Translation, Cognition & Behavior - Volume 7, Issue 2, 2024

The aim of the study was to investigate the effects of interpreter experience and interpreting direction on cognitive load of dialogue interpreters. The study was conducted on two groups of interpreters (experienced and inexperienced) during a simulated interpreter-mediated encounter and cognitive load was operationalized with disfluency durations, and disfluency counts. The data suggest that both experience and directionality modulate cognitive load in dialogue interpreters. All participants manifested an increase in cognitive load while interpreting into their weaker language (L2), which suggest that directionality had an effect on cognitive load of dialogue interpreters. The results also suggest that inexperienced interpreters are likely to exhibit higher cognitive load compared to experienced interpreters. The study underscores the significance of experience and directionality in understanding the cognitive demands faced by dialogue interpreters, contributing valuable insights to research on interpreting and cognitive load.

This study examines how morphological complexity affects cognitive effort in neural machine translation (NMT) post-editing across six languages. Analysis of the DivEMT dataset shows that morphologically richer target languages like Ukrainian and Turkish require more editing time, keystrokes, and frequent pauses, indicating higher cognitive demands. Vietnamese, despite simpler morphology, also showed high cognitive effort, suggesting other factors like syntax influence processing load. Mean Size of Paradigm (MSP) analysis confirmed Ukrainian and Turkish’s high morphological complexity compared to isolating languages like Vietnamese. Higher error rates in morphologically rich languages demonstrate increased editing needs. While user perceptions varied, the data reveals that greater linguistic distance correlates with higher cognitive effort in NMT post-editing, showing typological divergence impacts beyond morphology alone.

In the ever-changing language industry, translation experts are required to be ‘adaptive agents’. This study uses metaphors of self-concepts (i.e. personal metaphors) to access the cognitive aspects of the self that are enacted by students and professionals before and after performing a task. Translators whose metaphors show richness are expected to perform better, with their metaphors changing as they are exposed to different tasks. Such metaphor dynamicity is thus thought to be related to adaptability.

Personal metaphors were collected via linguistic prompts and naturally in questionnaires/interviews about participants’ backgrounds and professional opinions. While the study found a relation between metaphor dynamicity and performance, no definitive conclusions could be drawn on metaphor richness. This led to the hypothesis that dynamic personal metaphors of adaptability (e.g., a translator is a chameleon; a jack of all trades) can be linked to better performance. The study also suggests that incorporating metaphors and self-concepts in translator training could enhance adaptability in translation.

Speech fluency and comprehensibility have been examined more thoroughly in L1 and L2 research than in interpreting studies. We studied the fluency and comprehensibility of 25 interpreter trainees, whose spontaneous English speech samples were rated on 9-point scales by 6 American and 6 Persian judges, who were either interpreting experts or lays. Samples were rated once as originally recorded, and a month later with the silent pauses edited out. Native English judges proved less lenient (−.21) in their fluency ratings than the Persians; no nativeness effect was found for comprehensibility. Editing out pauses increased the fluency ratings (+.71) across all four rater groups but comprehensibility was not significantly affected (+.10). Both American and Persian experts rated the edited versions as more fluent (+1.80) than lays (+1.03). Experts may judge fluency more analytically, which will be helpful to them when providing feedback.

This paper critically discusses how crowdsourcing relate to the emergence of the so-called translation augmentation paradigm. As a broad generalization, these two technology-driven phenomena emerged as possible solutions to achieve better, more efficient or speedier solutions to complex problems than those that single individuals can solve on their own. They both can be contextualized within 4EA approaches to cognition, reframing problem-solving towards a distributed and extended perspective that goes beyond the limitations of the individual mind. Nevertheless, augmented translation refers to the coupling of human and machines in a technology-centric approach that attempts to increase the cognitive capabilities of translators, who remain in the loop and in charge (Lommel 2020; Angelone 2023). It relates to existing technological integrated systems beyond just the automation brought by TM and MT that learn and adapt from humans. The paper (1) interrelates distributed and extended cognitive approaches and crowdsourcing (Risku, Pein-Weber & Rogl 2016; Jiménez-Crespo 2017; Risku & Windhager 2020) in relation to AI and intelligence augmentation (IA), (2) analyzes the existing fuzziness in the theoretical and conceptual constructs to translation augmentation, (3) explores issues related to the locus of control and the center of the cognitive system in augmented crowdsourcing, as well as (4) looks ahead by introducing unexplored areas such as social augmentation. In doing so, the paper will attempt to answer whether human translators in crowdsourcing initiatives are, in fact, “in charge” and – or augmented.

We present an exploratory observational study on A-B interpreting in the European Parliament. In contrast to popular belief, it is a widespread practice for some of the newer EU languages to be interpreted into English by native speakers of the source language. The phonetic-acoustic properties of speech in English by eight Polish retour interpreters are assessed on the basis of material extracted from a large bidirectional EP-Poland corpus. We use the methodology from L2 speech research, based on identifying segmental and suprasegmental departures from native norms. The speech samples were inspected aurally and visually by means of spectrogram and waveform. The general detected departures emerge mostly from the differences between Polish and English sound systems. Although some of the interpreters may have more native-like accent than others, all the analysed interpretations are phonetically proficient and fully intelligible. In three of the interpreters, interesting regularities concerning pronunciation fluctuations were discovered.