동문 실시간 문장 처리에 있어 여전히 어려움을 겪는다는 것을 의미한다.

This study investigated whether L1 Japanese learners of Korean can acquire the word order alternations of the Korean passive. In self-paced reading, Korean native speakers (n = 30) showed greater sensitivity to structural cues than semantic ones, while Japanese L2 Korean speakers (n = 60) showed no difference in reading time, regardless of proficiency, as well as showing no sensitivity to structural or semantic cues. These results indicate that Japanese L2 Korean speakers have difficulty with the on-line processing of sentences despite the morphological congruency between L1 and L2.

This study investigates the distributional patterns of the English modal verbs can and could in native (L1) and non-native (L2) spoken discourse using a multifactorial corpus-based approach. Drawing on spoken data from the ICNALE Spoken Dialogue corpus, utterances containing either can or could were annotated for their morphosyntactic and pragmatic features. A multiple logistic regression analysis revealed that modality, sentence type, and subject person significantly predict their distributional patterns, with interrogative and epistemic contexts most strongly associated with could. Importantly, a significant interaction between language group and sentence type was observed, suggesting that L1 speakers frequently employ could in interrogatives to convey politeness or mitigate face-threatening acts, while the same does not hold for L2 learners. Contrary to prior research, L2 learners in this study did not show reduced usage of epistemic modals, indicating that spoken discourse may afford more opportunities for epistemic modals than previously reported. These findings shed light on how grammatical and pragmatic factors jointly shape L2 learners’ modal usage and highlight the importance of register in understanding L2 speakers’ acquisition of modal expressions.

Prior studies have reported that L1 lemma mediation is valid in processing L2 words. On the other hand, this finding has been criticized from a methodological point of view. Therefore, this study investigates L1 lemma mediation by comparing how intermediate and advanced Korean users of English respond to the decontextualized word pairs and contextualized sentence completion items. A two-part lexical judgment questionnaire, which consists of decontextualized word association pairs and contextualized sentence completion items, was distributed to the nine research participants. And the following recall sessions were conducted with the participants to examine their perceived similarity of the word pairs. Findings from the first part of the questionnaire, the word association task, portray that the same translation pairs seem to be more semantically related in comparison to different translation pairs. This result can validate the prior findings regarding the effect of L1 lemma mediation of L2 processing. However, in the sentence completion task, advanced Korean users of English showed low reliability on the same translation pairs and they tended to choose one word for the same translation pairs. These results imply that L1 mediation may not operate in the contextualized word processing for advanced users. Further recall sessions also suggest that L1 lemma mediation may be much stronger in the decontextualized word association task and for intermediate level users of English, while the advanced level users may access L2 lexicon more conceptually. Prior studies have reported that L1 lemma mediation is valid in processing L2 words (Jing, 2002, 2004; Kwon, 2005). On the other hand, this finding has been criticized from a methodological point of view (Vespoor, 2008). According to the criticism, though L1 lemma mediation has been captured, it is mainly based on a difference in reaction time of the decontextualized lexical judgment test. Therefore, this study investigates L1 lemma mediation by comparing how intermediate (n=3) and advanced Korean users of English (n=3) respond to the decontextualized word pairs and contextualized sentence completion items. Due to the contradictory findings and the methodological development in cross-linguistic studies (Fizpatrick, 2007; Jarvis & Pavlenko, 2008), native speakers of English (n=3) were also recruited for the study in order to determine the strength of the semantic relations of the word pairs. A two-part lexical judgment questionnaire, which consists of twenty decontextualized word association pairs and ten contextualized sentence completion items, was distributed to the nine research participants. And the following recall sessions were conducted with the participants to examine their perceived similarity (Javis & Pavlenko, 2008; Kellerman, 1977) of the word pairs. The first part of the questionnaire covers ten same translation word pairs and ten different translation pairs; these questions asked the participants’ perception regarding semantic overlap of the noun pairs. Participants showed their semantic judgment through a six-point Likert scale. The second part of the questionnaire consists of ten sentence completion questions; these items asked the participants to choose either the same or the different translation pairs. Findings from the first part of the questionnaire, the word association task, portray that the same translation pairs (M=4.87, SD=.75) seem to be more semantically related in comparison to different translation pairs (M=4.52, SD=.58). This result can validate the prior findings regarding the effect of L1 lemma mediation of L2 processing (Jing, 2002, 2004; Kwon, 2005). However, in the sentence completion task, advanced Korean users of English showed low reliability (Cronbach's 𝛂 = -.275) on the same translation pairs and they, like the intermediate group participants, also tended to choose one word for the same translation pairs [ X2 (1, 0.05)=8.58, 𝑃 ＜.001, two-ways]. These results imply that L1 mediation may not operate in the contextualized word processing for advanced users. Further recall sessions also suggest that L1 lemma mediation may be much stronger in the decontextualized word association task and for intermediate level users of English, while the advanced level users may access L2 lexicon more conceptually.

This study examines whether L1-Korean L2 learners of English are sensitive to morphological agreement violations in sentence processing and whether task type affects their sensitivity. One experiment evaluates online grammaticality judgment accuracy and uses reading times to assess sensitivity to ungrammaticality. A second experiment evaluates offline grammaticality judgment accuracy. Forty-six English native speakers and 28 advanced L2 learners of English completed the online task, and 13 native speakers and 11 advanced L2 learners completed the offline task. The L2 learners showed a lack of sensitivity to the agreement violations in the online task but native-like sensitivity in the offline task. These findings suggest that non-linguistic factors such as task differences play an important role in modulating L2 learners’ sensitivity to agreement errors. The findings also provide evidence that supports the computational deficit approach as an explanation for L2 learners’ insensitivity to morphological violations.

This study examines the Split phenomenon in EFL learning, in which a word in the L1 is manifest in two or more forms in the L2. This phenomenon is most difficult to learn in the L2 when the L1 and the L2 are syntactically and typologically different, as in the case of Korean (L1) and English (L2). Communication verbs in Korean and English are a good example of the Split phenomenon, because the Korean verb (Malhata) splits into four verbs (Tell, Say, Talk, and Speak) in English. Korean university students who had read four novels during an extensive reading course were tested to determine how correctly they used the four verbs and their subtypes in context. Results show that these verbs are not equally difficult for L2 learners to learn: Talk appears to be the easiest and the [say + to NP] subtype construction the most difficult. Rates at which Korean learners used these four verbs correctly were not associated with their frequencies in either the British National Corpus or a corpus based on the novels used in the study. This suggests that L2 learners’ knowledge of verb semantics and argument structure information is not guided exclusively by input frequency or by contextual information in the process of learning an L2. Semantic transfer in the form-meaning mapping in L2 vocabulary learning is discussed.

This study investigates the processing of literal and figurative meanings of L2 idioms by Korean learners of English. Based on Cieslicka’s (2006) literal-salience resonant model, which proposes that literal meanings of idioms are accessed before figurative meanings in L2 idiom processing, the present study examines the model using a different experimental task (cross-modal lexical priming vs. self-paced sentence processing in context) and learners with different L1 background (Polish L1 vs. Korean L1). Ten English idioms were embedded in sentences (e.g., He kicked the bucket) and presented in conversation-based contexts biased toward either literal or figurative meaning (He kicked the bucket = He died). A novel non-idiom sentence (He filled the bucket) was also presented for each idiom in an appropriate context. Sixty seven Korean learners of English read the contexts line by line and their reading times were measured for each line. Literal sentences were read faster than figurative sentences. This suggests that, in L2 idiom processing, the literal meaning of an idiom was processed before its figurative meaning and that figurative meaning was more difficult to compute than both literal and novel non-idiom meanings even in the presence of strong contextual cues. In general, the results of the present study are in accordance with the literal-salience resonant model when different methods and learners with different L1 background were employed.

The present study tested whether adult Korean learners of English are able to immediately integrate verb-specific structural information (i.e. subcategorization restrictions imposed by individual verbs) as they read sentences in English. Seventeen native speakers of English and 22 Korean learners of English participated in an eye-tracking experiment, in which they read English sentences that contained a wh-dependency between the head NP and the verb of a relative clause. The relative clause verbs consisted of three types that varied in terms of subcategorization properties: transitive verbs that take an NP complement more often than a PP complement, transitive verbs that take a PP complement more often than an NP complement, and unaccusative verbs that are strictly intransitive. The results suggest that whereas native speakers use verb subcategorization information in making initial parsing decisions regarding wh-dependency formation, Korean learners are only able to integrate the information in later stages of sentence processing. Possible accounts of this discrepancy in the time course of verb information integration are discussed.

This study explores how native English speakers, non-native English speakers, and bilingual speakers process formulaic phrases compared to non-formulaic phrases and investigates whether processing different types of formulaic sequences gives a comparative advantage or incurs processing costs. Three groups of speakers performed grammaticality judgment tasks: L1 native English speakers (ES), English bilingual speakers (EB), and Korean speakers learning English as an L2 (KS). For the type of stimuli under consideration, English idioms, phrasal verbs and collocations were used for psycholinguistic experiments. First, for all three groups, our results show that formulaic phrases are processed more rapidly and accurately than similar nonformulaic phrases. We suggest that for all three speaker groups, formulaic phrases are represented as whole units as if they are single lexical items and processed as holistically as other morphologically complex words. Second, each formula type is processed with different response times and accuracy for all three groups. This suggests that different formulas are stored, retrieved and organized in a distinct manner in each group’s mental lexicons, reflected in response latency, conceivably due to different formulaicity or semantic opacity. Furthermore, we show that conventionalized phrases, such as idioms and collocations, are processed more quickly and accurately by L1 English or English bilingual speakers, compared to Korean L2 English speakers. This implies that these formulas are problematic for L2 learners’ acquisition or vocabulary expansion. Finally, this study shows that the n-gram frequency or length of the formulas mediate the processing by all three groups, suggesting the possibility of frequency being encoded as part of lexicalized formulas.

As a replication study of Hanaoka and Izumi (2012), the present research investigated the process of learners’ noticing of problems in their interlanguage (IL) while they were producing the second language (L2) and their process of relevant input provided in the form of two types of feedback (i.e., reformulation and a model text) in L2 writing. The data were collected from twenty-five university students in Korea, and they engaged in a four-staged picture description task consisting of the writing stage (Stage 1), the comparison stage (Stage 2), the immediate revision stage (Stage 3), and the delayed revision stage (Stage 4). It was found that output triggered learners’ noticing of problems; however, the problems did not always show up overtly in their writing. Some problems were manifested in writing (i.e., overt problems), while some were hidden (i.e., covert problems). A model text and reformulation played somewhat different roles in offering the solutions to the problems: a model text provided more solutions to covert problems while reformulation provided more solutions to overt problems. And, the extent of learners’ noticing was associated with the types of problems (overt vs. covert problems) and feedback (i.e., models vs. reformulation). However, the types of problems and feedback were not related to the rate of incorporation of feedback into subsequent writing.

본 연구는 영어 원어민과 비원어민 대학생의 주장문 쓰기에서 나타나는 어휘 묶음(lexical bundle) there are so many의 의미적 운율(semantic prosody)을 분석한다. 기존 연구들이 주로 구조와 기능적 특성에 집중해온 반면, 본 연구는 담화 맥락에서 드러나는 긍·부정·중립적 의미 함의에 초점을 맞춘다. 자연어처리 기반 감성 분석을 활용하여, 두 학생 집단이 해당 어휘 묶음을 통해 어떤 입장과 태도를 드러내는지를 두 개의 비교 가능한 말뭉치를 바탕으로 분석하였다. 두 말뭉치는 각각 미국, 한국 대학생들이 작성한 영어 주장문으로 구성되어 있으며, 두 집단 모두 동일한 주제에 대해 제한된 시간 내에 글을 작성하였다. 연구 결과, 원어민 학생들은 해당 어휘 묶음을 주로 중립적이거나 긍정적인 맥락에서 사용하였으며, 이는 주로 우회적 표현이나 예시 제시에 활용되었다. 반면, 비원어민 학습자들은 이 어휘 묶음을 부정적인 맥락에서 더 자주 사용하였고, 그 결과 의미적으로 어색하거나 담화 상황에 부적절하게 사용되는 경우도 나타났다. 이러한 결과는 두 집단이 동일한 어휘 묶음을 문맥에 따라 다르게 활용한다는 점을 보여주며, 이는 담화 맥락에 적절한 표현 선택을 강조하는 글쓰기 교육이 필요함을 시사한다.

This study explores the semantic prosody of the lexical bundle (LB) there are so many in native and non-native novice English argumentative writing. While previous research on LBs has largely focused on their structural and functional patterns, this study shifts attention to their evaluative meanings in context. Using NLP-based sentiment analysis, it examines how this bundle conveys evaluative stance across two comparable corpora: one consisting of argumentative essays written by native English-speaking students and the other by Korean EFL learners. Both groups were incoming college freshmen who responded to the same prompts under identical time constraints. The results exhibited a clear difference in semantic prosody: native writers tended to use the bundle in neutral or positive contexts, often to hedge or provide examples, whereas non-native writers used it more frequently in negative contexts, which could lead to semantic incongruity or pragmatic misalignment. These findings move beyond existing accounts of LBs by highlighting the role of evaluative meaning. They suggest that L2 learners may benefit from explicit instruction on how stance is expressed in frequently occurring word sequences—an area particularly challenging for novice academic writers. This study highlights the value of sentiment-aware writing pedagogy, as informed by corpus-based insights into learner language.

This study investigates the effects of immediate repetition of auditory words on L2 pronunciation improvement of English schwa by Japanese EFL learners. More specifically, the study is concerned with how the amount of input (i.e., five or ten repetitions) and stimuli characteristics (i.e., the position of schwa in a word, word familiarity) influences the repetition effect. The study consists of a study and a test phase: In the study phase, participants listened to and repeated words. In the test phase, they repeated spoken stimuli including previously heard as well as new words, both ten minutes and one week after the study phase. The auditory materials consisted of 32 real words. The duration ratio of schwa to a stressed vowel in a word and the quality of the schwa were acoustically examined. A repetition effect was mainly found in the durational aspect. Repeating the stimuli more than five times did not differentiate the effect. One week after the study phase, the effect persisted only for initial schwa, which is weakly represented in the participants’ mental representation. The pedagogical implications for L2 pronunciation teaching from the view of language processing are discussed.

This paper investigates the use of reading strategies and analyzes the reading behaviors of mature English-as-a-foreign-language (EFL) readers. In brief, this study examines the reading performance of four Thai college students. Through the use of think-aloud protocols, interviews, and observations, I document EFL Thai students’ use of reading strategies while they are in the process of reading and comprehending four English and Thai texts. The findings reveal that three broad themes emerge from the analysis: a) the most frequentlyused strategies across readers in reading L1 and L2 texts, b) the effective use of reading strategies demonstrated by the students in reading L2, and; c) metacognition development in reading L2. However, the data did not provide adequate evidence to comment on the issues of the language transfer and cognates in reading L2 texts.

본 연구는 중국인 한국어 학습자의 한국어 양태성 이동동사 구문에 나타나는 ‘에서’와 ‘을/를’의 조사 교체 습득을 살펴보는 데 목적을 둔다. 이와 같은 현상은 형태와 기능 간의 일대일 대응에 어긋나는 조사 교체 현상으로 중국인 한국어 학습자들에게는 습득의 어려움을 예상할수 있다. 따라서 중국인 한국어 중 ․ 고급 학습자들을 대상으로 조사 교체 현상의 습득 양상을 살펴보고자 한다. 이를 위해 수용성 판단 과제 (acceptability judgment task)를 통해 문법성 인식에 대해 살펴본 후 조사 교체에 대한 지식을 실시간(online)으로 처리할 수 있는지 여부를 살펴보기 위해 자기조절읽기 과제(self-paced reading task)를 실시해습득 양상을 고찰하였다. 그 결과, 중국인 학습자 상, 하 집단 모두 양태성 이동동사 구문에서 ‘을/를’보다 ‘에서’를 수용하였으며 상 집단의 경우 역시 한국인 모어화자와 달리 조사 교체에 대한 지식이 부족 했다. 실시간 처리에서도 중국인 학습자들은 한국인 모어화자들과 달리 조사 교체에 대한 민감성을 보이지 않았다. 이와 같은 결과는 조사에 대한 교수가 개별 조사에 그칠 것이 아니라 조사 교체가 일어나는 구문들에 대한 명시적 교수로 확대되어야 함을 시사한다.

This study investigated whether L1 Chinese learners of Korean can acquire case particle alternation in motion verb constructions. An acceptability judgment and self-paced reading task were conducted with L2 advanced learners of Korean in tandem with native Korean speakers. The result of the acceptability judgment task demonstrated that the L1 group accepted the sentences in both case-marking conditions, whereas the L2 groups (both higher and lower proficiency learners) only accepted the sentences in the ‘ul/lul’ condition, while rejecting the sentences in the ‘eseo’ condition. From the results of the self-paced reading task, L2 participants did not show any reading time differences between the two case-marking conditions. However, given that the L2 learners failed to demonstrate target-like knowledge of ‘in manner of motion’ verb construction in the acceptability judgment task, their reading-time patterns in the self-paced reading task may not be interpreted as their acceptance of both conditions. These findings suggest that the L1 Chinese-speaking learners of Korean have a persistent difficulty acquiring the case alternation regarding motion verb construction. (Ewha Womans University)