University of Central Florida Undergraduate Research Journal - Cross-Modal Distraction on Simultaneous Translation: Language Interference in Spanish-English Bilinguals
US tab

Discussion

As the trials went on, all participants had higher accuracies and faster reaction times, as they had more practice. There were no significant differences in performance between male and female participants, nor any differences based on caffeine consumption, and having three left-handed participants did not impact the data. None of the music tendency variables returned significant results. Whether the participants listened to music while studying made no difference in their performance at this task. This result suggests there is no advantage to students who listen to music while studying, even at a high volume.

The purpose of the monolingual group was to act as a control group to see the effects of adding another language. This choice also allowed us to investigate if a bilingual advantage existed. Small sample size of participants can be a contributing confounding variable to the results. Monolinguals did not display interference by any of the variables tested except for screen word length. Having an auditory distractor word in any of the headphone ears did not seem to affect output performance. However, the length of the word on the screen did influence on their accuracy scores: the longer the word, the more errors the monolinguals made. A hypothesis for this result is that the shorter words did not span past the fixation cross, while the longer ones spanned significantly past the fixation cross where their eyes were focused. Therefore, the monolinguals’ line of vision cut off the whole word, creating an override effect for the auditory stimulus and allowing a switch of attention to the auditory modality. Monolinguals performed better than bilinguals at this task in both accuracy and reaction times. Their high performance could be a product of the simplicity of the task, as the words shown were very common words where the likely age of acquisition was low. These findings suggest there is no bilingual advantage in this cross-modal setup.

The bilingual group consisted of native speakers and Spanish learners, but first language did not have an influence on their performance outputs. Proficiency level was not a significant factor for performance either. Because of the high frequency of most of the words, novice and proficient bilinguals performed at equal levels of Spanish proficiency. Bilingual subjects performed worse when the word on the screen was different than the word spoken in the headphones. Accuracy was higher when the screen word was the same as the audio word, suggesting that the distractor word had an impact on subject performance. Repetition of the same word causes a faciliatory effect. It is more distracting to have counteracting information spoken while you are trying to read, comprehend, and repeat or translate the target word. When both words are presented simultaneously, a competition of word processing emerges. Participants performed more accurately when the auditory word was presented to both ears rather than the left or right individually. Having both ears stimulated is normal when having a conversation, so when it becomes fully panned left or right, the selective attention switches and the probability for interference increases. The biggest difference between panning conditions was right ear compared to both ears, with more errors in the right; this result goes against the findings in many other studies of a right ear advantage (Desjardins & Fernandez, 2018; Mägiste, 1984; Soveri et al., 2011). Accuracy was lowest when the word on the screen was in Spanish, and reaction times were slower when Spanish was present on the screen. For the heritage speakers, interlingual interference may have caused the decline in performance, while for the Spanish learners, this result is most likely due to the low proficiency in the language. Accuracy was lower and reaction times were slower when the word on screen and in the headphones was in Spanish compared to both words presented in English. Intralingual interference was higher in these two conditions, but Spanish elicited more errors since the task was to translate into English, and the dual English condition was faciliatory. The Spanish on screen and in the headphones allowed more interference to happen within that dictionary, causing competition of word processing. The condition with the highest accuracy was an English screen word with a Spanish audio word, compared to the dual Spanish condition (lowest accuracy). Having the target word in English eliminated the need for a translation, which means the task only required word repetition, like that of the monolingual group. With a simpler task, participants were able to better ignore the distractor stimuli. This condition shows the possibility that a bilingual has the capacity to not activate the dictionary of the opposing language, a feature described by Soveri et al. (2011). Contrary to the monolingual participants, screen word and audio word caused significant effects on subjects’ performance. The trials were randomized, forcing participants to switch between their two languages in almost every trial. The switching of languages and activation of dictionaries creates a higher cognitive load and may contribute to bilinguals’ lower performance.

Conclusion >>