Control Language Research Articles

Interactions between language and inhibitory control: Evidence from a combined language switching and Stroop paradigm

Aims and objectives: This study assessed bilinguals’ language control (LC) and inhibitory control (IC) performance (switch costs and Stroop effects) simultaneously in the same participants to investigate how these processes influence each other. Design: Seventy-four Turkish-English bilinguals were presented with Turkish (L1) or English (L2) color words printed either in congruent or incongruent ink color and instructed to name the ink color of these words in the presentation language. Stimuli’s language and congruency were either the same as in the previous trial or different. Data and analysis: Reaction times (RTs), switch costs (mean RTs on language repetition subtracted from switch trials), and the Stroop effects (mean RTs on congruent subtracted from incongruent trials) were analyzed using the linear mixed-effect model. Findings: The switch costs were larger on incongruent than congruent, and the Stroop effects were larger on language switch than repetition trials. This means that the LC performance decreased while resolving conflict, and the IC performance decreased during switching language, indicating that these two share a common IC mechanism. However, the switch costs and Stroop effects across L1 and L2 were symmetrical in all conditions, leaving the previous interpretation uncertain. Besides, the Stroop effects were larger when followed by congruent than incongruent trials during language repetition, whereas they were equal during switching. This means that the ability to adjust performance by previous experience was disrupted during language switching. Moreover, for the high-L2 proficiency group, this ability was diminished in language repetition trials too. These results indicate that rather than inhibition, other processes may primarily mediate bilinguals’ LC. Originality: This study provides evidence for how language and inhibitory control influence each other by combining language switching and Stroop paradigms. Furthermore, it investigates the sequential congruency effects (SCE). Significance/implications: This study shows that SCE investigation may provide significant theoretical implications.

An in-depth analysis of the individual impact of controlled language rules on machine translation output: a mixed-methods approach

Examining the general impact of Controlled Language (CL) rules in the context of Machine Translation (MT) has been an area of research for many years. The present study focuses on the following question: how do CL rules impact MT output individually? By analysing a German corpus-based test suite of technical texts that have been translated into English by different MT systems, this study endeavours to answer this question at different levels: the general impact of CL rules (rule- and system-independent), their impact at rule level (system-independent) as well as at rule and system level. The results of five MT systems are analysed and contrasted: a rule-based system, a statistical system, two differently constructed hybrid systems, and a neural system. For this, a mixed-methods triangulation approach that includes error annotation, human evaluation, and automatic evaluation was applied. The data was analysed both qualitatively and quantitatively in terms of CL influence on the following parameters: number and type of MT errors, style and content quality, and scores of two automatic evaluation metrics. In line with many studies, the results show a general positive impact of the applied CL rules on the MT output. However, at rule level, only four rules proved to have positive effects on the aforementioned parameters; three rules had negative effects on the parameters; and two rules did not show any significant impact. At rule and system level, the rules affected the MT systems differently, as expected. Rules that had a positive impact on earlier MT approaches did not show the same impact on the neural MT approach. Furthermore, neural MT delivered distinctly better results than earlier MT approaches, namely the highest error-free, style and content quality rates both before and after applying the rules, which indicates that neural MT offers a promising solution that no longer requires CL rules for improving the MT output.

Evaluation of the impact of controlled language on neural machine translation compared to other MT architectures

Many studies have shown that the application of controlled languages (CL) is an effective pre-editing technique to improve machine translation (MT) output. In this paper, we investigate whether this also holds true for neural machine translation (NMT). We compare the impact of applying nine CL rules on the quality of NMT output as opposed to that of rule-based, statistical, and hybrid MT by applying three methods: error annotation, human evaluation, and automatic evaluation. The analyzed data is a German corpus-based test suite of technical texts that have been translated into English by five MT systems (a neural, a rule-based, a statistical, and two hybrid MT systems). The comparison is conducted in terms of several quantitative parameters (number of errors, error types, quality ratings, and automatic evaluation metrics scores). The results show that CL rules positively affect rule-based, statistical, and hybrid MT systems. However, CL does not improve the results of the NMT system. The output of the neural system is mostly error-free both before and after CL application and has the highest quality in both scenarios among the analyzed MT systems showing a decrease in quality after applying the CL rules. The qualitative discussion of the NMT output sheds light on the problems that CL causes for this kind of MT architecture.

Helping the helpers: Evaluating the impact of a controlled language checker on the intralingual and interlingual translation tasks involving volunteer health professionals

ABSTRACTCochrane is a non-profit organization which mainly relies on volunteer health professionals for the production, simplification, evaluation, and multilingual dissemination of high-quality health content. The approach that Cochrane volunteers adopt for the simplification (or intralingual translation) of English health content is non-automated and involves the manual checking and implementation of plain language guidelines. This study investigated whether and to what extent the introduction of a controlled language (CL) checker – which would make the simplification approach semi-automated – increased authors’ satisfaction and machine translation (MT) quality. Twelve Cochrane authors completed a standardized questionnaire and answered follow-up questions on their level of satisfaction and preferences. Forty-one Cochrane evaluators assessed the quality of the Spanish MT outputs of simplified texts. Authors showed a preference for the introduction of a CL checker. Differences in MT quality scores were slight.

Evaluating the Usability of a Controlled Language Authoring Assistant

AbstractThis paper presents experimental results of a usability evaluation of a controlled language (CL) authoring assistant designed to help non-professional writers create machine translatable source texts. As the author drafts the text, the system detects CL rule violations and proscribed terms. It also incorporates several support functions to facilitate rephrasing of the source. In order to assess the usability of the system, we conducted a rewriting experiment, in which we compared two groups of participants, one with the aid of the system and the other without it. The results revealed that our system helped reduce the number of CL violations by about 9% and the time to correct violations by more than 30%. The CL-applied source text resulted in higher fluency and adequacy of MT outputs. Questionnaire and interview results also implied the improved satisfaction with the task completion of those participants who used the system.

The use of simplified English language for writing technical documentation: The proposal for a research project

Natural language permits an enormous amount of expressive variation. Writers, especially technical writers, tend to develop special vocabularies (jargons), styles, and grammatical constructions. Technical language becomes opaque not just to ordinary readers, but to experts as well. The problem becomes particularly acute when such text is translated into another language, since the translator may not even be an expert in the technical domain. Controlled Languages (CL) have been developed to counter the tendency of writers to use unusual or overly-specialized, inconsistent language [1, Section 7.6]. Engineering professionals know that making text understandable is very challenging in domestic environment (Serbian language), and especially in an international environment (mostly English language). Simplified Serbian language could help in domestic environment. Simplified English language already helps in international environment. Simplified English is sometimes used as a generic term for a controlled language. Simplified English is the original name of a controlled language originally developed for aerospace industry to facilitate the use of maintenance manuals by engineers for whom English is not the native language. It is now officially known under its trademarked name as Simplified Technical English (STE). Although STE is developed for use in the aerospace and defense industries, other industries have used it as a basis for developing their own controlled English standards, which are mostly used for making appropriate technical documentation.

A controlled language-based evaluation approach to ensure image accessibility during web localisation

In spite of recent improvements in non-visual web access, images on the web still present an accessibility barrier to screen reader users. For this population group the presence of inappropriate text alternatives for images, or simply their absence, usually results in a poor web user experience. In this paper, we propose a controlled language (CL) rule-based approach that enables translation professionals to ensure image accessibility during the web localisation process. We describe the set of 40 CL rules we developed and then present the results of the evaluation of a selection of ten rules from the set. During the study, which sought to assess their impact on the appropriateness of text alternatives in French, the ten rules were applied usingAcrolinx, a state-of-the-art CL checker. The results of the evaluation suggest that this sub-set of ten rules can help translators significantly improve the level of image accessibility obtained in the localised web product.

Computerization of a ‘Controlled Language’ to Write Medical Standard Operating Procedures (SOPs)

Abstract Accreditation of hospitals includes items regarding the existence of Standard Operating Procedures (SOPs); but these documents can be sources of misunderstanding and patient's safety may be jeopardized. We proposed a solution based on the Controlled Language (CL) concept and developed software services to make CL user-friendly to writers. We carried out: 1) deep linguistic analysis of SOP corpora in two medical domains; 2) language modelling to establish two adapted CLs, 3) improvement of home-made CL Authoring Software by developing software modules and a collaborative corpus-based web-accessible platform for the building of terminological and non-terminological resources, 4) evaluation through focus groups and computer-aided CL-writing test sessions. Health professionals and linguists cooperated closely in a field that is quite new to the health domain. The optimized Prolipsia CL Authoring Software appeared to be a good compromise between users’ needs and CL requirements. All actors agreed that benefits would be gained by using the proposed tools, in terms of patient safety and of work organization, institutional cohesion, and decreased liabilities. They also suggested that software solutions able to analyse the quality of existing texts and help correct them would better fit the situation of institutions which have already got a large corpus of (unsatisfactory) texts at their disposal. Such software is currently at an advanced stage of development, with a first version available.

The role of executive control in bilingual language production: A study with Parkinson's disease individuals

The basal ganglia are critically involved in language control (LC) processes, allowing a bilingual to utter correctly in one language without interference from the non-requested language. It has been hypothesized that the neural mechanism of LC closely resembles domain-general executive control (EC). The purpose of the present study is to investigate the integrity of bilingual LC and its overlap with domain-general EC in a clinical population such as individuals with Parkinson's disease (PD), notoriously associated with structural damage in the basal ganglia.We approach these issues in two ways. First, we employed a language switching task to investigate the integrity of LC in a group of Catalan–Spanish bilingual individuals with PD, as compared to a group of matched healthy controls. Second, to test the relationship between domain-general EC and LC we compared the performances of individuals with PD and healthy controls also in a non-linguistic switching task. We highlight that, compared to controls, individuals with PD report decreased processing speed, less accuracy and larger switching costs in terms of RT and errors in the language switching task, whereas in the non-linguistic switching task PD patients showed only increased switching cost in terms of errors. However, we report a positive correlation between the magnitudes of linguistic and non-linguistic mixing costs in individuals with PD. Taken together, these results support the notion of a critical role of the basal ganglia and connected structures in LC, and suggest a possible link between LC and domain-general EC.

Adaptations in Controlled Cultural Writing for Effective Machine Translation: A Register-specific Probe

Machine translation (MT) serves as a fast way of transmitting web information across countries and MT performance can be highly improved by adapting the source text using a controlled language (CL). Drawing on Halliday’s register-specific accounts as the theoretical framework, this paper investigates multi-faceted adaptations that undergo in the controlled cultural writing (CCW) considering the three dimensions of a language activity, such as field, tenor and mode. By analyzing 296 controlled sentences adapted from 22 folk cultural texts of online Encyclopedia of Taiwan , along with their corresponding good/excellent-level English machine translations, as a case study, this paper identifies some register-specific rules for adaptations. The findings show that variation in grammatical adaptations in CCW, related to the linguistic differences between English and Chinese, supports the mode of online MT application. The paraphrasing of all cultural references for clear transmission of cultural information helps achieve the function of intercultural communication (field), and the conversion from a heavy into lighter context meets western audiences’ pragmatic expectations (tenor). Above all, the MT-driven adaptations in CCW are performed within a context where the interplay of complex factors, such as the function of intercultural communication, cultural inadequacy of western audiences, and MT operational constraints, are taken into account for the creation of readable, scannable cultural machine translations to serve western audiences.

HVDC Intelligent Controller

The Purposes of this paper describes Gezhouba to Nanqiao rectifier direct current transmission project PI controller, through the intelligent control theory and the combination of the traditional PI controller are intelligent controller—HVDC fuzzy-PI controller, using the improved genetic algorithm domain Fuzzy- PI controller's language of control theory. The result of improved intelligent controller has a simple structure, fast response characteristics. Using MATLAB / SIULIKMN the fuzzy toolbox and genetic algorithm toolbox for system simulation, simulation results show that the Conclusions of the intelligent controller has better adaptability and dynamic features to improve the robustness of the system and transient stability for the intelligent control of high voltage DC reference

Zombie language: The parody of newspaper columns in Quim Monz's radio commentaries

In postmodernist fiction and new journalism, parody is used to combat the manipulation of classical literary and journalistic genres, based on objectivism as a commonplace, which usually impose an unequivocal vision of facts preventing any kind of dissidence. This shrewd control of language, revealed in clich use, also characterizes political discourse. Parody has been studied in Quim Monz's fiction and non-fiction, but not in his radio and television shows. The aim of this article is to examine parody in the radio show Articles que no farem/Articles we won't write (20052006), where Monz imitates the trite style of archetypal columnists issuing banal opinions on society, culture and politics. Rather than express original ideas, they merely strengthen mainstream ideas using commonplaces. Monz's parody mainly operates through the manipulation of individuals and their column titles; he also uses humoristic devices such as paronomasia and comparison. As a complement to previous research into Monz's journalism, it will help us understand his conception of columns and of the mainstream ideas he humorously censures, particularly political correctness.

Utilising scripting language for unmanned and automated guided vehicles operating within row crops

A flexible high-level control language is an important element in the ongoing task of introducing automated guided vehicles (AGV) to new application domains. A new application domain is row crops, where small AGV's will perform weed control around individual crop plants. This paper defines the requirements and scope of a process- and behaviour-based scripting language needed to control the weeding AGV in an agricultural row crop. The goal is to traverse and cover the whole field with no human auxiliary input during the field operation. The basis is the transparent and tactical real-time control language (CL) for small mobile robots (SMR). This SMR-CL has been modified to include some necessary motion commands and a supplemental supervisory function to monitor and record the progressive coverage of the field. The control language was then tested by applying it to a scenario representing typical field conditions for row crops. The construction of a suitable SMR-CL script for use in a field clearly demonstrates the feasibility of adapting behaviour-based control systems to field structures. The conducted case study indicated the importance of including goal-directing modules. Such a module is described here as the ‘supervisory field coverage monitor’ (SFCM), which acts to coordinate the behaviours. The applicability of this modified SMR-CL has been successfully demonstrated using a vehicle test in a specially designed artificial row crop field. The analysis of the operational performance verified that it is possible to cover all rows in a field without conducting time-consuming planning procedures.

Predicting potential COBOL performance on low level machine architectures

As a COBOL host, a computer architecture should efficiently execute those language constructs that are most frequently used in actual programs. However, when the language's control and data structures are at a far higher level than the control and data structures of the underlying machine, the compiler designer is faced with a large number of potential choices for mapping these high level structures to the low level architecture. This is the case for implementing COBOL on a typical mini-computer architecture. Restricting the choices to an implementable set meeting a given performance criteria and predicting the resulting performance level requires a complex analysis. This paper gives an example of the methodology used in one such analysis -- the performance of COBOL on low level architectures. This area was chosen because of the difficulty of the problem caused by the lack of support for COBOL constructs and data types on low level architectures used in most mini-and micro-computers. The Motorola MC68000 and a RISC-type architecture were selected as representative architectures.