Dr Özlem Temizöz

The advent of AI-supported, cloud-based collaborative translation platforms has enabled a new form of online collaborative translation — ‘concurrent translation’ (CT). CT refers to commercial translation performed on such platforms by multiple agents (translators, editors, subject-matter experts, etc.) simultaneously, via concurrent access. Although the practice has recently gained more ground, research on CT is scarce. The present article reports on selected key findings of a study that investigates translators’ experiences with CT via a survey of 804 professional translators working in CT mode across different commercial platforms. Despite the affordances such as peer learning, positive competition, speed, flexibility of the volume of work and working time, and reduced responsibility and reduced stress, CT workflow comes with its substantial challenges such as time pressure, negative competition, reduced self-revision and research, all of which result in quality compromised for speed.

The optimal approach to establishing communication between healthcare providers and
patients who do not share the same language has been discussed for quite some time; the
debates have ultimately served to highlight the crucial role that linguistically and culturally
competent human interpreters play in these instances of language-discordant interpersonal
healthcare communication.
However, there are many barriers to accessing human interpreters in healthcare settings,
including waiting times, lack of financial resources, restricted local availability of qualified/
trained interpreters in some languages and, at times, a lack of practicability ( for instance,
spontaneous communication in a hospital ward). Alternative approaches to accessing human
interpreters and approaches to providing language support without the need for an
interpreter has been developed, which involves drawing on and interacting with different
types of technology. They range from the use of audio and video communication tools enabling
remote interpretation to volunteer language support through crowdsourcing via digital
platforms, machine translation ( MT), and bespoke translation tools/ apps. Some of these
approaches were initially developed for situations of medical emergency and/ or humanitarian
crisis but have subsequently been explored in other settings of interpersonal healthcare
communication, shifting the patterns of interaction in these settings from human- to- human
interaction towards increasing involvement of human-machine interaction.
This chapter will review technology-assisted approaches to facilitating language-discordant
interpersonal healthcare communication and explore the extent to which they
are currently capable of meeting patients’ and healthcare providers’ communication needs.
The focus is on situations of dialogic communication and interaction normally involving an
interpreter, a patient and/ or caregiver and one or more healthcare providers. Technology-assisted
methods of translating written texts are not included ( on the latter, see Dew et al.
2018; Haddow, Birch and Heafield 2021; Vieira, O’Hagan and O’Sullivan 2021). Following
a brief epistemological consideration of technology use in interpersonal healthcare settings,
each approach will be considered in a separate section, beginning with a review of the relevant
scholarly literature and main practical developments, followed by a discussion of critical
issues and challenges arising.

This study compares the quality of postediting performed by
subject-matter experts as opposed to professional translators. A
total of 10 professional translators and 10 engineers postedited a
482-word technical text pre-translated from English into Turkish
using data-based machine translation system, Google Translate.
The findings suggest that, for this particular task (technical
translation), translators’ and engineers’ postediting quality is
similar as far as the categories of mistranslation, accuracy, and
consistency are concerned. Engineers performed significantly
better than translators only in the terminology category. In the
language category, translators made significantly fewer (minor)
errors than engineers. The qualitative data analysis revealed that,
for this particular task, a degree in translation does not directly
correlate with postediting quality, unless it is combined with
subject-matter knowledge and professional experience in
translation. Finally, the present study indicates that – both for the
engineers and the professional translators – expertise and
experience in the subject matter are important factors
determining postediting quality.

Counting and not counting recurring errors are two different methods that have been employed in translation quality evaluation without paying due attention to how the difference between the results of each method, if any, affects the quality score of the end product, thereby affecting the validity of the quality evaluation method in question. This paper reports on a study which shows that penalizing or not penalizing recurring errors in the target text significantly affects the quality score. The results reveal a need for a more critical approach in handling recurring errors in translation quality evaluation.

The paper reports on a pilot study conducted to test the methodology to
replicate the study by Jensen & Pavlović (2009) which investigates the effect
of translation directionality on cognitive processing by means of eye-tracking.
The following hypotheses are retested: (1) In both directions of translation,
processing the TT requires more cognitive effort than processing the ST;
(2) L2 translation tasks on the whole require more cognitive effort than L1
tasks; (3) cognitive effort invested in the processing of the ST is higher in
L1 translation than in L2 translation; (4) cognitive effort invested in the
processing of the TT is higher in L2 translation than in L1 translation. The
results showed that the findings of three out of four hypotheses were the
same as the findings of Jensen & Pavlović (2009). Both studies suggest that
neither processing the texts in L2 (ST or TT) nor translation into L2 leads to
a higher amount of cognitive effort. The findings are important in that they
challenge the traditional view of directionality that is based on traditional
assumptions rather than empirical data. This pilot study is distinctive in that
it is the first study in Turkey that uses eye-tracking to explore the translation
process (Temizöz 2009).

In parallel with the rise of MT and the integration of machine-translated segments into the translation workflow as TM input, empirical research on MT has gained momentum from the turn of the new millennium. This report covers empirical studies on machine translation and the postediting of MT output. It includes a synoptic table giving the author and year of each experiment, the number of participants, brief information on the participant profile, the type of text used, the number of words in the texts, language pair and direction used, and the name of the machine translation and/or translation memory system used.

Technological innovations bring about new practices, urging translators to transform their work processes. Our knowledge of these processes is based on research, which is related to the tools/workflows prevalent at the time. The advances in technology necessitate constant re-examination of the translation process to understand rapidly changing, contemporary translation production methods. The present article gives us a glimpse into how the translation process changes when translators work collaboratively and concurrently within technological environments that enable synchronous horizontal and vertical collaboration on a single text. It presents reported data on process-related aspects from an online survey of 804 translators. Findings suggest that, in the concurrent workflow, the translation process is very different compared to when an entire text is translated by an individual translator. Notable process-related changes reported include a shift in the distribution of time and tasks across the phases of the translation process, limited self-revision during the drafting phase, earlier segment confirmation, precluded end-[self-]revision, and decreased research time.

The aim of this article is to illustrate the goals, process and results of the first stage of a European project: MentalHealth4All. Partners from 9 different European countries have worked together to design a multilingual validated repository targeted at third-country nationals with limited language proficiency (LLP TCNs) and healthcare providers, which contains resources to improve this population’s knowledge and understanding of how to access interlinguistic support in mental healthcare services in the main languages of the countries involved in the project. The present article provides a detailed overview of the assessment process of resources and its main results.

There is a strong need for language support in mental health services.•Migrants, refugees and healthcare professionals are not aware of language support options.•Systemic, interpersonal, and intrapersonal factors affect uptake of language support options.•Improving language support and cultural competency in mental health services is essential.•Seeking, providing and accessing mental health services is a complex system.. Migrants and refugees may not access mental health services due to linguistic and cultural discordance between them and health and social care professionals (HSCPs). The aim of this review is to identify the communication needs and barriers experienced by third-country nationals (TCNs), their carers, and HSCPs, as well as the strategies they use and their preferences when accessing/providing mental health services and language barriers are present.. We undertook a rapid systematic review of the literature (01/01/2011 – 09/03/2022) on seeking and/or providing mental health services in linguistically discordant settings. Quality appraisal was performed, data was extracted, and evidence was reviewed and synthesised qualitatively.. 58/5,650 papers met the inclusion criteria. Both TCNs (and their carers) and HSCPs experience difficulties when seeking or providing mental health services and language barriers are present. TCNs and HSCPs prefer linguistically and culturally concordant provision of mental health services but professional interpreters are often required. However, their use is not always preferred, nor is it without problems.. Language barriers impede TCNs’ access to mental health services. Improving language support options and cultural competency in mental health services is crucial to ensure that individuals from diverse linguistic and cultural backgrounds can access and/or provide high-quality mental health services.

Background

Migrants and refugees with low language proficiency (LLP) have a higher risk of experiencing certain mental health disorders compared to non-migrant populations. They are also more likely to experience a lack of access to mental healthcare due to language- and culture-related barriers. As part of the MentalHealth4All project, a digital multilingual communication and information platform was developed to promote access to mental healthcare for LLP migrants and refugees across Europe. This paper describes the study protocol for evaluating the platform in practice, among both health and/or social care providers (HSCPs) and LLP migrants and refugees.

Methods

We will conduct a pretest-posttest cross-national survey study to perform the platform’s effect-evaluation (primary objective) and process-evaluation (secondary objective). The primary outcomes (measured at T0, T2 and T3) are four dimensions of access: availability, approachability, acceptability, and appropriateness of mental healthcare. Secondary outcomes (measured at T2) are: actual usage of the platform (i.e. tracking data), ease of use, usefulness of content, comprehensibility of information, attractiveness of content, and emotional support. Participants will be recruited from nine European countries: Belgium, Germany, Italy, Lithuania, the Netherlands, Poland, Slovakia, Spain, and the United Kingdom. Using convenience sampling through professional networks/organisations and key figures, we aim to include at least 52 HSCPs (i.e. 6-10 per country), and 260 LLP migrants (i.e. 30-35 per country). After completing a pretest questionnaire (T0), participants will be requested to use the platform and HSCPs will participate in an additional personalised training (T1). Next, participants will fill out a posttest questionnaire (T2), and will be requested to participate in a second posttest questionnaire (T3, about 6-8 weeks after T2) to answer additional questions on their experiences through a brief phone interview (T3 is optional for migrants/refugees).

Discussion

The findings of this prospective pretest-posttest cross-national study will deepen our understanding of how a multilingual platform may promote access to mental healthcare services for LLP migrants and refugees. If successful, this intervention could be used to improve access to mental healthcare services, as well as HSCPs’ competencies in delivering such services, for any LLP migrants and refugees across Europe (and beyond).

This study explores the use of Google Translate (GT) for translating mental healthcare (MHealth) information and evaluates its accuracy, comprehensibility, and implications for multilingual healthcare communication through analysing GT output in the MHealth domain from English to Persian, Arabic, Turkish, Romanian, and Spanish. Two datasets comprising MHealth information from the UK National Health Service website and information leaflets from The Royal College of Psychiatrists were used. Native speakers of the target languages manually assessed the GT translations, focusing on medical terminology accuracy, comprehensibility, and critical syntactic/semantic errors. GT output analysis revealed challenges in accurately translating medical terminology, particularly in Arabic, Romanian, and Persian. Fluency issues were prevalent across various languages, affecting comprehension, mainly in Arabic and Spanish. Critical errors arose in specific contexts, such as bullet-point formatting, specifically in Persian, Turkish, and Romanian. Although improvements are seen in longer-text translations, there remains a need to enhance accuracy in medical and mental health terminology and fluency, whilst also addressing formatting issues for a more seamless user experience. The findings highlight the need to use customised translation engines for Mhealth translation and the challenges when relying solely on machine-translated medical content, emphasising the crucial role of human reviewers in multilingual healthcare communication.

Collaborative translation has evolved significantly over time, especially in the last 20 years, driven by advancements in digital communication technologies. The evolution of technology contributed to the changes in the collaborative processes and enabled collaboration to assume different forms in terms of proximity of collaborators (on-site/remote), time factors (synchronous/a-synchronous), and configurations of collaborators (horizontal/vertical). This article aims to contribute to the understanding of the nature of collaborative translation as affected by evolving technologies and how translators are adapting to these changes. The focus will be on Concurrent Translation (CT), i.e., synchronous translation production activity carried out for commercial reasons on cloud-based collaborative platforms by multiple, predominantly trained translation professionals and on translators’ experiences with this new workflow based on a qualitative analysis of a survey of 804 translators.  

Migrants face higher risks of developing mental health issues, such as posttraumatic stress disorder and psychosis, compared to individuals without a migrant or refugee background. However, they can encounter significant barriers to accessing mental healthcare, due to language difficulties, lack of interpreters, cultural stigma and system-related barriers. Furthermore, limited knowledge of the healthcare system further restrict access. Providing culturally tailored online multilingual health materials can enhance their access to mental healthcare, with co-created animated videos being a promising medium. Therefore, this study aimed to: 1) identify key factors influencing the evaluation of mental health educational videos by migrants and their healthcare providers, focusing on critical mental health themes, and 2) actively involve these groups in in co-creating the videos based on feedback.

Eight animated videos were developed through an iterative process involving the entire multidisciplinary MHealth4all consortium, including numerous feedback sessions. The videos were evaluated by using a hybrid approach that combined think-aloud observations and interviews with migrants (N = 38) and their healthcare providers (N =33). Thematic coding of the verbatim transcribed interviews revealed four essential aspects: comprehensibility, attractiveness, reliability, and relevance, with specific elements identified for both migrants and for their healthcare providers. These insights can guide the creation of effective health information materials for migrants facing mental health challenges.