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

Methods

Participants

Sixteen Spanish-English bilinguals, and 17 English monolinguals were studied. All subjects had normal cognitive functioning and no auditory, visual, or physical impairments. Three out of the 33 participants were left-handed, and 30 were right-handed. All participants were recruited from the University of Central Florida using the psychology recruitment website SONA. All participants received one SONA participation credit. Age for the 33 participants ranged from 18-30, with bilingual averaging 20.25 years (SD = 3.73), and monolinguals averaging 18.64 years (SD = 1.32), while 51.51% of the participants were male. Half of bilinguals listed Spanish as their first language, 25% listed English, and 25% stated they learned both languages at the same time.

Materials

All subjects were given a consent form and a general questionnaire, consisting of background information, handedness, caffeine intake, studying habits, and music listening tendencies. Bilinguals were given English and Spanish proficiency tests as well as self-reported language fluency/acquisition questions. It was stated to select the most grammatically correct answer, in order to decrease the effect of varying dialects. Words from most commonly-used Spanish and English compilation websites were presented. Spoken forms of the words were recorded in both Spanish and English. Word recordings panned either 100% to the right ear, 100% to the left ear, or 50% left and 50% right (both ears). All recordings were less than a second long, said by the same speaker, and spoken neutrally. The speaker was a 21-year-old male native-born Puerto Rican, fluent in Spanish. There were two main conditions of stimuli for both groups: match (control) and mismatch (experimental). For the control, the word on the screen was the same as the word spoken in the headphones. In the experimental condition, the word on screen was different than the word spoken. In the monolingual group, both the words on the screen and the spoken words were always in English. The bilingual group consisted of four conditions: English on screen with English in headphones, English on screen with Spanish in headphones, Spanish on screen with Spanish in headphones, and Spanish on screen with English in headphones. Word relationship to audio panning were matched based on length of the words.

Procedure

The study was available in-person only and took place at UCF in the psychology building. Participants were given a general explanation of the experiment and voluntarily signed up for a specific time slot via the SONA website. In the lab they were given a packet containing the consent form, pre-survey, proficiency tests for English and Spanish (for bilinguals only), and a summary of the experiment. The task was explained by the researcher but was more detailed on the welcome screen before the start of the experiment. Participants were given these verbal instructions: “You are going to translate or repeat what you see on the screen while ignoring what you hear in the headphones. Use this button (1st button on the serial-response box, labeled ‘1’) to continue to the next word. Read the instructions on the screen before starting”. Bilingual participants were also given the instruction to “translate everything you see into English”. Participants sat at a desk with the computer at eye level and at least 12 inches from their faces; a black trifold board was placed behind the computer to minimize distractions during the task. Subjects then put on the headphones, read the instructions and began. Monolinguals were instructed to repeat the word on the screen while ignoring the words spoken in either or both ears. Bilinguals were instructed to translate the word seen on the screen (English or Spanish) into English, while ignoring the spoken words (English or Spanish) in the headphones. To control for any ear advantage, the audio was panned randomly between left, right, and both channels. The word was presented in the middle of the screen for 1400 milliseconds, followed by a centered fixation cross. Participants pressed the first button on the serial-response box to continue onto the next word. Each participant saw 72 words. Word order was randomized and the two groups received different lists, since the monolinguals cannot translate Spanish words. The serial-response box recorded reaction times in milliseconds while the computer software recorded their translations/responses. Subjects were given a brief post-task survey asking how the task went, how they felt, and any feedback.

Results >>