L2 Production Research Articles

Musical Aptitude and the Acquisition of Pitch and Duration in L2 Vowels

Music and language have been believed to be processed similarly in the brain. Studies have compared the processing of vowels and music, as they share spectral and temporal features such as pitch and timing. When comparing music and language skills the question arises whether music skills are inherent, measured with aptitude, or acquired, measured with experience. This article focused on musical aptitude studies. It was found that musical aptitude predicted non-native speech-sound processing better than musical experience. It is proposed that music and language skills are interconnected as greater overlaps in neural processing of music and speech were found for musicians than nonmusicians, and as individuals with no musical experience but with high L2 production skills were found to have high musical aptitude. Furthermore, spectral sensitivity was found to be the primary partial mediator of the link between musicality and non-native speech-sound processing, with temporal sensitivity having no role.

Update on the GOSAT TANSO–FTS SWIR Level 2 retrieval algorithm

Abstract. The National Institute for Environmental Studies has provided the column-averaged dry-air mole fraction of carbon dioxide and methane (XCO2 and XCH4) products (L2 products) obtained from the Greenhouse gases Observing SATellite (GOSAT) for more than a decade. Recently, we updated the retrieval algorithm used to produce the new L2 product, V03.00. The main changes from the previous version (V02) of the retrieval algorithm are the treatment of cirrus clouds, the degradation model of the Thermal And Near-infrared Spectrometer for carbon Observation–Fourier Transform Spectrometer (TANSO–FTS), solar irradiance spectra, and gas absorption coefficient tables. The retrieval results from the updated algorithm showed improvements in fitting accuracies in the O2 A, weak CO2, and CH4 bands of TANSO–FTS, although the residuals increase in the strong CO2 band over the ocean. The direct comparison of the new product obtained from the updated (V03) algorithm with the previous version V02.90/91 and the validations using the Total Carbon Column Observing Network revealed that the V03 algorithm increases the amount of data without diminishing the data qualities of XCO2 and XCH4 over land. However, the negative bias of XCO2 is larger than that of the previous version over the ocean, and bias correction is still necessary. Additionally, the V03 algorithm resolves the underestimation of the XCO2 growth rate compared with the in situ measurements over the ocean recently found using V02.90/91 and V02.95/96.

Temperature and Relative Humidity Profile Retrieval from Fengyun-3D/VASS in the Arctic Region Using Neural Networks

In this study, a new technique is proposed to retrieve temperature and relative humidity profiles under clear sky conditions in the Arctic region based on the artificial neural network (ANN) algorithm using Fengyun-3D (FY-3D) vertical atmospheric sounder suit (VASS: HIRAS, MWTS-II, and MWHS-II) observations. This technology combines infrared (IR) and microwave (MW) observations to improve retrieval accuracy in the middle and low troposphere by reducing the sensitivity of the neural networks (NNs) to cloud coverage. The approach was compared against other methods available in the literature on retrieving profiles only from FY-3D/HIRAS data. Furthermore, its retrieval performance was tested by comparing the NNs’ prediction accuracy versus the corresponding FY-3D/VASS and Aqua/AIRS L2 products. The results showed that: (1) NNs retrieval accuracy is higher during the warm season and over the ocean; (2) the retrieval accuracy of NNs has been significantly improved compared with satellite L2 products; (3) referring to radiosonde observations, the retrieval accuracy of NNs below 600 hPa is effectively improved by adding the information of the MW channel, especially on land where cloud clearing is more difficult. The root mean square error (RMSE) of temperature and relative humidity in the cold season were reduced by 0.3 K and 2%, respectively. The advanced NNs proposed herein offer a more stable retrieval performance compared with NNs built only by FY-3D/HIRAS data. The study results indicated the potential value in time and space domain of the NN algorithm in retrieving temperature and relative humidity profiles of the Arctic region from FY-3D/VASS observations under clear-sky conditions. All in all, this work enhances our knowledge towards improving operational use of FY-3D satellite data in the Arctic region.

The Distribution of Interactional Space and Collaboration in EFL Task-Based Peer Interactions

This paper focuses on the influence of task types, namely divergent and convergent tasks, on the interactional space of learners and the amount of collaboration in EFL adult learners’ interactions naturally emerging from performing each type of task. To define interactional space, behavioural engagement measurement was adopted, and the total number of turns and words were quantified to define the distribution of learner talk in each task. Additionally, the categories of collaborative behaviours were quantified for both task types to illustrate their distribution in the tasks. The data of this study involve eleven hours of learner interactions collected from an EFL context where the learners voluntarily participated in a speaking club and completed eight tasks. The interactions were audio-recorded and used for the analysis of the influence of task types on learners’ L2 production and the distribution of collaboration in each task. The results yield a difference in the amount of L2 production between tasks; for example, convergent tasks facilitate more L2 turns and more L2 words than divergent tasks although the mean length of utterance is higher in divergent tasks. Similarly, the distribution of collaborative behaviours is mostly higher in convergent tasks. The results are discussed in relation to both interactionist and sociocultural theories of L2 learning and some implications are also provided based on the results.

L1 category precision hypothesis in L2 production: Korean learners’ English front vowels

It has been hypothesized that learners with more precisely defined categories in their native language (L1) are better at learning nonnative sounds in a second language (L2). Some perception studies have supported this hypothesis, but not many production studies have been conducted. Our study aims to fill this gap. Focusing on front vowels in F1/F2 acoustic space, we calculated compactness scores of L1 vowels and distances between learners’ and native speakers’ productions to operationalize category precision and L2 accuracy, respectively. Then, we examined whether learners with a lower compactness score would produce L2 vowels with a closer distance. The learners’ productions were taken from a corpus containing native Korean speakers’ reading and retelling of the North Wind and the Sun in Korean and English and comparable native English speakers’ productions from ALLSSTAR corpus. Results showed that learners with lower compactness scores produced more accurate English /i/ and /ɪ/, but not for /ɛ/ or /æ/, which partially supports the hypothesis's application to L2 production. We speculate that the merger between /e/ and /æ/ in Korean, the Northern Cities Vowel Shift in English, and/or perceptual mapping between native and L2 categories are responsible for the English non-high vowel results.

Asymmetrical effects of cross-linguistic structural priming on cross-linguistic influence in L2 learners

AbstractThe present study investigates current proposals that priming is a mechanism of cross-linguistic influence (CLI) in bilinguals by aiming to boost CLI through priming. In two cross-linguistic structural priming experiments with less-proficient adolescent (Study 1) and more highly proficient adult German-English learners (Study 2), we assess whether structural priming enhances CLI for well-formed, dispreferred, and ungrammatical structures. L2 learners in both studies showed CLI in their production, in particular for structures with word order overlap. They also exhibited short-term cross-linguistic priming of grammatical L1-L2 word orders in L2 English, which extended to longer-term priming among the more highly proficient learners in Study 2. However, there was no evidence that cross-linguistic priming increased the use of dispreferred or ungrammatical L1-based word orders in L2 English in either study. Rather, the overall production of these word orders decreased. Together, these results suggest that, while cross-linguistic priming leads learners to increase the use of shared, grammatical L1-L2 word orders, it leads to the inhibition of non-shared ungrammatical structures in L2 production. We conclude that priming has asymmetrical effects on CLI of grammatical and ungrammatical L1-based structures in the L2.

The acoustics of the alignment of focus prosody and lexical stress: A production study of native speakers and second language learners of English

In English, the tonal target of a pitch accent is aligned with lexical stress. Therefore, the phonetic realization of contrastive focus prosody, phonologically marked by a H*/L + H* pitch accent, is expected to vary depending on the stress pattern of the focused word. This study explored acoustic effects of the alignment of lexical stress and focus prosody and examined how it is learned by Korean learners of English, whose first language lacks lexical stress. Ten native English speakers (L1) and 19 Korean learners of English (L2) participated in a read-aloud task, in which they produced nine trisyllabic words with primary stress on one of the three syllables with contrastive focus (e.g., Animal, vaNILla, engiNEER). For each syllable, three acoustic measures were extracted: the height of pitch peak, periodic energy mass, and the direction of pitch movements (Albert et al., 2021; Cangemi et al., 2019). Results revealed that the height of pitch peak, a conventional measure of both lexical stress and contrastive focus, did not properly demonstrate their alignment. Instead, the stressed syllable of the focused words was marked by its overall prominence (mass) or the direction of pitch movements, depending on the stress pattern. The L2 speakers’ productions were similar to the L1 productions, but there were indications of an influence of their L1 prosody and English proficiency. [This work was supported by the Ministry of Education and the National Research Foundation of Korea (NRF-2022S1A5A8052482).]

Factors Affecting L2 Production of English Affricates and Fricatives in Word-Final Position

Factors Affecting L2 Production of English Affricates and Fricatives in Word-Final Position

Formal Genre-Specific Knowledge as a Resource-Dispersing Feature of Task Complexity

Recent second language (L2) writing research informed by task-based theories of second language acquisition has enthusiastically adopted task complexity frameworks to describe the specific cognitive demands of a given writing task and the effect of those cognitive demands on written L2 production. However, missing from many studies on the effects of task complexity on L2 written production is a discussion of genre as a potential source of task complexity. This paper examines the potential of genre as a resource-dispersing form of task complexity that is unique to writing. The article summarizes the predictions of task-based theories of second language acquisition particularly the predictions of the Cognition Hypothesis and its intersection with Kellogg’s widely-cited model of working memory in writing. It then argues that formal genre-specific knowledge constitutes a resource-dispersing form of task complexity that is distinct from general L2 proficiency and general writing proficiency.

Behavioral and ERP evidence of differences in pitch feedback control in late bilinguals’ L1 and L2 speech production

AbstractThis study compared late bilinguals’ pitch feedback control in L1 and L2 production using a frequency-altered feedback paradigm in which participants read target words while presented with unexpected pitch-shift in their voice feedback. Variables of language (L1 or L2) and perturbation magnitudes (0, 100, 200, or 400 cents) were manipulated. Behaviorally, participants produced larger magnitudes but longer latencies of vocal compensation in L2 than in L1 production, suggesting that L2 pitch feedback control has greater importance but lower efficiency. Event-related potential findings demonstrated that 400-cent shifts elicited greater N1 amplitudes than those in the 0-cent baseline condition in L1 production. This difference was non-significant in L2 production, implying different neural processing of unaltered feedback and externally-generated feedback in L1 and L2 production. Participants’ vocal compensation and P2 amplitudes were similarly modulated by pitch-shift in L1 and L2 production, implying a similar gating mechanism to correct internal and external errors.

Matchup Strategies for Satellite Sea Surface Salinity Validation

Satellite validation is the process of comparing satellite measurements with in-situ measurements to ensure their accuracy. Satellite and in-situ sea surface salinity (SSS) measurements are different due to instrumental errors (IE), retrieval errors (RE), and representation differences (RD). In real-world data, IE, RE, and RD are inseparable, but validations seek to quantify only instrumental and retrieval error. Our goal is to determine which of four methods comparing in-situ and satellite measurements minimizes RD most effectively, which includes differences due to mismatches in the location and timing of the measurement, as well as representation error caused by the averaging of satellite measurements over a footprint. IE and RE were obviated by using simulated Argo float, and L2 NASA/SAC-D Aquarius, NASA·SMAP, and ESA·SMOS data generated from the high-resolution ECCO (Estimating the Climate and Circulation of the Oceans) model SSS data. The methods tested include the all-salinity difference averaging method (ASD), the N closest method (NCLO), which is an averaging method that is optimized for different satellites and regions of the ocean, and two single salinity difference methods—closest in space (SSDS) and closest in time (SSDT). The root mean square differences (RMSD) between the simulated in-situ and satellite measurements in seven regions of the ocean are used as a measure of the effectiveness of each method. The optimization of NCLO is examined to determine how the optimum matchup strategy changes depending on satellite track and region. We find that the NCLO method marginally produces the lowest RMSD in all regions but invoking a regionally optimized method is far more computationally expensive than the other methods. We find that averaging methods smooth IE, thus perhaps misleadingly lowering the detected instrumental error in the L2 product by as much as 0.15 PSU. It is apparent from our results that the dynamics of a particular region have more of an effect on matchup success than the method used. We recommend the SSDT validation strategy because it is more computationally efficient than NCLO, considers the proximity of in-situ and satellite measurements in both time and space, does not smooth instrumental errors with averaging, and generally produces RMSD values only slightly higher than the optimized NCLO method.

Is Full-Time Equivalent an Appropriate Measure to Assess L1 and L2 Perception of L2 Speakers with Limited L2 Experience?

The revised version of the Speech Learning Model (SLM-r) regards full-time equivalent (FTE), which involves the amount of L2 use during the length of residence (LOR) in an L2 setting, as the main factor in L2 speech acquisition. Previous studiesshowed that LOR has a significant effect on L2 and L1 production and perception but does not explain differences between populations (i.e., L1-Spanish L2-English vs. L1-English L2-Spanish). A reanalysis of the data has been conducted by calculating the FTE of the experienced participants. The aim was also to investigate whether the assumptions of the SLM-r are applicable to L1 and L2 perception. A series of correlation tests between FTE and category boundary—between voiced and voiceless stops—was conducted, yielding non-significant results. The relatively short LOR of participants, the quality of the input and differences in terms of L2 instruction between participants could explain the lack of a clear effect of FTE in this study. Therefore, FTE on its own may not be sufficient to account for L2 accuracy in perception, at least for L2 speakers with limited L2 input, and other factors should be considered.

Production and perception of English consonants by Yemeni EFL learners

Many previous studies on the perception-production correlation focused mainly on transcriptions and native English speaker evaluations for production accuracy assessments; only a few included acoustic measurements. The purpose of this study is to examine the production and perception of six English consonants by Yemeni EFL learners of English using an acoustic measurement of L2 production in order to better understand the link between production and perception. It has three objectives: (1) To investigate the extent to which word position influences the accuracy of Yemeni EFL learners’ production of the target consonants, (2) to examine the extent to which word position affects their perception, and (3) to examine the relationship between the overall production and perception of the investigated sounds. A quantitative research method was employed to collect data from six Yemeni EFL postgraduate students from Universiti Utara Malaysia (UUM). In the production tasks, the speakers were asked to produce seventy-two words with the target consonants three times in isolation (Experiment 1) and the other three times in a carrier sentence (Experiment 2). The researchers recorded the speakers’ productions with an iPhone and later evaluated them acoustically via Praat. In the perception test, an AXB experiment was conducted. The stimuli were read by a native male speaker of Australian English. The findings showed that word position significantly affected the production, yet not the perception of the target sounds. Moreover, an insignificant positive moderate correlation was revealed between the overall production and perception of the target consonants. Results have implications for L2 speech and pronunciation instruction and learning. Teachers may focus more on specific sound environments that lead learners to struggle while producing/perceiving particular English sounds.

Sea Surface Height Estimation from Improved Modified, and Decontaminated Sub-Waveform Retracking Methods over Coastal Areas

Coastal zones are challenging areas for sensing by satellite altimeters because reflected signals from non-water surfaces and from calm sea surfaces in small bays and ports inside the radar footprint lead to erroneous powers in return waveforms. Accordingly, these contaminated waveforms do not follow the so-called Brown model in conventional retracking algorithms and fail to derive qualified ranges. Consequently, the estimated water level is erroneous as well. Therefore, selecting an optimized retracker for post-processing waveforms is significantly important to achieve a qualified water level estimation. To find the optimized retracker, we employed a methodology to minimize the effect of erroneous powers on retracked range corrections. To this end, two new approaches were presented, one based on a waveform decontamination method and the other based on a waveform modification method. We considered the first meaningful sub-waveforms in the decontaminated waveforms and in the modified waveforms to be processed with a threshold retracker. To assess their performance, we also retracked the decontaminated and modified full-waveforms. The first meaningful sub-waveform and full-waveform in the original waveforms were retracked to compare the performance of the modified and decontaminated waveform retracking with the original waveform retracking. To compare the results of our sub-waveform retracking algorithms with those of external sub-waveform retracking algorithms, the (Adaptive Leading Edge Sub-waveform) ALES database was also used. In our retracking scenarios, we used the Sentinel-3A SRAL Altimeter to estimate the water levels over the study area within 10 km from the coastlines in both the Persian Gulf and the Bay of Biscay from June 2016 to October 2020. The water levels from processing L2 products were estimated as well. We evaluated our retracking scenarios and L2, as well as the ALES processing results, against the tide gauges. Our analysis showed that within 0–10 km from the coast, the first meaningful sub-waveform of the decontaminated waveforms had the best performance. We reached maximum RMS improvements in this scenario of 53% and 86% over the Persian Gulf and the Bay of Biscay, respectively, in comparison with L2 processing. Over these distances from the coast, the first sub-waveform from the original waveforms and the modified waveforms stayed in the second and third order of performance. The ALES database with an RMS ranging from 13 to 51 cm had a worse performance than all of our sub-waveform retracking scenarios.

The Predictive Role of Quantity and Quality Language-Exposure Measures for L1 and L2 Vocabulary Production among Immigrant Preschoolers in Italy.

In this study, we investigated the lexical ability in L1 and L2 of 60 immigrant children who were 37 to 62 months old and exposed to minority languages (L1) and Italian (L2). Using the MacArthur-Bates Communicative Development Inventories, we measured children's vocabulary production in L1 and L2. From interviews, we collected data on quantitative language exposure (parental input, child output, length of exposure to L2 at preschool, and parental oral fluency) and qualitative home-language exposure (HLE) practices (active, play, and passive) in L1 and L2. We conducted stepwise regression analyses to explore which factors predicted children's vocabulary production in L1 and L2. The child's chronological age and parental education were not predictors of vocabulary production. L2 parental input, L1 child output, and L1 HLE-active practices explained 42% of the variance in children's L1 vocabulary production. L2 child output and L2 HLE-active practices explained 47% of the variance in children's L2 vocabulary production, whereas length of L2 exposure in preschool was a predictor only when we included quantitative language-exposure factors in the model. The effects of the quantity and quality of language exposure on lexical ability among preschool immigrant children are discussed.

How adequately are elevated moist layers represented in reanalysis and satellite observations?

Abstract. We assess the representation of elevated moist layers (EMLs) in ERA5 reanalysis, the Infrared Atmospheric Sounding Interferometer (IASI) L2 retrieval Climate Data Record (CDR) and the Atmospheric Infrared Sounder (AIRS)-based Community Long-term Infrared Microwave Combined Atmospheric Product System (CLIMCAPS)-Aqua L2 retrieval. EMLs are free-tropospheric moisture anomalies that typically occur in the vicinity of deep convection in the tropics. EMLs significantly affect the spatial structure of radiative heating, which is considered a key driver for meso-scale dynamics, in particular convective aggregation. To our knowledge, the representation of EMLs in the mentioned data products has not been explicitly studied – a gap we start to address in this work. We assess the different datasets' capability of capturing EMLs by collocating them with 2146 radiosondes launched from Manus Island within the western Pacific warm pool, a region where EMLs occur particularly often. We identify and characterise moisture anomalies in the collocated datasets in terms of moisture anomaly strength, vertical thickness and altitude. By comparing the distributions of these characteristics, we deduce what specific EML characteristics the datasets are capturing well and what they are missing. Distributions of ERA5 moisture anomaly characteristics match those of the radiosonde dataset quite well, and remaining biases can be removed by applying a 1 km moving average to the radiosonde profiles. We conclude that ERA5 is a suitable reference dataset for investigating EMLs. We find that the IASI L2 CDR is subject to stronger smoothing than ERA5, with moisture anomalies being on average 13 % weaker and 28 % thicker than collocated ERA5 anomalies. The CLIMCAPS L2 product shows significant biases in its mean vertical humidity structure compared to the other investigated datasets. These biases manifest as an underestimation of mean moist layer height of about 1.3 km compared to the three other datasets that yields a general mid-tropospheric moist bias and an upper-tropospheric dry bias. Aside from these biases, the CLIMCAPS L2 product shows a similar, if not better, capability of capturing EMLs compared to the IASI L2 CDR. More nuanced evaluations of CLIMCAPS' capabilities may be possible once the underlying cause for the identified biases has been found and fixed. Biases found in the all-sky scenes do not change significantly when limiting the analysis to clear-sky scenes. We calculate radiatively driven vertical velocities ωrad derived from longwave heating rates to estimate the dynamical effect of the moist layers. Moist-layer-associated ωrad values derived from Global Climate Observing System Reference Upper-Air Network (GRUAN) soundings range between 2 and 3 hPa h−1, while mean meso-scale pressure velocities from the EUREC4A (Elucidating the Role of Clouds-Circulation Coupling in Climate) field campaign range between 1 and 2 hPa h−1, highlighting the dynamical significance of EMLs. Subtle differences in the representation of moisture and temperature structures in ERA5 and the satellite datasets create large relative errors in ωrad on the order of 40 % to 80 % with reference to GRUAN, indicating limited usefulness of these datasets to assess the dynamical impact of EMLs.

An Analysis of factors in the First Language (Thai) that Influence the Learning of the Second Language (English)

The purpose of this non-empirical study was to analyze the factors in L1 (Thai) that influence L2 learning (English) in Thai learners. It consisted of four sections. The first one is an introduction. The second one is an introduction to the Thai language and the English language. The third section illustrates the differences in syntax between Thai and English. Numerous supporting studies were reviewed, including a discussion of whether the Thai and English languages have an effect on L2 learning. The conclusion was drawn in light of numerous related studies and other recent empirical studies that support the conclusion in the final section. According to many studies reviewed in this paper, it was determined that intralingual errors, not interlingual errors or L1 interference, are the primary source of errors in learners’ L2 production.

제2외국어로서 영어와 한국어 폐쇄음 산출과 인지 연구

This study examines how L1 Korean learners of L2 English and L1 English learners of L2 Korean produce L2 stops in terms of voice onset time (VOT) and fundamental frequency (F0) and how they perceive L2 stops. The study also examines how the level of L2 proficiency functions as a critical factor in determining the L1-to-L2 influence. Two findings were derived from production experiments. First, the primary acoustic cues of L1 stops are negatively transferred to the production of L2 stops in L2 beginners. Second, advanced L2 learners acquire the primary cues of L2 stops. These results indicate that the acquisition of L2 stops is regulated by the different priority order of VOT and F0 between L1 and L2 stops and learners

제 3 언어로서의 스페인어 발음 학습에 관한 제언: 미국 내 한국어 화자를 중심으로

This study addresses the issue of how native Korean speakers living in the US have learned Spanish as a third language, particularly in terms of their pronunciation. Specifically, it delves into what characterizes their pronunciation, from what perspectives the pronunciation can be understood, and what researchers should take into account to examine their pronunciation. To do so, it first introduces models on L2 perception and production as well as models on L3 morphosyntax. Then, this study attempts to apply the main principles of the models to the learning contexts and characteristics of Korean Spanish learners in the US. Lastly, based on the theoretical underpinnings and their learning characteristics, it provides suggestions for empirical studies that intend to examine L3 Spanish pronunciation. This study can be useful for Korean learners of L3 Spanish and their parents, Spanish teachers, and L3 researchers. In addition, some theoretical and empirical suggestions of this study can be applicable to the Korean learners of L3 Spanish who live in Korea and English-speaking countries other than the US.

Linguistic and Cultural Differences Influencing Thai to English Translation: On the Qualitative Data Suggesting Techniques Employed by Third Year Students, Ubon Ratchathani Rajabhat University, Thailand

Linguistic and cultural differences between Thai and English underlie expected errors through both formal and informal observations of language productions in the Thai context. In a Thai-to-English translation task by 138 participants, contrastive analysis is used to analyze the structural similarities and differences between the source language and the target language together with error analysis and interlanguage theories (Spolsky, 1979; Lennon, 2008; Gass & Selinker, 2008; Kantè, 2014; Kantè, 2015) attempting to investigate the interlanguage system of the participants while dealing with the task in hand. With the use of formal cooperative learning groups under the Cooperative Language Learning method (Richards & Rodgers, 2001: 192-201) outside of the classroom for data collection, this 7-week long group work project of short story translation was assigned to third year students to see types of errors produced. The qualitative data indicate sources of L2 productions including direct translation, a lack of knowledge on grammatical features of lexical choices, overgeneralization of and innovations in culture-specific items, and rare exposure to target choices. The findings further suggest that the participants were prone to rely heavily on online translation machines or software and either online or printed dictionaries where all possible choices are provided without looking more carefully at which choices are appropriate for a certain context. In addition, for a task outside of the classroom in the digital age with easy online accessibility and distant teachers’ supervision, the ethical issue on plagiarism is to be intensively emphasized.