Attention and Simultaneous Translation

Abstract

In a simultaneous translation task subjects were required to translate a message, which they received in one ear, into another language. In the other ear these subjects received an irrelevant message in either of the two languages used in the experiment. It was thought that the irrelevant message would exert some influence over the translating procedure if the same dictionary units and possibly the same thought units had been involved. The results seemed to support the hypothesis in so far as the word omission rate of the translated product appeared dependent on the language of the irrelevant message. No particular conclusion could be drawn regarding the different types of prose utilised in the irrelevant message. It seemed that : (1) subjects do worse with the inputs in the same language than in a different language ; (2) when the input languages are the same, prose from different novels results in more omissions than prose from the same novel ; (3) when the input languages are different they seem to do better with prose from a different novel.