Bilinguals in the presence of other known bilinguals typically engage in codeswitching, generally described as the fluid alternation between languages in speech. Despite popular misperceptions, codeswitching is a highly intricate linguistic skill that is a hallmark of high levels of proficiency across a bilingual's two languages. Yet little is known in how bilinguals successfully integrate lexical, grammatical, and prosodic information in codeswitched contexts. In parallel, a recent hypothesis has implicated cognitive control (i.e. the general ability to negotiate between parallel representations) in the mediation of conflicting representations in both linguistic and non-linguistic domains. The current proposal works within this framework by viewing bilingualism as a case of parallel representations across a bilingual's languages. Codeswitching, therefore, presents a specialized linguistic situation in which parallel representations are brought into direct competition, particularly in comprehension. As such, the investigation of the comprehension of codeswitched speech presents an ideal case study for further testing whether conflicting representations incur heightened use of cognitive control.
To investigate this issue, the current project examines the comprehension of codeswitched speech in two studies. The first study examines general costs to comprehension in codeswitching. The second study takes a more fine-grained approach to investigating what kinds of linguistic cues might influence the comprehension of codeswitched speech. By simultaneously using both functional magnetic resonance imaging (fMRI) and eye-tracking methodologies, the proposed studies will provide converging evidence on the timecourse of comprehension (eye-tracking) and the role of cognitive control in specialized brain regions implicated for the successful integration of codeswitched speech (fMRI).
Intellectual Merits. These studies make significant contributions to research in how humans comprehend multiple languages both in unilingual, i.e. one language, and dual language, i.e. codeswitched, contexts. These studies constitute the first systematic investigation of the comprehension of codeswitched speech simultaneously using fMRI and eye-tracking. By investigating the role of cognitive control in language comprehension, the proposed studies contribute to our understanding of the claimed cognitive advantage in bilinguals. Moreover, the anticipated results help uncover the types of linguistic cues that facilitate or delay language comprehension, further informing current psycholinguistic models of sentence processing. Because codeswitching is hypothesized to involve the successful integration of competing representations across languages, these studies directly address how humans negotiate multiple representations. Pursuing this line of research has great implications for understanding the cognitive mechanisms implicated in language contact and language change.
Broader Impacts. As the U.S. increasingly becomes multilingual, studies of the linguistic skills available to bilinguals offer a more representative account of U.S. society and expand current understanding of language comprehension processes. The project provides training opportunities to the Fellow and to undergraduate bilingual researchers who are members of underrepresented groups, thus increasing diversity in STEM fields. In light of the fact that codeswitching continues to face negative stigmas, the proposed studies directly counter such negative stereotyping by providing empirical support for the view that codeswitching is a highly intricate linguistic skill requiring considerable cognitive flexibility.
Broadening Participation. By including bilinguals who codeswitch, these studies contribute crucial empirical evidence while broadening representation of underrepresented groups in scientific inquiry. The Fellow is committed to disseminating the results through public lectures to target audiences in order to better communicate research that directly impacts multilingual families. These lectures are intended to increase awareness of the importance of maintaining a multilingual society, mainly through the preservation of minority languages spoken at home. Finally, through a series of interactive activities that allow high school students to explore the types of experiments that language researchers utilize, the Fellow plans to heighten and spark the interest of high school students from underrepresented groups to pursue higher education in STEM-related fields.
Research on bilingual comprehension is becoming more important because most people around the world speak more than one language. The primary goal of the fellowship tenure was to investigate the cognitive and neural processes that support comprehension in bilingual speech, which often includes the frequent alternation between languages, known as code-switching: pero no encontraron el flag out there? "but didn't they find the flag out there?" During comprehension, switches to the other lanugage should be more difficult, yet bilinguals do not appear to suffer in comprehension. Nevertheless, we know little about the mental processes that support comprehension of code-switching. The Principal Investigator (PI) trained on the use of functional magentic resonance imagine (fMRI, brain imaging) to investigate comprehension. The PI compares one form of language unexpectancy common in language, referred to here as semantic unexpectancy, with that of code-switching. Examine the following sentence frame: 1) Finally, we are able to cross the river becuase the engineers completed the ________ When asked, overwhelmingly people complete the sentence with bridge, indicating a high expectation for that specific completion. In contrast, tunnel is a plausible way to complete the sentence, but it is less expected. Previous studies show that greater neural activity is found in the left prefrontal cortex of the brain when encountering semantic unexpectancy. This area has also been implicated in activities involving cognitive control--such as ignoring irrelevant information or switching between tasks. The innovative question then is how code-switching compares to semantic unexpectancy. One common analytic tool with fMRI is to compare the neural activity recording during the experiment with some other task that is well understood. Thus, we compared performance on a classic psychology task called the Stroop task with the results of the experiment. Here, participants must respond to the color that a word is presented in (press the blue button if you see a word written in blue). This task is fairly simple, but the critical manipulation is that occassionally participants will see a word that means a color presented in a different color, for example, the word blue presented in green. The correct response is to press the green button; however, people are driven to respond blue due to the meaning of the word. This conflict between meaning and goal of task is an illustration of cognitive control (referenced above) and also heightens neural activity in the left prefrontal cortex. Using this method of comparison, we found a rather striking effect where encountering code-switches elicited greater overall activity in the left prefrontal cortex, regardless of the semantic unexpectancy manipulation. We illustrate this effect, included as an other image, where we show that the code-switched sentence endings result in higher neural activity than the non-switched endings. Interestingly, when we compare the results to general language comprehension, we now see both the effect of code-switching and semantic unexpectancy as illustrated in the primary image. Here, we see a gradual increase in neural activity, from non-switched expected completions to code-switched and unexpected completions. These findings suggest that code-switching engages cognitive control, a mental process shared with more general cognition, but that there is also a language component which engages the actual meaning of the sentences being comprehended. These findings are interesting becuase they suggest that code-switching is actually a finely controlled skill thus contributing to the intellectual merits of the tenure fellowship. In terms of broader impacts, the experiments target an underrepresented group in STEM fields. In so doing, the PI broadens the inclusion of groups of language speakers not typically included in experimental language research. During debriefing, many of the participants expressed surprise, curiousity, and pride in being told that the research fundamentally investigates their way of speaking. In addition, the PI was able to work very closely with an undergraduate Hispanic research assistant and aided the research assistant in developing fundamental research skills in experimental design and analysis. The PI also gave public lectures on the importance of maintaining bilingualism to groups of underprivileged immigrants, many of whom continue to believe that their children should grow up as monolingual speakers and to undergraduate classes. During the fellowship tenure, the PI presented at several conferences and continues to do so. The PI additionally had one publication come out, a manuscript that has been revised and resubmitted, and two manuscripts that are in preparation to be submitted in 2015--all on work carried out during the fellowship tenure. Finally, the intial work carried out during the tenure fellowhsip led to stronger collaborations between the PI, the Sponsoring Scientist, and supporting scientists, leading to the successful submission and funding of an NIH R21 grant, which will further explore the production and comprehension processes underlying code-switching using a conversation task with fMRI. After the tenure fellowship, the PI began a tenure-track faculty position at an R-1 public university.
