This paper aims to investigate the eye-voice span (EVS), the distance between eyeandvoice,insighttranslatingmetaphoricalexpressions(MEs).24MAtranslation students, with no professional translation or interpreting exper...This paper aims to investigate the eye-voice span (EVS), the distance between eyeandvoice,insighttranslatingmetaphoricalexpressions(MEs).24MAtranslation students, with no professional translation or interpreting experience, were asked to conduct a sight translation (STR) task, and the processes were registered by eye-tracker and audio recorder. The qualified eye-tracking and audio data were further analysed by Tobii Studio and Audacity audio processing software. Our findings suggest that the time of the pause preceding an ME was largely, but not entirely, spent in processing the ensuing ME. However, due to the general existence of reading ahead activities in STR, the planning step for sight translating an ME takes place prior to the preceding pause; moreover, due to local processing difficulty causedbytheME,thetimeforreadingaheadintoME(temporalEVS)ismostlygreater than for reading ahead beyond ME. Our findings also reveal that the rate of methodological deviation(causedbythetwodifferentcalculationapproaches)forMEprocessingtimeis around10%,butthetwoprocessingtimeshavedemonstratednostatisticallysignificant difference, validating the processing time calculated by audio data in Zheng&Xiang (2013).We conclude this paper with some reservation on eye-tracking translation research:though powerful in providing solid and informative process data, it has some limitations in clearly probing into intricate human cognitive process.展开更多
文摘This paper aims to investigate the eye-voice span (EVS), the distance between eyeandvoice,insighttranslatingmetaphoricalexpressions(MEs).24MAtranslation students, with no professional translation or interpreting experience, were asked to conduct a sight translation (STR) task, and the processes were registered by eye-tracker and audio recorder. The qualified eye-tracking and audio data were further analysed by Tobii Studio and Audacity audio processing software. Our findings suggest that the time of the pause preceding an ME was largely, but not entirely, spent in processing the ensuing ME. However, due to the general existence of reading ahead activities in STR, the planning step for sight translating an ME takes place prior to the preceding pause; moreover, due to local processing difficulty causedbytheME,thetimeforreadingaheadintoME(temporalEVS)ismostlygreater than for reading ahead beyond ME. Our findings also reveal that the rate of methodological deviation(causedbythetwodifferentcalculationapproaches)forMEprocessingtimeis around10%,butthetwoprocessingtimeshavedemonstratednostatisticallysignificant difference, validating the processing time calculated by audio data in Zheng&Xiang (2013).We conclude this paper with some reservation on eye-tracking translation research:though powerful in providing solid and informative process data, it has some limitations in clearly probing into intricate human cognitive process.
