An empirical investigation of cognitive effort required to post-edit machine translated metaphors compared to the translation of metaphors

Abstract

This paper reports on a study investigating cognitive effort required to post-edit machine translated metaphors in comparison to translating metaphors by means of eye tracking and key logging. We hypothesize that post-editing will be less effortful than manual translation. In order to test this hypothesis, an experiment was carried out with two different groups of subjects. Fourteen participants were asked to post-edit a newspaper text and eight participants were asked to translate the same source text. The analysis focuses on eye-tracking data related to total fixation duration and key-logging data (insertions, deletions, pauses). Data analysis shows that cognitive effort required to post-edit MT output is lower in comparison to manual translation.