LING 620 Research Methods Final Proposal

Makiko Ogawa

Working title: The preference of shape and color in Japanese bilinguals

  1. INTRODUCTION: RESEARCH AREA

While a theory of Universal Grammar seeks to the innateness in language learning in monolinguals, the fact of research in support of linguistic relativity hypothesisremains and an emerging area of research in linguistics has been re-evaluating the hypothesis and left behind valuable insight in the field of SLA and Bilingualism. Although a linguistic relativity hypothesis has been a controversial issue in the past few decades, cognitive psychologists and anthropologists have become curious about the relationship between languages and thought in light of recent progress in methodological investigation, specifically how language learning may influence human cognition.

The present research will replicate a study conducted by Athanasopoulos & Kasai (2008), but will be addingthe variable of response time. The main focus of the present research is investigation into the relationship between humans’ perception and language learning by implementing non-linguistic tasks to native speakers of English (monolinguals) and native speakers of Japanese (bilinguals). Results will be compared and correlations between linguistic and non-linguistic will be sough if significant differencescan be foundbetween these two groups.

2.AIM/JUSTIFICATION

The present study is a replication of the original study conducted by Athanasopoulos & Kasai (2008), whichwill add the component of response time. Cook, Bassetti, Kasai, Sasaki, and Takahashi (2006)address the need of research focusing on response time. The study is important for threereasons. First little is known about the relationship between linguistic and non-linguistic process among speakers of different languages. Secondly, there are few studiesfocusing on response time in the research of human’s perception and language learning. Thirdly, little is known about the process of monolinguals and bilinguals in terms of non-linguisticperformance. Therefore, the present study will investigate an association between perception and language learning by measuring response time of speakers of English (monolinguals)and speakers of Japanese (bilinguals).

  1. LITERATURE REVIEW

The linguistic relativity hypothesis is widely known through the work of Benjamin Lee Whorf (1956), which is sometimes called the Sapir-Whorf Hypothesis, the language we speak determines the way we think. Lucy (1992), for example, claims that a grammatical marking in languages has a significant impact on human cognition. Lucy hypothesizes that the differences in the number marking systems in Yactec and English would cause monolingual speakers of Yactec and English to pay attention differently to either the material or the shape of an object when implementing a triads matching task. In Lucy’s experiment, participants were first shown a standard object, such as a wooden spoon. Then, the participants were asked to select from the two choices, one that had a different shape but made from the same materials such as a wooden spatula, and one that was in the same shape but made from different materials, such as a plastic spoon. Results showed that the English monolinguals selected shapes alternatives significantly more than Yactec monolinguals and found different preference of choices between these two groups of monolinguals. Lucy argues that English speakers tend to focus on the shape of an object because of the obligatory plural marking system in English whenever a count noun is used. Therefore, attention is given to the shape of an object and become habitual awareness.

Mazuka (2003), however, argues that these different preferences of Yactec monolinguals and English monolinguals are not due to the grammatical markings of the two languages but non-linguistic factors such as educational and cultural backgrounds of these two groups. Given the fact that English monolinguals in Lucy’s experiment were typical American college students between 19 and 27 years old, and Yactec monolinguals were from a village with fairly little education ranging in age from 18 to 45 years. Mazuka claimed that these non-linguistic factors could account for these two groups of monolinguals showing different preferences.

Although Mazuka pointed out the problems in Lucy’s experiment, the author replicated Lucy’s study with English monolinguals in the U.S and Japanese monolinguals in Japan and Japanese bilinguals in the U.S to evaluate the hypothesis of Lucy’s grammatical marking systems. Japanese and Yactec share similar kinds of numerical classifiers, therefore, the differences in the non-linguistic backgrounds of the two groups can be minimized if controlled. In the replicated study, it was assumed that if the grammatical marking system is crucial, Japanese monolinguals would behave similarly to Yuctec monolinguals. However, the results of the study stated that not only the American college students but also the Japanese college students and bilinguals living in the U.S showed preference for shape alternatives in triad matching task; therefore, disproving Lucy’s hypothesis.

Mazuka’s results do not suggest that different number marking systems have no cognitive consequence for the speakers as discussed in the Mazuka’s paper (p.371). There are a series of studies that have examined children’s acquisition of substance and objects. In a study of Imai and Gardner (2007), they used three different types of stimuli, substances such as sand in an S-shape, simple objects such as kidney-shaped pieces of paraffin, and complex objects such as wood whisks. American children and Japanese children were first shown one of three stimuli, and then were asked to select from the two alternatives; a shape alternative or a material alternative. The authors found that American and Japanese children could differentiate between substances and objects when they were as young as 2 years old. However, American and Japanese children reacted differently to simple objects and substance selection. American children were more likely to select the shape alternative compared to Japanese children as young as 2 years old. This implies that “linguistic structure affects the weighting of dimensions and the way in which speakers classify entities into different categories”(p.196).

Now a question arises, what is happening in the mind of adult’sbilinguals? Cook et al. (2006) replicated the study of Imai and Gardner (1997) to see how adult bilinguals perform in the triad matching task. In the study, Cook et al. (2006) hypothesized that experience of English would increase the selection of shape alternative in simple objects and the preference of bilinguals in triads matching task differs from both Japanese monolinguals and English monolinguals. Japanese bilinguals were divided into two groups according to the length of stay in an English speaking country. The study showed that longer stay groups tend to select the shape alternative more than the short-stay group, overall. Participants took the vocabulary test(Nation,1990) to measure their English proficiency; however, there is no correlation between the scores of the tests and the preferences of shape and materials alternatives.

Athanasopoulos and Kasai (2008) conducted a similar kind of study conducting triads matching task that are non-linguistic, shape versus color, instead of shape versus material. The reason they chose the color was for the advantage of stimuli not to be lexically labeled as counters or mass nouns. As stated in the article, “color name is used as an adjective, and as such, it is inherently neutral with regard to any semantic or grammatical content denoting count or mass status” (p.109) Participants were monolingual Japanese recruited in Japan, Japanese bilinguals in Japanese and in English recruited both in Japan and in the U.K, and English monolinguals recruited in the U.K. The participants were first shown the standard object with color, and then were asked to choose either shape or color alternative. Their results show English monolinguals selected shape significantly more than Japanese monolingual; on the other hand, bilinguals shifted their cognitive preference according to their L2 proficiency in the triad matching task. Also, the studies show that there was no significant difference between the task tested in L1 or L2 instruction.

Athanasopoulos and Kasai (2008) address the crucial issues whether different linguistic structures influence non-linguistic similarity judgment or not. It is shown that there is a correlation between the level of proficiency and the preference of color versus shape matching task. One crucial issue of this study is unresolved, namely how bilinguals perform the task compared to monolinguals in terms of response time.Randsell and Fischler (1987) found that even accuracy was the same on list recognition and lexical decision tasks, adults bilinguals were slower in terms of response time. In the Athanasopoulos and Kasai (2008) study, the participants could take as much time as they wanted; therefore, the response time has not been sought yet. The current research is meant to add the new variable, response time into the original study in order to investigate the process and relationship between linguistic and non-linguistic tasks.

  1. RESEARCH QUESTIONS/HYPOTHESES

ResearchQuestions

1. Do English monolinguals selectshape significantly more than color materials compared to Japanese bilinguals in the triads matching task?

2. Do English monolinguals respond significantly quicker to triads matching taskcompared to Japanese bilinguals?

3. Does language learning influence the speaker’s perception?

Hypothesis

I expect to obtain the similar data tothe previous study. In addition to that, I expect that Japanese bilinguals would take longer time compared to English monolinguals in terms of the time of response in triads matching tasks.

METHODOLOGY

A. SUBJECTS/SOURCES

45 students ranging in age 18-30 years will participate. 15 students will be English native speakers enrolled in Ohio University, 15 students will be Japanese native speakers who are bilinguals in Japanese and English enrolled in Ohio University, and 15 students will be Japanese native speakers who will come for the study abroad program from Chubu University, Japan in March 2010. They will be all adults. The tests will be all instructed in English, which will be L1 for English native speakers, and L2 for Japanese native speakers.

B. MATERIALS/INSTRUMENTS

Questionnaires, “shape” versus “color” triads matching tasks, a cloze test, and picture description tasks will be given to participants in that order.

Questionnaires will ask participants their gender, age, their experiences of living in different countries, the length of stay in English speaking countries, languages they can speak other than their native languages, and the acquisition of L2. (See the attached Appendix A for details.)

As for the “shape” versus “color” triads matching tasks, thirty color illustrations of novel objects will be used as stimuli. The stimuli will be created with free web software called Sumo paint. They will be all created by the present researcher and the stimuli will be organized into 10 different triads as examples shown in the attached Appendix B. Each triad will be composed of the standard object and two alternatives, which will have the same shape as the standard but different color. As the previous study of Athanasopoulos & Kasai (2008) showed, precautions for creating the stimuli will be greatly taken into consideration. First, all the stimuli will be designed to the same scale. Second, the color used in the stimuli will be carefully selected and combined with rather contrasting colors such as red and green. Thirdly, the shape will be an arbitrary novel shape, not to be labeled as square, triangle or circle, to test their non-linguistic influence. Then the stimuli will be transfer into the software called Psyscope and record the participants’ response time.

A cloze task which was created and used by Oshita (1997) will be given to all participants to measure their English language proficiency holistically.Errors in part of speech, tense, styles and others are unacceptable as correct answers. See the attached Appendix C for details.

A picture description task will ask participants to describe orally a picture or a scene. An example of the picture is attached asAppendix D.

C. PROCEDURE

Participants will be tested individuallyin the computer lab at Ohio University.An interactive computer program, which enables us to measure the response time as well as save all responses, will be designed for the tests. The software that will be used to create the program is Psyscope. The program will be ready to run by the time the participants come.

There will be four steps for this study.First, participants will be given questionnaires addressing their language learning experiences.The questionnaire sheets will be handed out to the participants and the researcher will orally read the questions in front of the participants to make sure they can fill answer them. The researcher will answer questions from the participants as necessary.

Secondly, participants will be asked to complete a “shape” versus “color” triads matching tasks that will be presented on a computer screen.The researcher will briefly introduce the task that they need touse a computer mouse to click on the computer screen. After reading the instructions on the computer screen, the participants will be asked to click on the “next” button to begin the triad task. There will be in total 10 trials two stages in each trial. In the first stage, the standard object will appear on the screen and the participants will be instructed to click on it, which will guide them to the second stage. In the second stage, the two alternatives from the standard will appear side by side below the standard object on the screen and the participants will be instructed to click one alternative that they think is the same as the standard object. The instructions that will appear on the screen will be the same ones as in the previous study of Athanasopoulos & Kasai (2008); “Show me which is the same as this, please click.” The response time between the first click on the first stage and the second click on the second stage will be measured. Their response time will be recorded and saved as a data file in addition to their response choices.

Thirdly, participants will perform a cloze task (Oshita, 1997) to test their English language proficiency.A cloze task requires them to read a short paragraph (about 8-10 sentences long) and fill in missing words based upon context clues surrounding the gap.The researcher will pass out the written sheetsand collect them.

Lastly, participants will be asked to describe a picture orally, which will be recorded and stored in the computer for analysis of spontaneous oral production. For this test, the participants willuse headphones to record their oral depiction.Then the content of the oral depiction task will be transcribed by the researcher.It will take participants about 20-30 minutes to complete this entire experiment.

D. TYPE OF DATA (8 pts)

Responses will be scored as the number of times each participant chooses a shape or color. In addition to that, individual preference will be scored as shape or color when the participants make a choice of shape or color at least 7 out of 10 trials. When the participant makes a choice of four, five or six shape or color, they will be scored as no preference.

The mean response time of the triad task will be calculated and compared with the groups of English native speakers and the Japanese native speakers.

For scoring correct answers of the cloze test, acceptable-word method will be used as Oshita(1997)did. Mean score, standard deviation, minimum score and maximum score will be calculated.

For the picture description tasks, the mean percentage ratio of producing correct plural morphology, and mean raw scores of producing correct plural morphology/number will be calculated.

6. ANALYSIS

The study will use statistical analysis of PSPP. The correlation between shape preference and variables such as gender, the age of L2 acquisition, the length of stay in an English speaking country, the mean scores of a cloze test, plural marking scores in the picturedescription task and others will be analyzed.

7. ANTICIPATED PROBLEMS/LIMITATIONS OF THE STUDY

First of all, we may have technical problems in recording. The oral depiction tasks may not be very easy to transcribe in every detail. In the experiment, the participants may not take the tests seriously, or have technical trouble with using the mouse, which may affect their response time in triads matching task.

Secondly, the study is limited in terms of participants. They will reside in Athens and be enrolled in only Ohio University; moreover, the number of participants is small compared tothe previous study conducted by Athanasopoulos & Kasai (2008).

8. WHAT DO YOU EXPECT TO FIND?/ OR RESEARCH FINDINGS SO FAR

As this study is the replication study of Athanasopoulos & Kasai (2008), which adds the component of response time, I expect to obtain the similar data to support the finding of the previous study. In addition to the finding of the previous study, I expect to find that Japanese native speakers who are bilinguals in English and Japanese may perform differently in terms of the time of response. If the hypothesis of Lucy (1992), who claims that obligatory plural markings in an English direct speaker’s habitual attention toward shape, then the question arises of how bilinguals will perform; will they choose the alternatives in the triads tasks as quickly as English native speakers? The aim of this study is to find out if there is any significant difference between English native speakers and Japanese bilingual speakers in terms of response time in triads matching tasks.