Word Translation Task Research Articles

Positive and Negative Lexical Transfer in English Vocabulary Acquisition

The aim of the paper is to explore the appearance of positive and negative lexical transfer of plurilingual learners in English vocabulary acquisition. Cross-linguistic influences in the study are examined by word translation tasks from Croatian into English, including true, partial, and deceptive cognates or false friends in English, German, and Italian. The results have revealed different language dominances and positive or negative transfer manifestation. Lexical transfer from L4 German is manifested positively, but the Italian language seems to play a dominant role in the acquisition of English vocabulary. The effect of Croatian is manifested both positively and negatively. The study has confirmed previous psycholinguistic studies on the complexity of lexical connections in plurilingual learners and the dynamic interaction of various learning-based factors, such as language recency, proficiency, exposure to languages, the order in which languages are learned, and the formal context in language learning.

Mutual intelligibility of a Kurmanji and a Zazaki dialect spoken in the province of Elazığ, Turkey

Abstract We present the first results of a large project concerned with the mutual intelligibility between Zazaki and Kurmanji dialects spoken in Eastern Anatolia. There is an ongoing debate on the classification of Kurmanji and Zazaki as separate languages or as dialects of the same language, Kurdish. However, there is no scientific study of how well speakers of Zazaki and other dialects of Kurdish can understand each other. In this paper, we present the results of a pilot investigation where we tested the mutual intelligibility of 69 Kurmanji and Zazaki participants by means of a word translation task and asked the participants to estimate how well they could understand the other language variety. The results showed that overall the mutual intelligibility was rather low. There was a significant interaction between the effects of gender and language. Zazaki males identified more words correctly than Kurmanji males while Kurmanji females had higher intelligibility scores than Zazaki females. We suggest linguistic (lexical) and non-linguistic (attitudes and amount of exposure) explanations for the intelligibility results. We also have a closer look at the intelligibility of individual words to gain a greater understanding of the reasons for the asymmetric intelligibility results.

Pre-Training on Mixed Data for Low-Resource Neural Machine Translation

The pre-training fine-tuning mode has been shown to be effective for low resource neural machine translation. In this mode, pre-training models trained on monolingual data are used to initiate translation models to transfer knowledge from monolingual data into translation models. In recent years, pre-training models usually take sentences with randomly masked words as input, and are trained by predicting these masked words based on unmasked words. In this paper, we propose a new pre-training method that still predicts masked words, but randomly replaces some of the unmasked words in the input with their translation words in another language. The translation words are from bilingual data, so that the data for pre-training contains both monolingual data and bilingual data. We conduct experiments on Uyghur-Chinese corpus to evaluate our method. The experimental results show that our method can make the pre-training model have a better generalization ability and help the translation model to achieve better performance. Through a word translation task, we also demonstrate that our method enables the embedding of the translation model to acquire more alignment knowledge.

The multiple phonological activation in Chinese spoken word production: An ERP study in a word translation task

The multiple phonological activation in Chinese spoken word production: An ERP study in a word translation task

Effects of Continuous Theta Burst Stimulation Over the Left Dlpfc on Mother Tongue and Second Language Production In Late Bilinguals: A Behavioral and ERP Study.

Clinical, neuroimaging, and non-invasive brain stimulation studies have associated the dorsolateral prefrontal cortex (DLPFC) with the multilingual language control system. Here, we investigated if this role is increased during the processing of the non-dominant language due to the higher cognitive/attentional demands. We used an inhibitory repetitive transcranial magnetic stimulation (rTMS) protocol over the left DLPFC and investigated the behavioral and electrophysiological effects on (i) picture naming in the mother tongue and second language, (ii) forward and backward translation and (iii) non-verbal inhibition. To this end, we compared the effects of inhibitory rTMS (cTBS) vs sham-rTMS using a single-blind within-subject design including 22 late bilinguals. Behaviorally, response times were longer after cTBS compared to sham-rTMS in the picture naming task independent of language, while response times were not affected for the word translation task. These results were mirrored on the electrophysiological level showing an effect of stimulation in the picture naming task starting at 547ms post-stimulus onset, but not in the translation task. This late time range is likely associated with processes of conflict resolution and initiation of the articulation of the word rather than processes related to lexical selection or language switching. For the non-verbal inhibition task, behavioral outcome was not affected despite electrophysiological stimulation-induced changes. Overall, the results suggest that the DLPFC plays a role in top-down cognitive control in language production, but that this role is not increased with higher cognitive demand such as naming in a second language or in language switching during word translation.

Unsupervised Word Translation with Adversarial Autoencoder

Crosslingual word embeddings learned from monolingual embeddings have a crucial role in many downstream tasks, ranging from machine translation to transfer learning. Adversarial training has shown impressive success in learning crosslingual embeddings and the associated word translation task without any parallel data by mapping monolingual embeddings to a shared space. However, recent work has shown superior performance for non-adversarial methods in more challenging language pairs. In this article, we investigate adversarial autoencoder for unsupervised word translation and propose two novel extensions to it that yield more stable training and improved results. Our method includes regularization terms to enforce cycle consistency and input reconstruction, and puts the target encoders as an adversary against the corresponding discriminator. We use two types of refinement procedures sequentially after obtaining the trained encoders and mappings from the adversarial training, namely, refinement with Procrustes solution and refinement with symmetric re-weighting. Extensive experimentations with high- and low-resource languages from two different data sets show that our method achieves better performance than existing adversarial and non-adversarial approaches and is also competitive with the supervised system. Along with performing comprehensive ablation studies to understand the contribution of different components of our adversarial model, we also conduct a thorough analysis of the refinement procedures to understand their effects.

Mutual intelligibility of Finnish and Estonian vocabulary

This paper reports on an investigation on Finnish-Estonian mutual intelligibility carried out by means of a word translation task among 307 Finnish and 118 Estonian participants. The results confirm previous findings by Kaivapalu (2015) that both in the spoken and in the written mode Estonians understand isolated words in Finnish better than Finns understand isolated words in Estonian. Older participants performed better than younger participants. Many of the participants had at least some previous exposure to the test language and were therefore familiar with some words and sound correspondences. However, we were also interested in testing how well speakers of the two languages could understand the other language purely on the basis of the similarity of the languages. We therefore separately analyzed the data produced by a subset of participants with no or little exposure to the test language. The results were still asymmetric, but only in the spoken mode. We looked at various linguistic and non-linguistic factors that could explain why some cognates are more easily recognized than others. To put our results into perspective, we compared them to the results of the similar experiments involving 70 Germanic, Romance and Slavic language combinations.

The Effect of Lexical Frequency in Modulating Translation Asymmetric on L2 Learners

Mao Cui, Eunhae Oh. 2018. The effect of lexical frequency in modulating translation asymmetries on L2 learners. Language and Information, 22.1, 57-71. The study examined response time in Chinese-English two-way translation in order to explore the role of word frequency in modulating translation asymmetries between forward and backward translation. 23 Chinese learners of English at two different proficiency levels performed a word translation task of 90 concrete nouns, one-syllable in English words and two characters in Chinese words. The results showed that there were asymmetries between the two translation directions, with backward translation taking significantly longer time than forward translation. Word frequency played a role in modulating the asymmetries in that translating high-frequency items, regardless of direction, took significantly shorter time than the mid- and low-frequency items. But backward translation displayed larger magnitude of sensitivity to the variation of frequency than forward translation. It was suggested that both forward and backward translation were conceptually-mediated with differing strength in the linking between the lexical items and their conceptual representations, being stronger with high-frequency items than mid- and low-frequency items. Conceptual activation is easier for high frequency items.

The architecture of the mental lexicon and the selection of lexical nodes

This study presents the architecture of the mental lexicon of third language learners by focusing on three representation levels: letter, word and language. In particular, this analysis attempts to examine the extent of the influence of the first and second languages known by bilingual learners of English. The study is guided by Dijkstra’s (2003) Multilingual Interactive Activation (MIA) model, and the hypothesis of the language selective or language nonselective access of third language learners is tested. The method involved in this analysis is the word translation task as a tool for investigating the organization of the mental lexicon. The results obtained as a result of the translation task claim that trilingual speakers can operate with three languages during the process of learning.Keywords: multilingual processing, mental lexicon, language typology.

It Is Not Necessary to Retrieve the Phonological Nodes of Context Objects for Chinese Speakers.

The issue of how activation is transmitted from semantic to phonological level in spoken production remains controversial. Recent evidences from alphabetic languages support a cascaded view. However, given the different architecture of phonological encoding in non-alphabetic languages, it is not clear whether this view applies in Chinese, as a non-alphabetic script. We therefore investigated whether the not-to-be named pictures activate their phonological properties in Chinese speech production. In Experiment 1, participants were presented a target English word and a context picture (semantically related or unrelated, phonologically related or unrelated to target word in Chinese) and were asked to translate the English word into a Chinese word. The translation latencies were faster in the semantically related condition than in the unrelated condition. By contrast, no difference between phonologically related and unrelated was observed. In Experiment 2, in order to promote participants phonological sensitivity in a word-translation task, we increased the proportion of phonologically related trials from 25 to 50%. In Experiment 3, we employed a word association task that was more sensitive to phonological activation of context objects than a word translation task. The phonological activation of context objects were absent again in Experiments 2 and 3. Bayes Factor analysis suggested that the absence of phonological activation of context pictures was reliable. Results consistently revealed that only target lemma could activate the corresponding phonological node to guide articulation whereas no phonological activation of non-target lemma’s in Chinese. The present findings thus support a discrete model in Chinese spoken word production, which was contrastive with the cascaded view in alphabetic languages production.

Directionality and context effects in word translation tasks performed by conference interpreters

AbstractProfessional interpreters employed by international institutions usually work into their L1 from their L2, while freelance interpreters tend to work both into and from their L1. A study was devised to see if the long-term interpreting unidirectional practice (in the L2–L1 direction only), in contrast to bidirectional practice (in the L2–L1 and L1–L2 direction), influences the speed of lexical retrieval manifested through shorter translation latencies. Forty-eight professional conference interpreters produced oral translations of nouns presented in isolation, in high context constraint sentences and in low context constraint sentences. Contrary to predictions, unidirectional interpreters did not manifest directionality asymmetry and their L2–L1 translation latencies were not shorter than L1–L2 translation latencies. Surprisingly, the L2–L1 direction advantage was found in the group of bidirectional interpreters. The data suggest that the dominant directionality in interpreting practice has little impact on the strength of interlingual lexical links in the interpreter’s mental lexicon or that other factors (such as language use, exposure and immersion) might offset any such impact. The study also revealed an expected context effect, which shows that interpreters use semantic constraint to anticipate sentence-final words.

Mutual intelligibility between West and South Slavic languages

In the present study we tested the level of mutual intelligibility between three West Slavic (Czech, Slovak and Polish) and three South Slavic languages (Croatian, Slovene and Bulgarian). Three different methods were used: a word translation task, a cloze test and a picture task. The results show that in most cases, a division between West and South Slavic languages does exist and that West Slavic languages are more intelligible among speakers of West Slavic languages than among those of South Slavic languages. We found an asymmetry in Croatian-Slovene intelligibility, whereby Slovene speakers can understand written and spoken Croatian better than vice versa. Finally, we compared the three methods and found that the word translation task and the cloze test give very similar results, while the results of the picture task are somewhat unreliable.

Language Selectivity in Lexical Access: an Experimental Study on Bilinguals

In this study, the locus of language selection in bilingual lexical access is investigated based on some basic factors such as first languages, second language proficiency, age of acquisition and multilingualism. In particular, this study explores competition between bilinguals’ languages and proposes two language selection models; Inhibitory Control Model (Green, 1998) and Concept Selection Model (La Heij, 2005). In experiment 1, the participants were asked to perform a word translation task from their second language (L2) to first language (L1). Each target was accompanied by a distracter item in the form of a picture or a word which was related/unrelated to the target word semantically. As a result, all participants translated target words faster when they are accompanied with semantically related/unrelated word distracters than picture distracters. On the other hand, they translated target words faster when they are accompanied with unrelated word distracters than related word distracters. Finally, they translated target words faster when they are accompanied with related picture distracters than unrelated picture distracters. In experiment 2, participants were asked to perform a switching task with the numbers in their first language and second language according to the background color of the digits. Finally, the language switching cost was larger when switching from L2 to L1 than vice versa. The results have shown that while the factors such as L1 andage of acquisition do not affect the locus of language selection during lexical access, proficiency in L2 and multilingualism factors affect the locus of language selection.

Mutual intelligibility of Dutch-German cognates by children: The devil is in the detail

AbstractSeveral studies (e.g., Ház 2005) have found German to be easier to understand for Dutch listeners than Dutch for German listeners. This asymmetry has been attributed to the fact that German is an obligatory subject in Dutch secondary school and that many Dutch people watch German television. In contrast, it is much less common for German children to learn Dutch at school and for German people to watch Dutch television. It cannot be excluded, however, that in addition to the extralinguistic factor of language contact, linguistic factors also play a role in the asymmetric intelligibility between German and Dutch. The present study aimed at gaining insight into the phonetic-phonological factors playing a role in Dutch-German intelligibility at the word level for speakers of the respective languages in a first confrontation (i.e., assuming no prior language contact).We presented highly frequent Dutch and German cognate nouns, recorded by a perfect bilingual speaker, to Dutch and German children between 9 and 12 years in a word translation task. The German and Dutch children were comparable in that they did not know the other language or a related dialect and expressed equally positive attitudes towards the other language, its speakers and the country. It was thus ensured that language contact and language attitude could not play a role in the present study.Our results revealed that the Dutch subjects were significantly better at understanding the German cognates (50.2% correct translations) than the German subjects were at understanding the Dutch cognates (41.9%). Since the relevant extra-linguistic factors had been excluded, the asymmetry must have a linguistic basis. A thorough analysis of the 16 cognate pairs with an asymmetry larger than 20% showed that (combinations of) neighbors (lexical competitors), phonetic detail and asymmetric perceptions of corresponding sounds play a major role in the explanation of the asymmetry.

Raspberry, not a car: context predictability and a phonological advantage in early and late learners' processing of speech in noise.

Second language learners perform worse than native speakers under adverse listening conditions, such as speech in noise (SPIN). No data are available on heritage language speakers’ (early naturalistic interrupted learners’) ability to perceive SPIN. The current study fills this gap and investigates the perception of Russian speech in multi-talker babble noise by the matched groups of high- and low-proficiency heritage speakers (HSs) and late second language learners of Russian who were native speakers of English. The study includes a control group of Russian native speakers. It manipulates the noise level (high and low), and context cloze probability (high and low). The results of the SPIN task are compared to the tasks testing the control of phonology, AXB discrimination and picture-word discrimination, and lexical knowledge, a word translation task, in the same participants. The increased phonological sensitivity of HSs interacted with their ability to rely on top–down processing in sentence integration, use contextual cues, and build expectancies in the high-noise/high-context condition in a bootstrapping fashion. HSs outperformed oral proficiency-matched late second language learners on SPIN task and two tests of phonological sensitivity. The outcomes of the SPIN experiment support both the early naturalistic advantage and the role of proficiency in HSs. HSs’ ability to take advantage of the high-predictability context in the high-noise condition was mitigated by their level of proficiency. Only high-proficiency HSs, but not any other non-native group, took advantage of the high-predictability context that became available with better phonological processing skills in high-noise. The study thus confirms high-proficiency (but not low-proficiency) HSs’ nativelike ability to combine bottom–up and top–down cues in processing SPIN.

Word reading and translation in bilinguals: the impact of formal and informal translation expertise.

Studies on bilingual word reading and translation have examined the effects of lexical variables (e.g., concreteness, cognate status) by comparing groups of non-translators with varying levels of L2 proficiency. However, little attention has been paid to another relevant factor: translation expertise (TI). To explore this issue, we administered word reading and translation tasks to two groups of non-translators possessing different levels of informal TI (Experiment 1), and to three groups of bilinguals possessing different levels of translation training (Experiment 2). Reaction-time recordings showed that in all groups reading was faster than translation and unaffected by concreteness and cognate effects. Conversely, in both experiments, all groups translated concrete and cognate words faster than abstract and non-cognate words, respectively. Notably, an advantage of backward over forward translation was observed only for low-proficiency non-translators (in Experiment 1). Also, in Experiment 2, the modifications induced by translation expertise were more marked in the early than in the late stages of training and practice. The results suggest that TI contributes to modulating inter-equivalent connections in bilingual memory.

Losing control of your languages: A case study

Bilingual speakers are usually quite good at restricting their lexicalization output to the desired language while preventing all sorts of language intrusions from the language not in use. However, brain damage can affect these abilities of language control, leading to striking and flagrant linguistic behaviours, such as pathological language mixing (pLM) and pathological switching (pLS). In this paper we report the performance of a Catalan–Spanish bilingual individual (R.R.T.) who, due to a neuroinflammatory disease and subcortical lesions, shows pLS. We tested R.R.T. in several tasks of language production and control, such as picture naming (objects and actions), word translation, blocked naming, and language switching task. R.R.T. was also tested in executive control (EC) tasks, such as task switching and a flanker task. We found several interesting results. First, cross-language intrusions were present much more frequently when R.R.T. was asked to speak in her first (and dominant) language (Catalan) than when she was asked to do so in the nondominant language (Spanish). Second, the results provide evidence suggesting that damage to certain subcortical structures may lead to problems in controlling the language output during verbalization in bilingual speakers. Third, we observed that R.R.T. seemed to show more difficulties in language control with verbs. Fourth, R.R.T. showed impaired performance compared to controls in both task switching and a flanker task. The results are discussed in relation to other findings of pLM and pLS in published single-case reports and in relation to EC deficits.

The Hands and Mouth Do Not Always Slip Together in British Sign Language

In contrast to the single-articulatory system of spoken languages, sign languages employ multiple articulators, including the hands and the mouth. We asked whether manual components and mouthing patterns of lexical signs share a semantic representation, and whether their relationship is affected by the differing language experience of deaf and hearing native signers. We used picture-naming tasks and word-translation tasks to assess whether the same semantic effects occur in manual production and mouthing production. Semantic errors on the hands were more common in the English-translation task than in the picture-naming task, but errors in mouthing patterns showed a different trend. We conclude that mouthing is represented and accessed through a largely separable channel, rather than being bundled with manual components in the sign lexicon. Results were comparable for deaf and hearing signers; differences in language experience did not play a role. These results provide novel insight into coordinating different modalities in language production.

The Distractor Picture Paradox in Speech Production: Evidence from the Word Translation Task

Several naming studies show that distractor pictures, even when intentionally ignored by the speaker, are still capable of activating their respective phonological representations. However results from word translation studies suggest that distractor pictures are only conceptually activated. Here we tested the reliability of the word translation results. In four experiments, bilingual participants translated words from one language to the other one while ignoring the presentation of pictures. In Experiment 1a phonological related pictures sped up translation latencies. However, the effect disappeared when the percentage of related trials was reduced (Experiment 1b). In Experiment 2a translation latencies were faster when the words were accompanied by semantically related pictures than by unrelated pictures. Importantly, the effect was still reliable when the proportion of related trials was reduced and the total number of semantic categories was increased (Experiment 2b). Theoretical implications of the influence of distractor pictures during speech production are discussed.

Cross-language lexical connections in the mental lexicon: Evidence from a case of trilingual aphasia

Despite anecdotal data on lexical interference among the languages of multilingual speakers, little research evidence about the lexical connections among multilinguals’ languages exists to date. In the present paper, two experiments with a multilingual speaker who had suffered aphasia are reported. The first experiment provides data about inter-language activation during natural conversations; the second experiment examines performance on a word-translation task. Asymmetric patterns of inter-language interference and translation are evident. These patterns are influenced by age of language learning, degree of language recovery and use, and prevalence of shared lexical items. We conclude that whereas age of language learning plays a role in language recovery following aphasia, the degrees of language use prior to the aphasia onset and of shared vocabulary determine the ease with which words are accessed. The findings emphasize the importance of patterns of language use and the relations between the language pair under investigation in understanding lexical connections among languages in bilinguals and multilinguals.