Abstract Cognitive accessibility in video games warrants further exploration, as it encompasses a wide range of disorders and learning difficulties that may benefit from easy-to-understand (E2U) language. This study investigates players’ reception of E2U language, comparing individuals with and without dyslexia. A video game was developed, and participants were assigned to one of four conditions in a 2×2 experimental design, based on the type of language used in the game (standard or E2U) and the presence or absence of dyslexia. Both quantitative and qualitative data were collected, with a particular focus on player experience and language acceptability. Players with dyslexia reported a more positive experience when playing the E2U version, whereas players without dyslexia reported largely similar experiences across both versions. Nonetheless, participants in both groups expressed a general preference for the standard language. Overall, the findings of this exploratory study suggest that E2U language has the potential to enhance the gaming experience for players with dyslexia.
Abstract Social network analysis (SNA) posits that individuals’ professional activities are shaped by their relationships with other actors and should therefore be understood through a network-based perspective. With generative artificial intelligence (GenAI) increasingly demonstrating agentic features and becoming integrated into translation workplaces, translators are required to interact with both human stakeholders and GenAI tools, which complicates their work. To investigate how translators’ workplace practices are shaped in the GenAI era, this explorative study adopts an ego-centred SNA approach and employs the three-concentric-circles model in semi-structured interviews with seven in-house translators in China. By examining the structure of translators’ professional networks, the study elucidates their interactions with both human stakeholders and GenAI within their workplace settings. It further provides practical insights into the design of AI-integrated translation workflows from a socio-technical perspective and offers implications for translation training programmes in preparing students for AI-integrated translation workplaces.
Abstract We conducted a comparative reception analysis of Spanish‑to‑Czech translations by OpenAI’s GPT‑3.5, GPT‑4, and DeepSeek‑V3 across two text domains (marketing and literary), two evaluation criteria (naturalness and grammar), and two prompting strategies (simple vs. detailed). Additionally, the consistency of these translations was assessed using the Levenshtein distance metric. A reading task in which 132 Czech native speakers gave ratings of the translations on a 5‑point Likert scale revealed that, contrary to previous findings, simple prompts produced reliably higher‑quality translations than detailed prompts. Next, literary translations were rated lower than marketing translations, and grammar ratings exceeded those for naturalness. Additionally, GPT‑4 outperformed the other two models on the literary translations only. Finally, DeepSeek‑V3 showed greater consistency but lower quality of literary translation, suggesting that increased consistency may be at the expense of creativity. These findings provide empirical insights into how prompting strategy, text type, and model choice may influence machine translation quality.
