Language switching in aviation : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Aviation at Massey University, Manawatū, New Zealand

Abstract

Clear and precise communication between pilots and air traffic controllers is a precondition for safe operations. Communication has long been identified as a major element of the cockpit–controller interface, explaining one third of general aviation incidents (Etem & Patten, 1998). Yet, despite multilingualism with English as the lingua franca being a characteristic of aviation communication, little research appears to have investigated the efficiency of operation of bilinguals alternating between their dominant, usually native, language and English in a bilingual air traffic environment. The studies undertaken for this research sought to rectify this situation by examining the cognitive aspects of situation awareness during language switching in aviation. Quantitatively and qualitatively analysed responses to an online-distributed survey aimed at investigating the current bilingual situation in aviation revealed that while situation awareness for the majority (76%) of native-English speakers was adversely affected by bilingualism, almost 30% of bilinguals also reported their situation awareness being affected. Subsequent experimental analyses using a language switching paradigm investigated how participants recognize a target call sign, identify an error and predict in bilingual compared with monolingual English conditions. The effect of the language condition participants’ native Chinese only, English only, or a mix of both, varied across the three tasks. Call sign recognition performance was found to be faster in the English condition than in the bilingual condition, but accuracy did not differ, a finding that was attributed to the effect of call sign similarity. However, when the task was more complicated, the difference between the conditions diminished. No effect on performance was found for simultaneously listening to two speech sources, which is potentially analogous to cockpit communication and radio calls. The error analyses served to test for response bias by calculating sensitivity, d', and decision criterion C in accordance with Stanislaw and Todorov’s (1999) Signal Detection Theory calculations. Several cognitive implications for practice were proposed, for example, in Crew Resource Management (CRM) training and personal airmanship development, exploration of own behavioural biases might be used to adjust the placement of the criterion. The cognitive implications largely focused on affecting attitudes to increase awareness. Attention was focused on performance of bilinguals to identify which language condition facilitated faster and more accurate responses. The findings were unable to support any of the conditions, leaving the question: Would a universal language for communication on radio frequencies be worth considering, to allow everyone to understand what is said? Disentangling the effects of language switching on the performance of bilingual pilots and air traffic controllers remains a task for future studies.