Reading for comprehension vs. sight interpreting/translation: Revisiting the influence of purposes

Reading behaviour has been shown to be affected by the purpose of reading (Alves, Pagano & da Silva 2011; Jakobsen & Jensen 2008; Macizo & Bajo 2004; Shreve, Lacruz & Angelone 2010; Wang & He 2018). However, most studies rely on task-based indices, such as task time, number of fixations, regression rate, or mean fixation duration, while analysis of different reading stages is largely absent. This study examined the reading behaviour in three tasks—silent reading (SR), reading aloud (RA), and unprepared, self-paced sight interpreting/translation (SiT, also known as sight translation in the literature). Three groups of participants were invited, including 17 experienced interpreters, 18 interpreting students, and 18 untrained bilinguals. The materials were three texts of 175 words, adapted from three diplomatic speeches on the same topic, and global and local eye-movement data were analysed. Overall, SiT took longer time and more fixations to complete than RA and SR, with the latter two being statistically equal. This finding generally corroborates previous research. Yet, this study offers more by looking at how reading was done in various stages: the effort of vocalisation in RA was clearly visible early on, leading to longer first fixation duration (FFD) and gaze duration (GD) than SR and SiT; while FFD was identical in the latter two tasks, GD was shorter in SiT, which could be a signal of the participants’ attempt to accelerate reading to begin the next stage of processing sooner. On the other hand, longer fixation durations in later reading stages evidently delineated the additional cognitive processes involved in SiT. While the effort of vocalisation affected all groups, participants, with training, managed to shorten the average fixation duration in later-stage reading to the same level as the other less demanding tasks.