Neural Differences in Early Bilinguals and Late Bilinguals

The University of Texas at Austin

Language is one of the most intriguing methods of the human communication because we use it in our everyday life. This system of communication is unique to humans in that we are able to create an infinite number of utterances from finite words. We are the only animals that have the capacity to communicate with proper grammar and form structural sentences. An interesting topic of research is how the human brain processes this great mechanism. Researchers have investigated the phenomenon of human language processing for many years. As a result of increase of the amount of bilinguals among citizens of USA, it became very important for the neuropsychologists to face the possible challenges in the new field. The researchers have begun to explore bilingual language processing, because it can have a positive effect learning a second language in an early age, while affecting the plasticity of the brain structures and impact on the thickness of the grey matter’s level in the inferior parietal cortex. Most importantly, recent advances in neuroimaging had allowed researchers to explore how the underlying neural mechanisms of the brain processes language. One technique commonly used is functional magnetic resonance imaging (fMRI), which measures the brain’s activity by detecting changes associated with blood flow. This tells us which parts of the brain are activated during language processing. Especially, using fMRI techniques would tell us, how bilinguals are able to process and switch between two languages, and would provide researchers with needed material to investigate the areas of the brain associated with language switching.Language switching occurs when a speaker alternates between two or more languages in the context of a single conversation. One prominent debate regarding the underlying neural mechanisms associated with language switching is whether they are domain general or domain specific.

Many researchers have proposed that the mechanisms involved in language switching are domain specific. For example, a study by Hernandez, A. et al. (2001), proposes that language switching is domain general after finding that the Dorsolateral Prefrontal Cortex, a region associated with executive function, was also activated during language switching. However, many support a domain specific theory and propose that there may be a mechanism specific to language control. Abutalebi,J. et al. (2008), has compared areas of activation during intralanguage switching and interlanguage switching. The Anterior Cingulate Cortex and Dorsolateral Prefrontal Cortex were activated during intralanguage switching between a verb and a noun, within the same language. However, the left caudate area was only activated during interlanguage switching between English and Spanish bilinguals. Additionally, it should be mentioned, that there are no researches about bilingual differences with other thanIndo-European language speakers (for example bilingual speakers with English and Japanese languages).

Crinion, J. (2006) used fMRI to demonstrate that the neuronal responses in the head of the left caudate are sensitive to both the semantic content and the language of written words. The most notable case is a study of a tri-lingual patient with a lesion to the white matter surrounding the head of the left caudate. This patient had preserved comprehension in all three of her languages. Her picture naming was also above 80% accuracy. However, during language production tasks, she spontaneously and involuntarily switched from one language to another. These and other findings with bilingual patients suggest that the left caudate is required to monitor and control lexical and language alternatives in production tasks. Research has shown that executive control activates a different area in the brain compared to language switching. Additionally, behavioral studies have shown different patterns of switch costs in the linguistic and the non-linguistic switching tasks suggested that the bilingual language control system is not completely subsidiary to the domain general executive system. However, it may depend on when they learn a second language because we all have an age in life when learning a second language does not come as automatic, this period is called sensitive period. According to Montessori M., the sensitive period for language learning lasts until the sixth year. However, it is obvious, that until fifteen the language understanding and semantic definitions are still forming, and, as a result – the second language can be learned easier than in the later ages. Bilinguals who started learning a second language before the sensitive period use a specific language control mechanism, whereas people who acquire a second language after the sensitive period use a more domain general mechanism for language switching. Beck G. (2004) suggests that both behavioral and neural have a sensitive period for language acquisition. Case studies of individual feral or abused children, isolated from exposure to their first language until after puberty (critical period) have shown extreme deficits in phonology, morphology, and syntax (Beck, G., 2004). In studies of populations of normal individuals, one can systematically examine proficiency in relation to age of linguistic exposure without concern about the physical status of the learning brain. This study shows a strong relationship between age of exposure to a language and the ultimate proficiency achieved in that language. Learning during the first few months or years if exposure may show an advantage for adult learners, particularly in the acquisition of vocabulary and the speed of using certain complex sentences forms; however, long-term outcome clearly favors those who start learning that language during childhood. Those whose exposure to that language begins in infancy or early childhood display peak proficiency in the language, in control over the sound system as well as the grammatical structure. Such early learners show not only flawless control over the accent and rhythm of the language but also full productive control over the syntax and morphology. With increasing ages of exposure there is a decline in average proficiency, beginning as early as ages 4 to 6 and continuing until proficiency plateaus for adult learners (Beck 2004).

According to all mentioned above, it would be crucial to examine if there are any differences in the neural language control mechanisms between bilinguals who acquired a second language early in childhood compared to those who acquired one later in life. Past research suggests that there may be unique neural mechanisms that control language switching compared to non-language switching. However, there is no comparison of these processes in early versus late bilinguals. To address this gap, 100 simultaneous bilinguals (bilinguals who acquired both languages at the same time) and 100 sequential bilinguals (bilinguals who acquired their second language after the age of fifteen) will be recruited. The study will consist of interlanguage (English and Spanish) switching and intralanguage (verb and noun in the same language) switching.

The hypothesis is that during the interlanguage switch task, participants who learned a second language since childhood will have a faster switch rate and will also show stronger activation in the left caudate area proposed to be the control mechanism for language switching. Those who learned a second language later will have slower responses during the interlanguage switch task and will involve domain general regions including DPFC and ACC associated with overall higher executive function. In the intralanguage switch tasks, it is expected that both groups of bilinguals will show stronger activation in the DPFC and ACC instead of in the left caudate. In conclusion, the present study will fill an important gap in the literature that will provide insight on the neural differences of language control in early and late bilinguals.

Study Design

The Experimental Hypothesis

The exact age when the individuals have learned the second language (which is the independent variable of the experiment) influences the effectiveness of the language switching process and localization of a language switching mechanism in the cerebral cortex (being the dependent variable).

According to McLeod (2008), the dependent and independent variable can be determined in this way:

Independent variable. The factor of age when the second language has been learned will be used as an independent variable, in order to define how it will affect the switching speed of the language and the location of the switching center in the cerebral cortex.

Dependent variable. The switching speed of the language and the location of the switching center in the cerebral cortex will be used as a dependent variable since we cannot manipulate with this variable.

The Sample

In accordance with the requirements of the experimental method in psychology, the sample should be representative. Thus, the group of interest would be recruited from monetary or ideologically motivated people by means of conducting a search in the Institute with the help of information boards or through the Internet and newspapers, and would be randomized on one criterion, namely, the age when a second language was studied. The first group of 100 people is experimental. It will consist of only those, who started to learn the second language at an early age (under 6 years). The second experimental group, also consisting of 100 people, will comprise of those, who started to learn the second language after the sensitive period, namely, since 15 years and up. The third group, numbering 100 people, will be the control group composed of the people who did not learn the second language at all. The assignment of the control group will be to compare whether there will be any statistically significant difference between the results of all three groups. In order to obtain the valid data, the age of the sample should be in the range of 18-27, with mean age – 23.2. Other ages have their specific psychological features.

External Variables

There might be problems during the experiment that can be characterized as external variables. Slang and profanity, as well as a mixture of languages, can be considered as a separate language which can influence the experiment. Somatic and psychosomatic disorders might cause the data interpretation difficulties along with the difference between declared and real level of the language competence. The issue of hipo- and hipermotivation can also harm the validity of the experiment. There can be other factors, all of which cannot be foreseen. The researcher will try to control and to reduce the impact of external valuables to a minimum and expect to use the preliminary interview to identify the disparities with the requirements of the sample and individual features. The whole process of the experiment will be carefully recorded and documented in order to facilitate the search of the possible errors or for re-checking the results.

Type of Experiment

Laboratory experiment is a type of experiment in the scope of which the conditions are specifically organized by the experimenter. The main task is to ensure a high level of the internal validity. The main way to control the external variables is their elimination. External validity, that is validity that determines how the results of a particular study can be extended to the entire class of similar situations / events / objects, is lower than in a field experiment.

The Stimulus Material

The randomly taken 80 pictures of different objects, which were used by Snodgrass J. G. and Vanderwart M. (1980) in their research will be the stimulus material used in the experiment. The pictures will be printed on paper and be of the same colors. Additionally, the researcher will use a list of 10-30 nouns in order to form the verb.

Validity

Internal validity is the degree of influence of the independent variable, i.e. the exact age one learns a second language, on the dependent variable, i.e. the speed of language switching and the mechanism of its localization in the cerebral cortex. This characteristic grows with the increase of a chance that the changes in the dependent variable are caused by changes in the independent variable. According to the theoretical basis, the chances of the internal validity are very high. The external validity is the type of validity that determines how the results of specific research can be extended to the entire class of similar situations / events / objects. In order to increase the external validity, further experiments with other bilinguals (English/Russian; English/ Swahili; English/ Japanese; and so on) should be conducted to represent the greater amount of the possible bilinguals.

The Procedure of the Experiment

During the research procedure, the participants will be connected to the fMRI in order to gather information about the cerebral cortex activity. In order to collect the information about the interlanguage switching the next steps would be done. First of all, the participants will be presented with stimulus material in a form of the serially disposed pictures. The respondent has to name the object / creature / phenomenon, drawn on the picture, first using English and then Spanish languages. Further, the participants will be asked to speak a little about the image using languages as in the previous task. During these two tasks, the experimenter will monitor the accuracy and speed of answers, their accuracy and the switching speed of the language. The next step will be in completing the story based on any 10 of the 80 proposed pictures, using different languages, as it was previously. Thus, we will be able not only to measure the speed of language switching but also locate the center of switching. Finally, the respondents will have a task to write 3 to 4 sentences about the provided painting using each of the languages as it was done previously. As a result, it will be possible to determine efficiency of the language switching and then to define whether the quality of information would suffer from switching between languages.

To gather information about intralanguage switching the next steps will be undertaken. The participants of all three groups will be given a task to form verbs from the noun, which were printed on the paper. The respondents will have to name the noun, explain its meaning, and form the verb from of the noun. In this way, it would be possible to trace the speed of intralanguage switching.

The procedure of the experiment will be changed in order to test the control group. The members of the control group will have to use only one language. The results of the observation of the control group will be represented as the statistic level of an average person and give an opportunity to compare the bilingual person with people speaking a single language. The next stage of the research will be the statistical analysis of the results of the experiment and comparison between data groups. At this stage, it is important to identify the main categories of the data that will be received and the comparison inside them. The first group of data will be the speed of the interlanguage switch. Statistical analysis of the results of this group will make it possible to state the presence or absence of significant difference in the speed of switching between languages of first and second experimental groups. The second group is the intralanguage switching speed. According to the data of this category, it will be possible to compare the intralanguage speed switching of all three groups of the respondents. Results would allow to determine whether the switching speed suffers after the second language learning, and if yes, than how much and in which group. The third group is the localization of interlanguage switching control centers in the cerebral cortex. Systematization of this group data will allow confirming or denying the assumption about the age determination of the location of the language switching centers after the learning of the second language. In comparison with intralanguage switching center, these data will assist in determining whether the control centers of interlanguage and intralanguage switching are localized differently. Thus, by comparing the results of intra-group data and summarizing them, we would get a structured image of the process of switching of the brain between the languages and their correlation with age when the second language was learned.

Implications of the Findings

The researcher expects that the results of the experiment will confirm the hypothesis about the difference in the localization of the switching language control centers according to the age, when the second language was learned, the greater speed of language switching, and the plasticity of the brain of those who learned the second language in the sensitive period. The findings will provide an opportunity to reconsider the concept of the second language acquisition in order to improve the education system. Apart from that, the results of this study will help optimize the program of second language acquisition by people who are out of the sensitive period. In addition, the results of this experiment will allow a better understanding of neural processes during communication between people. Developing this topic further may allow studying the cultural dimension of the brain parts responsible for language and speech brain structures differentiation. If there would be no statistical difference in the speed of switching between languages, and the localization of the switching control centers will be different, it will be possible to assume that the age difference of the second language acquisition forms the interchangeable control centers of the language switching. Thus, as a result of the experiment, it would be getting the knowledge about compensatory centers of the languages switching control. Accordingly, these results will be more demanded by neurology and neuropsychology in order to provide the information about the treatment and understanding of the functions of certain areas of the cerebral cortex.

In case the research will provide the data that, depending on the age when the second language was learned, about the reduction of the intralanguage switching speed comparing with other groups, it will be possible to make the following two assumptions. If the reduction in the switching speed of the language will be observed among those who learned second language in the sensitive period, it can be assumed that learning a second language after 15 years makes it impossible to fully acquire both languages. For assumptions that are more specific, it would be necessary to conduct additional research. Moreover, it is important to crosscheck the results because, in spite of the rule of the falsifiability in any science, there are experiments disproving these assumptions. If the reduction is observed among those who learned a second language after the sensitive period, it can be assumed that learning a second language somehow replaces the previous one. To confirm or disapprove this hypothesis, further studies would be necessary to conduct.

The results of this experiment will allow to speak about the connection and the amount of memory available for learning languages, and the need for an early study of as much as possible number of them. If there is no statistically significant difference in the speed of switching between languages, but there is a difference in the location of the language switching control center, it will be possible to suggest the need to review the functions of the areas of the cerebral cortex that were supposed to be responsible for language switching. Further experiments will define this issue.

There are also possible difficulties related to the interpretation of the results of the experiment that are influenced by individual characteristics of the functioning of a linguistic system. For example, a person who has studied the second language by the book reading may have differences in the location of the center of the language switch control as opposed to the person who acquired the language mostly through the audio recording or talking with the native speakers. In addition, there may be problems associated with the quality of language and its constituent parts, such as phonetics, vocabulary, grammar, and semantics. These problems may affect the results of the experiment significantly, and therefore, they require the experimenters’ control.

It is expected to use interviews and experts assessment in order to avoid endogenous factors of competence in the language. However, it is barely possible to control or determine the social factors and limit their influence. Moreover, it is possible to detect unforeseen correlation by dividing the groups into subgroups during processing of the results. Such unforeseen correlation should be investigated further in order to determine their real significance. On the other hand, the results would be defective if there would be no comparison with bilingual people who learned any other language as a second language except the one from Indo-European group. The importance of this factor is associated with ethnical linguistic features of other nations and races. The Asian group of languages is characterized by ambiguity and the lack of single letters. Thus, the format of the language resembles the language of images. The comparison with the groups of people who know more than two languages, as well as a comparison within this group based on already described factor (whichever during sensitive periods they learned their languages or after) would be also very important for understanding the value of the gathered results and their exogenous validity.

In conclusion, it should be mentioned that the experiment has a high chance of success due to the high level of internal validity. The hypothesis about the age when the individual has learned the second language has high chances to influence the effectiveness of the language switching process. Additionally, the assumptions about localization of a language switching mechanism in the cerebral cortex are theoretically substantiated and can be properly studied within the proposed experiment. These results have significant scientific novelty and could be used for further research.