The Effects Of Thinking Aloud On Cognitive Load In Translation:An Empirical Study

Think-aloud protocols(TAP)has been a popular data collection method in translation process research since the 1980s.Yet,regarding its validity and appropriateness there have been controversies.For example,Toury(1991)believes that thinking aloud may have influence on translation process.Studies such as Krings(2001)and Jakobsen(2003)find that thinking aloud does affect translation process in segmentation.To date,there has been a scarcity of strong evidence to support that thinking aloud has critical influence on translation process.Now that new research methods are available,methodological concerns related to think-aloud protocols prompt for further research.This study investigates the effects of think aloud protocols on cognitive load in translation in terms of time on task,duration of different translation phases,cognitive load of processing the source and target texts,as well as translation difficulty.Twenty translation students were recruited,and thirteen of them took part in the formal experiment.As a within-participant study,the main experiment consisted of two sessions—translation without thinking aloud and translation with it.Source text difficulty was also taken into account to explore the effects of TAP in detail.The whole translation process of participants was recorded by a Tobii TX300 eye-tracker and Translog Ⅱ,a key-logging program.An adapted version of NASA Task Load Index was employed to assess translation difficulty.In addition,the participants’ translations were graded.All the data from eye tracking and key logging were analyzed using descriptive and inferential statistics.In ferential statistics analysis was conducted with linear mixed effects models.Data from NASA Task Load Index and verbal repots were used for triangulation.The results shows that:(1)cognitive load in translation is increased by thinking aloud;(2)the effects of thinking aloud on cognitive load vary in accordance with source text difficulty;(3)cognitive load in terms of duration in the draft phase is increased due to thinking aloud;(4)cognitive load in terms of fixation duration and counts in source text processing is increased due to thinking aloud;(5)As a complementary finding,the inferential analysis of translation performance indicates that thinking aloud had no influence on translation quality of difficult text group,but does lower translation quality of easy text group.The findings shed more light on the effects of TAP on the translation process in terms of cognitive load,which need to be taken into account when using TAP to study the process of translation.