Exploring the association between emotional dynamics and performance in technology-mediated translation

Translation technologies have changed the professional workflows in the translation industry, bringing cognitive and emotional demands to translators. We investigated the association between student translators’ emotional dynamics and their performance in technology-mediated translation. Sentiment analysis was adopted to analyze student translators’ positive, neutral and negative emotions, as well as the polarity and subjectivity of their responses during the project. Our findings showed that neutral emotions took the largest proportion, followed by positive emotions, with negative emotions accounting for the smallest proportion overall. Descriptive week-by-week analyses showed variations in emotions: positive emotions decreased during the mid to late stages with an increase in negative emotions, and positive emotions recovered by the end of the project. Last but not least, Bayesian Linear Regression analysis suggested a potential positive association between student-level sentiment and technology-mediated translation performance, with higher sentiment scores associated with better performance. The findings highlight the importance of considering emotional factors in translator training alongside technical skills, particularly in technology-mediated learning environments. Both may be relevant to their performance and professional development in the changing landscape of the translation industry.