Cognitive shift in the bilingual mind: Spatial concepts in Korean–English bilinguals

Bilingualism: Language and Cognitionhttp://journals.cambridge.org/BIL

Additional services for Bilingualism: Language and Cognition:

Email alerts: Click hereSubscriptions: Click hereCommercial reprints: Click hereTerms of use : Click here

Cognitive shift in the bilingual mind: Spatial concepts in Korean–Englishbilinguals

HAE IN PARK and NICOLE ZIEGLER

Bilingualism: Language and Cognition / Volume 17 / Issue 02 / April 2014, pp 410 - 430DOI: 10.1017/S1366728913000400, Published online: 08 October 2013

Link to this article: http://journals.cambridge.org/abstract_S1366728913000400

How to cite this article:HAE IN PARK and NICOLE ZIEGLER (2014). Cognitive shift in the bilingual mind: Spatial concepts in Korean–Englishbilinguals . Bilingualism: Language and Cognition, 17, pp 410-430 doi:10.1017/S1366728913000400

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/BIL, IP address: 198.24.31.70 on 21 Apr 2014

http://journals.cambridge.org Downloaded: 21 Apr 2014 IP address: 198.24.31.70

Bilingualism: Language and Cognition 17 (2), 2014, 410–430 C© Cambridge University Press 2013 doi:10.1017/S1366728913000400

Cognitive shift in the bilingualmind: Spatial concepts inKorean–English bilinguals∗

H A E I N PA R KGeorgetown UniversityN I C O L E Z I E G L E RUniversity of Hawai‘i at Manoa

(Received: May 28, 2012; final revision received: June 21, 2013; accepted: June 26, 2013; first published online 8 October 2013)

Recent research shows that speakers with different first languages perceive various cognitive domains in different ways andthat bilinguals’ conceptualization patterns are affected by the concepts of both languages. The present study extends Choiand Bowerman (1991) by examining the categorizations of spatial concepts in adult Korean–English bilinguals. Usingsimilarity judgments, the study compares the conceptualization patterns of “put in” and “put on” by Korean–Englishbilinguals and Korean and English monolinguals. Results indicated significant differences between the monolingual andbilingual groups, demonstrating the process of convergence of the two languages in the bilingual mind. Regression analysesalso revealed that bilinguals’ conceptualizations are largely influenced by English proficiency and frequency of Korean use.These findings provide evidence of conceptual restructuring, supporting the claim of previous research that bilinguals’conceptualization patterns are susceptible to their language experience.

Keywords: cross-linguistic influence, cognitive shift, linguistic relativity, bilingualism

Introduction

Recent empirical studies have demonstrated cleardifferences between speakers with different nativelanguages in a variety of cognitive domains, includingcolor (e.g., Davidoff, Davies & Roberson, 1999;Roberson, 2005), time (e.g., Boroditsky, 2001), space(e.g., Levinson, 1996; Majid, Bowerman, Kita, Haun &Levinson, 2004), and grammatical number (e.g., Lucy &Gaskins, 2003). Overall, these studies lend support to theidea that speakers of different languages are influencedby their language-specific linguistic patterns to makedifferent types of evaluations and observations of theworld (Whorf, 1956). This hypothesis, known as thelinguistic relativity hypothesis, suggests that a speaker’slanguage directs him or her to perceive various cognitivedomains differently than speakers of other languages, thusresulting in speakers’ different perceptions. Although theidea of the validity of linguistic relativity has been asource of lively debate in the fields of linguistics andpsychology for the last few decades, recent research hasprovided cogent evidence that various language-encodedconcepts, including lexical and grammatical features,affect speakers’ different cognitive processes, such asclassification, categorization, and similarity judgments.

∗ This study was first developed in Cristina Sanz’s Bilingualism andCognition seminar at Georgetown University, and presented at SLRF2011 (Iowa State University, October 2011). The authors would liketo thank Cristina Sanz and Yunkyoung Kang for their insightfulcomments and guidance on an earlier version of this paper. Wewould also like to thank reviewers for their constructive commentsand suggestions. However, any remaining errors are exclusively theresponsibility of the authors.

Address for correspondence: Hae In (Lauren) Park, Department of Linguistics, Georgetown University, Poulton Hall 239, 1437 37th Street NW,Washington, DC 20057-1051, [email protected]

In addition, revived interest in cross-linguisticinfluence has also pushed bilingual research to explorethe effects of knowing and using two languages, thuscontributing to the expansion of the scope of linguisticrelativity. For example, the effect of using two or morelanguages on the structuring of one’s cognitive domainshas been carefully examined in relation to bilingualism’sassociated factors, such as age of onset, frequency oflanguage use, and language proficiency. More specifically,studies have sought to answer the following two questions:If a speaker’s second language (L2) contains differentgrammatical and lexical properties than the first (L1),how might the two languages interact and influence thespeaker’s categorization and conceptualization patterns?Will the L1 preferences maintain dominance or will theinfluence of the L2 result in cognitive restructuring? Thecurrent study adds to the growing body of bilingualresearch (Athanasopoulos, 2006, 2007, 2009, 2011;Athanasopoulos & Kasai, 2008) examining whetherbilinguals’ categorical perception remains in a staticstate, formed and maintained by the speaker’s L1, orwhether the learning of lexical and grammatical featuresin a new language results in cognitive restructuringin the bilingual mind. Using an experimental design,this research seeks to investigate the use of spatialdomains (i.e., containment and support) by Korean–English adult bilingual speakers in comparison to thoseof Korean and English monolinguals. By examiningvarious factors that influence conceptual restructuring,this study aims to offer a more complete account of theinfluence of novel language acquisition on the bilingualmind.

http://journals.cambridge.org


Cognitive shift in the bilingual mind 411

Categorical concepts of monolingual speakers

Cross-linguistic research in L1 acquisition has demon-strated that languages may differ in the way they interpretand categorize different conceptual domains such asquantification, color, and space (Choi & Bowerman,1991; Levinson, 1996, 2001). For instance, speakers oflanguages that make a clear distinction between countand mass nouns, such as English, were found to be morelikely to make similarity judgments based on commonshape, while speakers of languages that quantify all nounsas mass nouns, such as Japanese or Yucatec, were moreprone to base similarity judgments on material properties(Imai & Gentner, 1997; Imai & Mazuka, 2003; Lucy,1992; Lucy & Gaskins, 2001). In Lucy and Gaskins(2001, 2003), the findings indicated that adult Englishand Yucatec speakers categorized non-solid substances ina similar manner. However, when presented with triadscontaining solid substances in a similar judgment task,the English speakers matched items based on shape whilethe Yucatec speakers matched items based on material.This difference in categorization was attributed to the factthat English has count nouns that encode individuality,which is best conceptually represented by shape, whereasYucatec does not. Due to the absence of count nouns inYucatec, speakers’ attention is drawn to other conceptualrepresentations, such as material. The authors concludedthat these linguistically specific categorization patternsare formed in childhood, and vary between differentlanguages.

Another cognitive domain that has been the focus ofrecent bilingual research is that of color perceptions.A number of studies have suggested that speakers ofdifferent L1s divide the color spectrum differently anddemonstrate different lexical coding of color. For instance,speakers of Russian are able to distinguish betweenshades of blue more quickly and with finer distinctionsthan English speakers. Such conceptual differences canbe explained by the inherent differences in the colorproperties of the speaker’s L1: Russian has a requiredlexical distinction between light and dark shades of blue,while English does not make a lexical distinction betweenthese color shades (Roberson & Davidoff, 2000; Winawer,Witthoft, Frank, Wu, Wade & Boroditsky, 2007).

The influence of language-specific input on cognitionhas also been confirmed in cross-linguistic studies onspatial domains (Levinson, 1996; Talmy, 1975, 1985),suggesting that languages also differ in the way theysemantically structure space. For example, L1 research(e.g., Bowerman, 1996; Bowerman & Choi, 2001, 2003;Choi & Hattrup, 2012; Choi, McDonough, Bowerman &Mandler, 1999; McDonough, Choi & Mandler, 2003) hasshown that both child and adult speakers of Korean andEnglish differ significantly in grouping and partitioningvarious types of containment and support relations.

Their findings indicate that children learning Englishas their first language make a distinctive semanticcontrast between the spatial particles in and on basedon whether the relation involves containment or support,respectively. On the other hand, Korean children acquirea path verb kkita, which cuts across the English “on–in” contrast; that is, what English considers to be unifiedcategories (e.g., put on and put in) are subdivided intotight-fitting and loose-fitting categories for speakers ofKorean. Tight-fitting events can be described as kkitairrespective of support or containment, while verbsvary respectively with close containment and support(see Figure 1). Although Lucy and Gaskins (2001,2003) stated that language-specific representations areformed around the ages of seven to nine, Choi and hercolleagues have shown that children as early as 18 monthsold understand nonlinguistic spatial categorization in alanguage-specific way (Choi et al., 1999). These findingsclearly demonstrate that the semantic organization of aspeaker’s language plays a significant role in shapinghis/her nonlinguistic spatial concepts, as well as the factthat these conceptual organizations may be formed duringearly childhood.

Categorical concepts in the bilingual mind

Overall, cross-linguistic research suggests that languageand conceptual cognition are connected, and that linguisticinfluence on one’s mind is evident from a very early age.As interest in bilingual cognition research has grown, thetopic of linguistic relativity has been more extensivelydiscussed in relation to bilinguals. Given that speakersof different languages evaluate the perceptual attributesof their reality in different ways, recent research hasbeen concerned with the issue of how learning and usingtwo languages, particularly when they contain differentgrammatical and lexical categories, affects conceptualcategorization within the bilingual mind. Pavlenko(2011a) suggests that the bilingual mind undergoesdifferent processes of cognitive shift depending on wherethe speaker falls on the continuum of monolingualismand bilingualism. As speakers learn and use an L2, novelperspectives and frames of reference from the L2 maybe internalized, leading to the restructuring of the framesand categories of the speaker’s L1. This cognitive change,represented as transformations in speakers’ linguisticcategories, can be manifested in the bilingual mind ina number of ways.

One of the processes that the bilingual mind mayexperience is what Pavlenko (2011a) refers to as co-existence, in which bilinguals maintain the categoriesunique to each language and use them in agreement withthe constraints of specific languages. Sachs and Coley(2006) provided evidence for co-existence through theirinvestigation of Russian–English bilinguals’ emotional



412 Hae In Park and Nicole Ziegler

conceptualizations of envy and jealousy. Results indicatedthat bilinguals tested in English perceived envy andjealousy like English monolinguals, while bilingualstested in Russian categorized the emotions similar toRussian monolinguals. These findings offer support forthe idea that bilinguals are capable of using native-likeconceptualization patterns in both L1 (i.e., Russian) andL2 (i.e., English) emotional categorization tasks, and thatthese patterns can be conveniently accessed by bilingualspeakers depending on their language environment.

Another process that bilinguals commonly experienceis the influence of the L1 on the L2 (Pavlenko, 2011a).During this process, bilingual speakers’ L2 performanceis guided by L1 properties, preferences, categories,and frames of reference. Bilinguals experiencing L1influence may be thinking in their L1 in order to driveperformance in the L2, although this reliance may beimperceptible even to the speaker. Empirical evidencesupporting conceptual transfer has been found mainlyin studies examining object categorization, particularlyin unbalanced bilinguals of beginning and intermediateL2 proficiency. In Athanasopoulos (2006), in which thegrammatical representation of number in intermediateand advanced Japanese–English bilinguals was examined,findings confirmed that proficiency could be a significantfactor in shaping bilinguals’ conceptual cognition. Resultsindicated that the intermediate bilinguals continuedto match stimuli based on changes in the numberof Animals, adhering to a more L1-Japanese–basedconceptual representation, rather than matching stimulibased on the number of Implements or Substance,considered to be a more L2 English representation.This suggests that the stronger influence of the L1 onthe bilinguals’ L2 performance is more evident withbilinguals of lower L2 proficiency. Furthermore, thesefindings support Pavlenko’s (2005) claim that bilingualcognition is susceptible to the influence of languagedevelopment.

The third process, and the most relevant to thecurrent research, is the convergence of L1 and L2categories, whereby a unitary conceptual category distinctfrom both L1 and L2 concepts emerges. Since thenew category does not fully resemble either an L1-based or L2-based category, the change is also deemeddivergent (Jarvis & Pavlenko, 2008). During thisphase, bilinguals demonstrate conceptualization patternsthat are between those of monolingual speakers ofboth languages, exhibiting what is called “in-betweenperformance” (Pavlenko, 2011a, p. 247). Bilinguals’conceptualization patterns may contain some featuressimilar to L1 monolinguals and some features similar toL2 monolinguals, but they would not completely matchone particular language-specific pattern. Such L1–L2integration has been identified in a number of cognitivedomains, including color (Jameson & Alvarado, 2003)

and object naming (Cook, Basetti, Kasai, Sasaki &Takahasi, 2006). In addition, Hohenstein, Eisenberg andNaigles’ (2006) study of early- and late-onset Spanish–English bilinguals also provided evidence supportingthe process of convergence in the domain of motion.Using twelve videos depicting different motion events,the researchers examined whether the use of manneror path verbs in Spanish or English was affected bythe acquisition of English as an L2. After viewingthe videos, participants were interviewed about theactions represented in each video. Results indicated thatbilinguals’ in-between performance in conceptualizingmotion was reflected in their token frequency of mannerverbs: in their Spanish descriptions, bilinguals usedfewer manner verbs than the English monolinguals,but more than the Spanish monolinguals. Conversely,participants also used more manner verbs than the Spanishmonolinguals, but fewer than the English monolinguals,in their English descriptions. The bilinguals’ divergencefrom both L1 and L2 patterns clearly reveals thatSpanish–English bilinguals’ conceptualization of motionsno longer aligned with that of either monolingualgroup, suggesting convergence of L1 and L2 categories.However, convergence of the L1 and L2 is not alwaysmanifested in the form of in-between performance.Bassetti and Cook (2011) state that convergence alsoconsists of patterns that do not necessarily representconcepts that are in-between the two languages. Thus,concepts that are completely different from eitherlanguage could be seen as evidence of convergence sincethey diverge from the classification preferences of themonolingual groups.

Convergence can also be interpreted as evidence ofcognitive restructuring, or the restructuring of a speaker’smental representations of different concepts (Jarvis &Pavlenko, 2008) in the bilingual mind. As bilinguals learnand use their L2, their conceptual patterns may begin todiverge from the initial L1 pattern, changing to representmore closely the patterns of the L2. Evidence of cognitiverestructuring can be seen in recent work in the domain ofcolor (Athanasopoulos, 2009) and object categorization(Athanasopoulos, 2006, 2007; Athanasopoulos & Kasai,2008). Athanasopoulos and Kasai (2008) sought to answerwhether English–Japanese bilinguals would match objectsbased on color or shape. Using triads consisting of colorillustrations of novel objects in a similarity judgment task,participants were required to match the standard with oneof the two alternate illustrations. Results indicated thatadvanced bilinguals performed more like monolingualspeakers of their L2, while intermediate bilingualsperformed in a similar manner to speakers of their L1.These findings demonstrate L1 influence on the L2 forthe intermediate speakers, and convergence, or cognitiverestructuring in the direction of the L2, for the advancedspeakers.




The degree to which bilingual speakers’ conceptual-ization patterns may be affected is influenced by a varietyof factors, including age of onset (Bylund & Jarvis, 2011;Hohenstein et al., 2006; Pavelenko & Malt, 2011), lengthof stay in the L2-speaking country (Athanasopoulos,2007, 2009; Cook et al., 2006), and proficiency (Athana-sopoulos, 2009; Athanasopoulos & Kasai, 2008). Otherstudies have demonstrated that the L1 may be subject torestructuring of linguistic conceptualizations and attritiondepending on environmental factors, such as contextand frequency of use. For instance, Athanasopoulos,Damjanovic, Krajciova and Sasaki (2011) reported thatJapanese–English bilinguals who used English more oftenthan Japanese were more likely to identify shades of bluelike English monolinguals. That is, the color perceptionsof Japanese–English bilinguals were most significantlyinfluenced by the amount of L2 use in their daily lives.Overall, the degree of influence of linguistic and extra-linguistic factors on bilinguals’ cognitive shift remainsunclear, with more research needed to clarify the mostrobust predictors of bilinguals’ conceptual change.

Seeking to expand the scope of studies concerningcognitive change, the current research sets out toinvestigate the conceptualization of space in Korean–English bilinguals. Extending the work of Choi andBowerman (1991), this study will examine whetherKorean–English bilinguals’ conceptualization of space isdivergent from that of Korean and English monolinguals.Specifically, the use of the Korean path verb kkita amongKorean–English bilinguals will be investigated to promotebetter understanding of cognitive change in the bilingualmind. In addition, the current study aims to clarifythe influence of linguistic and extra-linguistic variablescommonly associated with bilingualism, such as age ofonset and arrival, proficiency, and frequency of use,on the cognitive restructuring of adult Korean–Englishbilingual speakers. Despite the fact that bilingual researchon cognitive change has become more productive in recentyears, more research is warranted to understand the role ofdifferent linguistic and extra-linguistic factors on variousconceptual domains, such as space, in the bilingual mind.The three research questions of the current study are asfollows:

1. Do Korean and English adult monolinguals andKorean–English adult bilinguals categorize spatialconcepts differently?

2. If so, how are their spatial conceptualizations different?

3. What is the relationship between bilinguals’ spatialconceptualizations and linguistic and extra-linguisticvariables associated with bilingualism (e.g., languageproficiency, frequency of language use, age of arrival,and age of onset)?

On the basis of previous cross-linguistic studies (e.g., Choi& Bowerman, 1991; Choi & Hattrup, 2012; Lucy & Gask-ins, 2001; McDonough et al., 2003) that have providedevidence for the different conceptual categorizations ofadult speakers of different languages, it is hypothesizedthat Korean and English adult monolingual speakerswill follow the spatial conceptualization patterns ofKorean and English children, respectively, as documentedin Choi and Bowerman (1991). Moreover, we expectKorean–English bilinguals in the current research todisplay evidence of cognitive change, which will bemanifested as L1–L2 convergence in the experiment.Their cognitive shift may be significantly influenced byage-related or language-related variables, such as L2proficiency, frequency of use, and age of onset and arrival,as supported by previous studies (Athanasopoulos, 2009;Athanasopoulos & Kasai, 2008; Athanasopoulos et al.,2011; Hohenstein et al., 2006; Pavlenko & Malt, 2011).

Method

Participants

The final sample consisted of 76 adults in Seoul, Korea,and in Arizona and Washington, D.C. in the U.S., rangingin age from 18 to 69 years old. The mean age ofthe 17 Korean monolingual speakers was 39.88 (SD= 14.29), while the mean age of the 28 monolingualspeakers of English, all undergraduates at a mid-sizeuniversity, was 20.13 (SD = 2.78). The mean age of the31 Korean–English bilinguals, who were either temporaryor permanent residents of the D.C. area, was 26.35years (SD = 3.68). For the current study, bilingualismhas been operationalized as the alternate use of twolanguages that are equivalent in terms of the speaker’sproficiency (Romaine, 1995). Monolingual speakers hadbeen minimally exposed to any foreign language, andtheir dominant language in everyday life was their nativelanguage. Following Bialystok (2006), and Bialystok,Craik and Ruocco (2006), language proficiency wasself-reported in a questionnaire. Proficiency levels werebased on the Common European Framework and potentialratings ranged from complete beginner to native speaker.Participants rated their current Korean and Englishproficiency in reading, listening, speaking, and writingon five-point Likert-type scales, with the sum of thefour ratings calculated to approximate overall proficiencyin each language (proficiency score range: 1–20). Thescores for the native-like proficiency level ranged from17 to 20, the advanced level from 13 to 16, andthe high-intermediate from 9 to 12. Participants alsoreported their current age (AoT), age of onset of L2acquisition (AoO), and age of arrival in the U.S. (AoA)in the language background survey. Among the bilingualparticipants, who were recruited through flyers posted at




Table 1. Descriptive statistics for participants’ age and language variables. Means are reported with standarddeviations (SDs) in parentheses.

Korean English Korean English

N AoT AoA AoO proficiency proficiency use use

Korean monolinguals 17 39.88 (14.29) – – – – – –

English monolinguals 28 20.13 (2.78) – – – – – –

Korean–English bilinguals 31 26.35 (3.68) 13 (5.91) 7.1 (5.19) 18.39 (2.81) 18.06(2.68) 57.52 (24.01) 67.03 (24.48)

AoT = Current Age; AoA = Age of Arrival; AoO = Age of Onset of English; Korean use = total use of Korean (hrs/week); English use = total use of English (hrs/week)

Korean churches in Centreville and Herndon, Virginia, 23bilinguals rated themselves as native-like (17–20 points),five bilinguals as advanced (13–16 points), and threebilinguals as high-intermediate (12 points). To gaugeparticipants’ Korean and English language use, weeklyaverage estimates of language use were reported in hours.Participants indicated the number of hours they spentperforming a list of activities (e.g., talking on the phone,texting, browsing the internet, watching TV, interactingsocially, etc.) in each language.

Korean monolingual participants were recruited fromuniversities and churches in Seoul, Korea, through theposting and distribution of flyers. Since the researcherswere not onsite in Korea, English instructors and teachertrainers with experience in conducting linguistics researchwere trained to serve as recruiters and proctors for thestudy. Although a monolingual speaker is by definitionan individual who speaks only one language, followingthe precedent set in previous research (Athanasopoulos,2009, 2011; Athanasopoulos et al., 2011; Brown &Gullberg, 2011), a more relaxed operationalization ofthe term has been adopted for the current study sincemost Koreans start receiving English education from avery early age. Therefore, Korean monolinguals recruitedfor the current study were functional monolinguals withminimum English proficiency. Also, it is noteworthy tomention that the mean age for the Korean monolingualgroup was higher than the other groups given that Koreanspeakers in the age range from 20 to 30 years generallyhave been exposed to English language experiencethroughout their secondary and post-secondary education.For English monolingual participants, compositioninstructors at the university where the participants wererecruited were trained as proctors for the study. Englishmonolingual participants’ prior language experience wasalso examined to ensure that they had minimal exposure toany additional language, including Korean. Although theoriginal sample included 88 participants, twelve Englishmonolingual participants had their data discarded due totheir experience with a second or third language, andtwo of the bilingual participants were eliminated fromthe sample for not completing the tasks and introspectivemeasures correctly. Descriptive statistics for age and

Figure 1. Categorization of some spatial events in Korean(adapted from Bowerman & Choi, 2003).

language variables of the three groups are presented inTable 1.

Materials

Materials for the current study have been designedfollowing Bowerman and Choi (2003), which examinedthe different spatial categorizations of English and Koreanmonolingual children. Twenty spatial events were usedfor Tasks 1 and 2, and 22 pictures were created by theresearchers to illustrate these concepts. Given the findingsof previous literature (Bowerman & Choi, 2003; Choi& Hattrup, 2012; Choi et al., 1999; McDonough et al.,2003), it was predicted that the Korean and Englishmonolinguals would categorize spatial events asrepresented in Figures 1 and 2.

Experimental tasks

The two experimental tasks used in the current studyhave been adopted from Sachs and Coley (2006), andwere utilized as nonverbal measures of participants’conceptualization of space. Instructions and tasks were




Figure 2. Categorization of some spatial events in English(adapted from Bowerman & Choi, 2003).

administered to both monolingual groups in their nativelanguage by a native speaker of that language, while bilin-gual participants were randomly assigned to receiveinstructions and tasks in either Korean or English. Thelanguage of instruction (i.e., language in which theexperiment is performed) was counterbalanced to avoidan instruction effect. Photographs used for both tasks wereextensively piloted with Korean and English monolingualsand Korean–English bilinguals to confirm that spatialconcepts from each language were accurately represented.To ensure that both tasks measured participants’ nonverbalcategorization as opposed to verbal categorization, anyword that may trigger linguistic processing (e.g., think,come up with, remember, represent, etc.) was avoided inthe verbal and written instructions. Therefore, participantsreceived jargon-free instructions that asked them toexamine the picture based on the action taking place andmake choices accordingly. In addition, the task was strictlytimed in an attempt to restrict participants from engagingin analytic thinking, which occurs more often when theallotted time is uncontrolled.

Task 1: Triad matching taskParticipants were presented with 40 sets of picture triads,consisting of 29 critical item sets and 11 distractor sets,which were administered using PowerPoint presentationson individual computers. Each set consisted of twophotographs describing the same spatial concept and onephotograph referring to a different spatial concept. Withinthe given time of seven seconds per set, participants wereasked to select the image that did not match the other

two images in spatial concept.1 Critical item sets weredesigned to yield two different interpretations so thatparticipants’ answers would vary depending on how eachindividual conceptualized space. For example, reliance onthe Korean conceptual system would prompt participantsto choose an answer that was different from thoseparticipants adhering to the English conceptual system.Thus, Korean and English monolinguals were expected tobehave differently in critical item sets while distractor setsallowed only one answer for both monolingual groups.Examples of expected categorizations by Korean andEnglish monolinguals in each set are shown in Figures 3and 4.

Participants were asked to indicate their answers foreach triad on the provided answer sheet. In order to preventparticipants from basing their decision on other aspectsof the pictures (i.e., the similarities in color or type ofthe object), they were encouraged to focus on the actiondepicted in each picture. Arrows were included in some ofthe images to highlight motions in the pictures. Since Task1 was intended to measure bilinguals’ intuitive responses,as opposed to their well thought out answers, the taskwas timed (7 seconds per set), and backtracking was notallowed during the task.

Task 2: Free sort taskParticipants were presented with nine photographs ofdifferent spatial concepts using PowerPoint presentationsand individual computers. They were instructed to sortthe pictures into as many groups as they liked accordingto the kind of spatial concept represented in the image.2

The number of items for Task 2 was smaller than Task

1 The English instruction for Task 1 was as follows:

You will be presented with 40 sets of pictures on the screen. Each setconsists of three pictures: two pictures describing the same action andone picture describing a different action. Pay attention to the arrow(s)on the picture and choose the picture that does not match with theother two pictures. To indicate your answer, circle the correspondingnumber on the answer sheet. You will have seven seconds (per set)to make your decision. If you cannot make a choice within the giventime, leave the question blank. You cannot backtrack your answers.

The equivalent instruction in Korean was:

��1� 40�� . � ��

� �� 2 ��

� 1�� . ��

� �� . ��

� 7��. 7��

� � �� . ��

� � ��.

2 The English instruction for Task 2 was as follows:

Categorize pictures into as many groups as possible according to thekind of action. A category may consist of a minimum of one picture.You may or may not use all six categories provided on your answersheet. However, you must categorize all nine items. You have threeminutes to complete the task.




Figure 3. (Colour online) Critical item set in Task 1: Expected categorization by Korean and English monolinguals.

1 given that the nature of Task 2 was more cognitivelychallenging; that is, instead of providing participants withoptions to choose from as in Task 1, Task 2 required themto create their own categories and sort items accordingly.Of the nine pictures, eight were critical items and onewas a distractor item. The eight critical items consisted offour complex items that were designed to be categorizeddifferently by Korean and English monolinguals (i.e.,

The instruction in Korean was

�� . ��

�� . � ��

�� . �� 6��

�� . �� ,9�� . ��

3��.

items referring to tight-fitting events that can be describedas kkita),3 and four simple items that would be classified

3 Of the four critical complex items, three items (Items #3, 8, 9 inAppendix A) represented “tight put-on” whereas one item (Item #4)referred to the “buttoning” motion. One of the limitations that mustbe addressed here is that no “tight put-in” items were included in Task2. To examine whether English speakers have distinct categories for“put in” and “put on”, it would be best to include critical complexitems that represent “tight put-in” as well as “tight put-on”. However,since Task 1 necessitated more items to create 40 triads, the limitednumber of “tight put-in” items were all used for Task 1. It should alsobe noted that Item #4 (to button) is different from the other complexitems in that it does not describe the motion of kkita that crosscutsthe categories of “put in” and “put on”. Nevertheless, “buttoning” is atype of kkita that extends to some situations that are neither “put in”nor “put on” in Korean, and examining a) whether Korean speakerswould categorize this item with other complex items; and b) whether




Figure 4. (Colour online) Distractor set in Task 1: Expected categorization by Korean and English monolinguals.

in the same way by both monolingual groups (i.e., itemsreferring to loose-fitting events). Critical items describedthe concepts of put in, put on, or kkita, while the distractoritem illustrated an irrelevant concept (i.e., rip). Thedetailed information regarding the nine items is shownin Appendix A. The maximum time limit for this taskwas three minutes to ensure that participants based theirdecisions on an intuitive response rather than a calculatedresponse, thus providing more accurate representations oftheir nonverbal conceptualizations. Examples of expectedcategorizations by Korean and English monolinguals areshown in Figure 5.

Procedure

Bilingual participants met either individually or insmall groups with the researcher(s) to complete thetasks. During the experiment, participants completedinformed consent paperwork and the two experimentaltasks, which were each preceded by a short practicetask. In addition, Task 2 was followed by a shortdebriefing questionnaire (see Appendix B) designed toobtain additional information about their decision-makingprocesses during the second task. For instance, thedebriefing questionnaire asked participants to label thecategories created during Task 2 (no modification of theexisting categories allowed) and to indicate the difficultylevel of each item (easy vs. confusing). Labels wereused as a reference to better understand and evaluateparticipants’ nonverbal categorizations. If items wereindicated as difficult to categorize, participants wereasked to explain the reasons for their difficulty. These

English speakers would single out this item from other complex itemswould still provide information about different conceptualizations ofthe two monolingual groups.

questions were designed to elicit information regardingthe participants’ categorization processes. Following thecompletion of the two tasks, all bilingual participants wererequired to complete an online questionnaire addressingtheir language background, language use, and educationalexperience. Those who did not complete the survey wereexcluded from the final sample pool.

The same tasks and debriefing questionnaire wereadministered to the Korean and English monolingualparticipants. Since the researchers were not onsite at thelocations where the English and Korean monolingualparticipants were recruited, English instructors andteacher trainers were trained to serve as proctors forthe task administration. Following completion of theexperiment, monolingual participants also participatedin an online background survey measuring languagebackground and educational experience.

Scoring

Each triad set in Task 1 consisted of three pictures, withthe possibility of three unique answers. Participants couldselect a picture that described a different spatial concept inthe Korean conceptual system or in the English conceptualsystem. However, there was also a third picture thatwould not be considered an odd spatial concept in eitherthe Korean or English categorization, but could still bepotentially chosen as an answer. For each critical triadset (N = 29), one point was given to one of the threecategories (Korean, English, and Other) depending onparticipants’ choices of a different action. For instance,if a participant chose a picture that described a differentspatial concept in the Korean conceptual system, one pointwas given to the Korean category. Participants’ selectionswere analyzed by calculating the proportion of times they




Figure 5. (Colour online) Task 2: Expected categorization by Korean and English monolinguals.

categorized pictures in the Korean, English, or Otherway. The possible maximum points that each conceptualcategory could earn were 29 points and the minimumwas zero. The three scores for each participant were thenconverted to a percentage score to facilitate between groupcomparison.

For Task 2, the same scoring measure was utilized withan addition of a subjective measure. For the interest ofthe present study, participants’ categorization of the fourcomplex items4 (Items #3, 4, 8, 9 in Appendix A), i.e.,the items that could potentially be categorized in either a

4 Simple items (i.e., items referring to loose-fitting events) were notsubject to analysis as participants were expected to categorize itemsin the same way irrespective of their language background.

Korean or English manner, were coded in order to accountfor similar conceptual categories and to identify language-specific conceptual categorizations. First, participants’answers were carefully analyzed with their categorylabels, which they were asked to indicate in the debriefingquestionnaire. However, since the current study focusedon the restructuring of non-linguistic concepts as opposedto linguistic concepts, category labels alone did not playa decisive role in scoring. For instance, if a participantcreated two distinct categories for “put on” and “placeon”, they were merged into one category since the mainfocus of the study was on distinguishing the “on”–“in”difference rather than on identifying the exact main verbfor the spatial concept. Next, the created categories wereexamined to determine whether they represented Korean,




Table 2. Descriptive statistics for participants’ spatial conceptualization scores in Task 1.

Korean category English category Other category

M SD M SD M SD

Korean monolinguals (N = 17) 86.06 6.19 6.35 4.30 7.65 5.02

Korean–English bilinguals (N = 31) 37.94 18.14 52.90 17.57 8.84 4.77

English monolinguals (N = 28) 7.61 3.67 85.79 4.68 6.39 4.47

English, or Other way of categorization. On the basis ofprevious research on spatial categorization of Korean andEnglish monolinguals (Bowerman & Choi 2001, 2003;Choi & Hattrup, 2012; Choi et al., 1999; McDonoughet al., 2003), Korean monolinguals were expected toclassify the four complex items into a single category(potentially labeled as kkita), while English monolingualswere expected to categorize them into two separatecategories (Items #3, 8, 9 in one category potentiallylabeled as “put on”; item #4 in another category potentiallylabeled as “button”). If participants failed to conceptualizethe motion according to the English or Korean spatialconceptual system, and therefore categorized an item in away that adhered to a different spatial concept independentof its related items, it would be considered as the “Other”way of categorization. For example, if a Korean speakerput Items #3, 4, 8 in one category labeled “kkita”, andcreated another category for Item #9 with a label thatrepresents neither Korean nor English conceptualization,he/she received three points for the Korean category andone point for the Other category. Each category receivedas many points as the number of complex items in thecategory. The possible maximum score that each category(i.e., Korean, English, and Other categories) could earnwas four points and the minimum was zero. The threecategory scores for each participant were then converted topercentage scores to facilitate between-group comparison.

Results

Task 1

Participants’ responses to Task 1 were converted topercentage scores and the mean score for each languagegroup in each of the three conceptual categories wascalculated. Table 2 shows the mean percentage scores andstandard deviations of the Korean, English, and Otherconceptual categories for each language group in Task 1.

Since the Task 1 scores did not meet the underlyingassumptions for normality, separate Kruskal-Wallis non-parametric tests were used to examine the mean scoresof the Korean, English, and Other conceptual categoriesacross the three groups. Although there were no significantdifferences in the Other category, H(2) = 2.94, p = .230,results showed a significant difference between the three

groups in both the Korean, H(2) = 62.68, p < .001,and the English categories, H(2) = 62.82, p < .001.Separate Mann-Whitney U-tests were then conductedto examine which groups were statistically significantfrom one another. The results indicated that there weresignificant differences between the Korean and Englishmonolinguals in the Korean category, U = .00, p <

.05, as well as in the English category, U = .00, p <

.05. Korean–English bilinguals significantly differed fromKorean monolinguals (in the Korean category, U = 10.50,p < .05; in the English category, U = 7.50, p < .05)and English monolinguals (in the Korean category, U =16.50, p < .05; in the English category, U = 17.50, p <

.05) in both the Korean and English categories. To explorethe relationship between participants’ categorization andvariables associated with bilingualism (e.g., languageproficiency, frequency of language use, age of arrival,and age of onset), correlational analyses were conducted.Table 3 shows a correlation matrix of the variables thatwere examined in Task 1.

Pearson’s correlation coefficient indicated significantcorrelations between English proficiency and bilinguals’Korean category scores (r = –.45, p < .05), as well asbetween English proficiency and English category scores(r = .44, p < .05). Moreover, bilinguals’ total Korean usenegatively correlated with English category scores (r =–.41, p < .05). Both age of onset (r = –.34, p = .064) andage of arrival (r = –.31, p = .095) approached significantcorrelations with participants’ English category, whiletotal Korean use approached significance with bilinguals’Korean category (r = .35, p = .056). Total English useand Korean proficiency were not significantly correlatedwith the English or Korean categories.

Following the correlational analysis, the independentvariables of English proficiency and total Korean use wereentered into a simultaneous multiple regression analysis.The overall regressions for both the Korean category,F2,28 = 5.62, p < .05 (R2 = .29), and the English category,F2,28 = 6.70, p < .05 (R2 = .32), were statisticallysignificant. Table 4 shows unstandardized Beta weightsand significance values for the Korean and Englishcategories, respectively.

English proficiency and total Korean use emergedas significant predictors for both Korean and Englishcategories. Next, separate hierarchical regressions were




Table 3. Results of Pearson’s correlation coefficient for the variables examined in Task 1.

Korean English Korean English Korean English Other

AoA AoO proficiency proficiency use use category category category

AoA 1 .49∗∗ .34 −.48∗∗ .12 −.12 .29 −.34 .11

AoO .49∗∗ 1 −.09 −.34 .17 0 .26 −.31 .12

Korean proficiency .34 −.09 1 .05 .26 .07 .05 −.08 .1

English proficiency −.48∗∗ –.34 .05 1 –.13 .36∗ –.45∗ .44∗ .1

Korean use .12 .17 .26 –.13 1 .56∗∗ .35 –.41∗ .16

English use –.12 .004 .07 .36∗ .56∗∗ 1 .15 –.19 .09

Korean category .29 .26 .05 –.45∗ .35 .15 1 –.95∗∗ –.3

English category –.34 –.31 –.08 .44∗ –.41∗ –.19 –.95∗∗ 1 –.01

Other category .11 .12 .1 .1 .16 .09 –.3 –.01 1

∗p < .05, ∗∗p < .01

AoA = Age of Arrival; AoO = Age of Onset; Korean use = total use of Korean (hrs/week); English use = total use of English (hrs/week)

Table 4. Results of multiple regressions applied toidentify outcomes related to the Korean/Englishcategories for Task 1.

Korean category as the

dependent variable

B weights

(unstandardized) t p-value

English proficiency –4.12 –2.56 .016

Korean use .29 1.82 .080

English category as the

dependent variable

B weights


English proficiency 2.58 2.51 .018

Korean use –.27 –2.31 .028

Note: For Korean as the dependent variable, R2 = .29; for English, R2 = .32.

conducted on both Korean and English categories in orderto partial out the effects of the predicting factors identifiedin the simultaneous regressions. The key factors enteredinto the hierarchical regressions were English proficiencyand total Korean use. English proficiency was introducedas the primary variable due to its high correlation and Betavalue, followed by total Korean use. Table 5 summarizesthe results of the hierarchical regression model for theKorean and English categories, respectively.

The results demonstrated that English proficiencyemerged as the only significant variable, accounting forapproximately 20% of the variance in the Korean category.Total Korean use did not show significant levels, butit increased the percentage of variance explained by

Table 5. Results of hierarchical regressionsapplied to identify outcomes related to theKorean/English categories for Task 1.


dependent variable

R2 change p-value

English proficiency .2 .011

Korean use .09 .08


dependent variable

R2 change p-value

English proficiency .19 .013

Korean use .13 .028

the model by 9%. As for the English category, Englishproficiency and total Korean use showed significant levels,with English proficiency accounting for approximately19% of the variance and total Korean use accountingfor approximately 13% of the total variance. Thesefindings suggest that English proficiency was predictiveof bilingual speakers’ performance on the English andKorean categories, while total Korean use was onlypredictive of their performance on the English category.

Task 2

Quantitative analysisFollowing the conversion of participants’ Task 2 scoresto percentage scores, mean scores of the three categorieswere calculated for each language group. Table 6 shows




Table 6. Descriptive statistics for participants’ spatial conceptualization scores in Task 2.

Korean category English category Other category

M SD M SD M SD

Korean monolinguals (N = 17) 94.12 10.93 1.47 6.06 2.94 8.30

Korean–English bilinguals (N = 31) 31.45 37.62 56.45 39.77 12.10 14.25

English monolinguals (N = 28) 10.71 27.58 76.79 31.13 11.61 18.61

Table 7. Results of Pearson’s correlation coefficient for the variables examined in Task 2.

Korean English Korean English Korean English Other

AoA AoO proficiency proficiency use use category category category

AoA 1 .49∗∗ .34 –.48∗∗ .12 –.12 .33 –.29 –.04

AoO .49∗∗ 1 –.09 –.34 .17 .004 .24 –.31 .23

Korean Proficiency .34 –.09 1 .05 .26 .07 .13 –.16 .11

English Proficiency –.48∗∗ –.34 .05 1 –.13 .36∗ –.34∗ .38∗ –.17

Kor use .12 .17 .26 –.13 1 .56∗∗ .21 –.23 .11

Eng use –.12 0 .07 .36∗ .56∗∗ 1 –.04 .09 –.16

Korean category .33 .24 .13 –.34∗ .21 –.04 1 –.93∗∗ –.03

English category –.29 –.31 –.16 .38∗ –.23 –.09 –.93∗∗ 1 –.33

Other category –.04 .23 .11 –.17 .11 –.16 –.03 –.33 1

∗p < .05, ∗∗p < .01

AoA = Age of Arrival; AoO = Age of Onset; Korean use = total use of Korean (hrs/week); English use = total use of English (hrs/week)

the descriptive statistics of the Korean, English, andOther categories for all three language groups inTask 2.

Task 2 scores did not meet the underlying assumptionsfor normality, necessitating the use of separate Kruskal-Wallis non-parametric tests to examine the mean scoresof the Korean, English, and Other conceptual categoriesacross the three groups. Results indicated significantdifferences between the three groups in the Koreancategory, H(2) = 49.08, p < .001, the English category,H(2) = 40.02, p < .001, and the Other category, H(2) =10.74, p < .05. Separate Mann-Whitney U-tests indicatedthat there were significant differences between the Koreanand English monolinguals in the Korean category, U =.50, p < .05, in the English category, U = .00, p < .05,and in the Other category, U = 129.00, p < .05. Similarly,Korean–English bilinguals significantly differed fromKorean monolinguals (in the Korean category, U =28.00, p < .05; in the English category, U = 34.00,p < .05) as well as from English monolinguals (inthe Korean category, U = 167.00, p < .05; in theEnglish category, U = 240.00, p < .05; in the Othercategory, U = 219.00, p < .05). These findings suggestthat the bilinguals demonstrate different categorizationpatterns from the two monolingual groups, confirmingthe results of Task 1. Furthermore, these findingsprovide cogent evidence that the bilingual group hadformed distinct spatial conceptualizations due to the

convergence of the Korean and English linguisticsystems.

Pearson’s correlation coefficient indicated a significantcorrelation between English proficiency and bilinguals’English category scores, r = .37, p < .05. Although notsignificant, age of onset approached significance, r = –3.14, p = .09, for the English category, while age of arrival(r = .33, p = .07) and English proficiency (r = –.34,p = .07) were shown to approach significance for theKorean category. Table 7 shows a correlation matrix ofthe variables examined in Task 2.

On the basis of the significant variable that emergedfrom the correlational analysis, English proficiency wasentered into a simple linear regression analysis as theonly predictor variable. The results demonstrated thatEnglish proficiency contributed to 11% of the explainedvariance for the Korean category, F1,29 = 3.67, p =.07 (R2 = .11), but the regression model did not reachsignificance at the .05 level. On the other hand, the overallregression model for the English category was statisticallysignificant, F1,29 = 4.86, p < .05 (R2 = .14). Table 8 showsthe unstandardized Beta weights and significance valuesfor the Korean and English categories, respectively.

Overall, results from the regression analyses supportprevious research, with proficiency (Athanasopoulos,2006, 2007) accounting for a significantly large proportionof the variance in the Korean and English categories ofspatial relations. These outcomes indicate that there is a




Table 8. Results of simple linear regressions applied toidentify outcomes related to the Korean/Englishcategories for Task 2.


dependent variable

B weights


English Proficiency –4.7 –1.91 0.07


dependent variable

B weights


English Proficiency 5.62 2.2 0.36

Note: For Korean as the dependent variable, R2 = .11; for English, R2 = .14.

significant relationship between bilinguals’ categorizationand their L2 proficiency. That is, the higher the bilingual’sproficiency in English is, the more likely it is forhis/her classification preferences to pattern with Englishmonolinguals.

Qualitative analysisA qualitative analysis was conducted with Task 2 in orderto provide a more fine-grained account of how spatialcategorizations created by bilinguals differed from thoseof monolinguals. On the basis of previous L1 literature(e.g., Bowerman, 1996; Bowerman & Choi, 2001, 2003;Choi & Hattrup, 2012; Choi et al., 1999; McDonoughet al., 2003), Korean monolinguals were expected tocategorize four complex items (i.e., critical items thatcan be categorized in two different ways) into one singlecategory that refers to the tight fit of the spatial motion. Incontrast, English monolinguals were predicted to providetwo separate categories for items that represent support(i.e., the verb put on) and the motion “to button”. The highmean percentages for (a) the Korean category for Koreanmonolinguals (94.14%) and (b) the English category forEnglish monolinguals (76.79%) in Table 6 suggest thateach monolingual group categorized complex items morein line with the semantics of their language. Examiningthe labels that participants created for each categoryduring the debriefing questionnaire also supported theconclusion that language was the primary guide to theirspatial categorization. Korean monolinguals labeled thecategory that represents tight fit with the verbs kkita orkkiwuta (rough English translation: “to fit in”, “to insert”,respectively), which is another verb that expresses tightfit containment and support. There were several Koreanspeakers who labeled a complex item with the verbcipenehta, which translates into the English verb insert.If they had placed this item in the cipenehta category

along with other motions of containment (Items #2 and5), such a classification was considered to be in line withthe English categorization, as “insert” suggests “to putin in a controlled way (e.g., because the space is small)”(Choi & Bowerman, 1991, p. 94). However, if participantshad created the cipenehta category in addition to the kkitaand put in categories (total number of categories createdfor the four complex items = 3), the complex item inthis particular category would be coded as belongingto the Other category as this shows that the speaker’sconceptualization of spatial relations was not based oneither the “on–in” contrast in English or the “tight-fit”relation in Korean. As for the English monolinguals,they categorized spatial concepts differently from theKorean monolinguals, following the manner predictedby previous research (Bowerman & Choi, 2003; Choi& Hattrup, 2012; Choi et al., 1999; McDonough et al.,2003). Most of them did not create categories or categorylabels in relation to the tight-fit motion, but relied heavilyon the “on–in” contrast. The most common labels for thecategories created by the English monolinguals includeplace on, put on, place down, and put down for supportand button, insert, fasten, put in, and put through for thebuttoning motion. Interestingly, there were three Englishmonolingual speakers who created categories based on the“tight-fit” relation and categorized all four complex itemsinto one “fit” category. These atypical categorizationsmay have contributed to the variations in mean andstandard deviation values. However, despite the variabilityobserved, the examination of participants’ labels in Task2 also suggests that the two monolingual groups wereguided by their L1 to categorize spatial concepts in aspecific manner.

Examining bilingual speakers’ classification in Task2 demonstrated that they exhibited three differentcategorization patterns. First, there were some bilingualspeakers who mirrored the categorization patterns ofeither Korean or English monolinguals. Three bilingualparticipants classified the four complex items in onesingle category according to the “tight-fit” relation(Categorization Type #1 in Table 9), while nine of themcreated two separate categories following the “on–in”contrast (Categorization Type #8 in Table 9). As can beseen in Table 9, the remaining 19 bilingual participants(who created Categorization Types #2–7) diverted fromboth Korean and English categorizations, exhibiting muchmore variety in classifying complex items. This suggeststhat their conceptualizations did not fully resemble eithermonolingual group and that a new conceptual categoryhad been created in the bilingual mind.

The second pattern observed among the bilingualparticipants is what Jarvis and Pavlenko (2008, p. 164)call “a hybrid or amalgam” of the two language-specificcategories. Six bilingual speakers received points in bothKorean and English categories (Categorization Types #2,




Table 9. Frequency of different categorizations createdby Korean–English bilinguals.

Different categorizations of the

four complex items

Categorization Korean English Other

Type N category category category

1 3 100% 0% 0%

2 1 75% 25% 0%

3 4 75% 0% 25%

4 4 50% 50% 0%

5 1 50% 25% 25%

6 1 50% 0% 50%

7 8 0% 75% 25%

8 9 0% 100% 0%

Note: Frequency (N) refers to the number of bilingual participants who createddifferent types of categorizations.

4, 5 in Table 9), indicating that their categorizationswere guided by both languages. For instance, fourbilingual speakers (Categorization Type #4) classifiedItems #4 and 8 as kkita, and Items #3 and 9 as puton. Participants accordingly received two points forthe Korean category and another two for the Englishcategory. These findings show that bilinguals combinedcharacteristics of both Korean and English semanticfeatures, providing evidence of in-between performancein their nonverbal conceptualization. The third patterndisplayed by 13 bilinguals (Categorization Types #3, 6,7) is a type of divergence that does not necessarilycombine both L1-mediated and L2-mediated concepts.Rather, these participants’ classifications seem to havebeen influenced by an alternative criterion that is irrelevantto either language-specific semantics. For instance, twobilinguals in Categorization Type #3 classified Items #3,4, and 8 together as kkita and Item #9 as cover. While itcould be predicted that the former category was guided bytheir L1-mediated concept, it is uncertain what motivatedthe bilinguals to single out Item #9 from the rest ofthe complex items. In cases like this, the three items inthe kkita category received three points for the Korean

category, and one point was awarded to the Other category.Notwithstanding that this type of categorization pattern isnot characterized as in-between performance, it is stilldeemed evidence of convergence as it represents a uniquecategory that is neither an L1-based nor L2-based concept(Bassetti & Cook, 2011).

In the debriefing questionnaire, participants wereasked to indicate the degree of difficulty (e.g., easy vs.confusing) for each item and explain why certain itemswere more challenging than the others. The number ofparticipants who reported each item to be challengingwas counted and converted to a percentage score. Table 10shows participants’ perceived difficulty scores for all nineitems.

With the Korean monolinguals, there was noobservable difference in their perceived difficulty betweensimple and complex critical items. Their responsesindicated that most of the difficulties were either relatedto comprehending the action represented in the pictures(e.g., some of the arrowheads were not clear enough)or irrelevant to the nature of the actions presented.The English monolinguals, however, indicated that oneparticular complex item (Item #4 to button) was morechallenging than the other items. According to theirresponses in the debriefing questionnaire, the difficultymainly arose from the dilemma of whether to classifyItem #4 separately from items that refer to the phrasalverb put in. There were also several participants whoindicated that Item #4 caused trouble because it did notbelong to either category that they had created (whichwere later labeled by the participants as put on and put in).These findings suggest that for the English monolingualspeakers, there was a tendency to group Item #4 withother items referring to the verb put in. Given that inEnglish a major distinction in spatial categorization isgoverned by containment relationships (i.e., the “on–in”contrast) (Choi & Harttrup, 2012), it is not surprisingthat English monolinguals were inclined to merge theconcept of buttoning, which essentially involves “putting”a button “inside” the hole, with the concept of put in, andclassify the items representing these concepts into a singlecategory.

Results of the Korean–English bilinguals’ question-naires demonstrated that all four complex critical items

Table 10. Descriptive statistics for participants’ perceived difficulty on each item in Task 2.

Simple critical items Complex critical items Distractor

Item No. N 1 2 5 6 3 4 8 9 7

Korean monolinguals 17 0 12 6 0 6 18 6 12 0

English monolinguals 27 4 7 4 0 4 30 7 4 0

Korean–English bilinguals 31 3 3 13 3 23 55 11 23 0

Note. Number represents percentage.




Table 11. Bilinguals’ comments about the items perceived as difficult in Task 2.

Item No. Comment

Participant 1 4 I wasn’t sure if the motion was referring to kkita or put in. (translation)

Participant 2 4 It is difficult to decide whether I need to separate #4 to a new category.

Participant 3 4 Depending on which language I use, this motion can be categorized differently. (translation)

Participant 4 8 It’s confusing because in English this could be categorized as wear, but in Korean, it could be

kkita. (translation)

Participant 5 3 There is more than one verb to refer to this motion.

Participant 6 3, 8 I would say “put on your glasses” and “put on your ring” but these kinds of put on are different

from the verb put on used in #1.

Note: The bilinguals were asked to fill out the debriefing questionnaire in the language of their preference. The comments written in Korean have been translated intoEnglish by the researcher who is a native speaker of Korean.

caused more confusion and categorical difficulty than theother items (i.e., simple critical items and the distractor) inTask 2. Table 11 summarizes some of the comments thatwere repeatedly reported by the bilinguals in the debriefingquestionnaire.

These comments indicate that for bilinguals, nonverbalspatial categorization is especially challenging when theirtwo languages employ different semantic categories forspatial motions. Furthermore, some of the bilinguals wereable to identify the source of the problem (see e.g.,comments from Participants 3 and 4 in Table 11) whenasked to reflect on the task. They reported that theirtwo languages differed semantically in the use of spatialrelations, leading to their categorical difficulty. However,as this piece of information was collected via post-taskquestionnaires, it does not necessarily mirror the completeprocess that participants may have experienced duringthe task, suggesting that their reflections on linguisticdifferences should be interpreted with caution.

Discussion

The first research question addressed whether Koreanand English adult monolingual speakers would categorizespatial concepts in different ways. Previous L1 studiesexamining Korean and English spatial categorization(e.g., Bowerman, 1996; Bowerman & Choi, 2001,2003; Choi & Hattrup, 2012; Choi et al., 1999) haveconsistently shown that spatial perceptions are formedfrom infancy to early stages of language learning andthat developmental patterns of spatial categorizationare susceptible to language-specific semantics. Whileinfants learning Korean were found to distinguishspatial concepts in terms of tight fit, English-speakinginfants were inclined to categorize spatial relationsbased on containment. Findings from the current studyprovide clear evidence that adult Korean and Englishmonolinguals exhibit similar categorization behavioras their infant counterparts, suggesting that nonverbal

spatial categorization patterns formed in infancy aremaintained when the primary language environment staysunchanged. The Korean monolinguals performed nearceiling levels on the Korean category in both Tasks 1and 2, indicating that their decisions were made usingfit properties (“tight fit”, “loose fit”). Likewise, theEnglish monolinguals displayed the strongest preferencefor the English category in both tasks, suggestingthat they classified spatial events based on their L1-specific semantics, i.e., containment (“put in”, “put on”).The conceptual differences observed between the twomonolingual groups support the Whorfian view thatlinguistic structure influences non-linguistic cognition andconceptualization (Levinson, 1996; Lucy, 1992; Lucy& Gaskins, 2001). However, the fact that three Englishmonolingual speakers in Task 2 categorized the complexitems according to the concept of tight-fit reveals thatsensitivity to the tight fit and loose fit distinction alsoexists among English monolinguals. This suggests thatmonolinguals of different languages display a preferencetowards certain concepts not because they lack otherconcepts in their language. Choi (2006) and Choi andHattrup (2012) attest that both Korean and Englishmonolinguals display the perceptual salience for tight-fit and containment. Among various ways to segmentmotions into different categories, monolingual speakersare likely to be guided by concepts that are oftenrepresented in their language and used when speakingin that language. In this way, language highlights a certainway of thinking, playing an important role in shapinghabitual thought (Bassetti & Cook, 2011; Boroditsky,2001).

The second research question examined the differencesin spatial concepts between Korean–English bilingualsand Korean and English monolinguals. The bilinguals’conceptualizations of space were significantly differentfrom those of the two monolingual groups in both Tasks 1and 2, with their patterns largely displaying convergenceof L1 and L2 conceptual domains. The qualitative analysis




of their responses in Task 2 demonstrated that threetypes of classification patterns emerged: (a) completeL1 or L2 resemblance, (b) in-between performance, and(c) patterns that did not resemble either L1-related orL2-related concepts. Approximately two thirds of thebilingual speakers (19 out of 31) exhibited patterns (b)and (c), which suggests that the bilinguals created a newunitary category that is distinct from either monolingualgroup. In pattern (b), bilinguals combined characteristicsof both language-specific semantic distinctions to formhybrid, in-between performance, while others displayedpatterns that do not lie between their L1 and L2conceptual patterns in pattern (c). These results indicatethat the use of two languages in one’s everyday lifemay result in the adjustment of bilinguals’ sensitivityto the degree of fit and containment that they initiallyformed as infants. Furthermore, the findings lend supportto the claim that the bilingual mind undergoes cognitiverestructuring as a result of using two different languages(Athanasopoulos, 2011). Although some studies havereported that bilinguals maintain two separate categoriesrelevant to both L1 and L2, accessing the language-specific concepts of each language in accordance withthe constraints posed by one’s environment (Sachs &Coley, 2006), the results of the current study providecontrary evidence for this hypothesis. The convergenceof the bilinguals’ conceptual systems found in the resultsdoes not provide support for the “Modular” view thatcontends that cognition is unaffected by language-specificsemantics (e.g., Malt, Sloman, Gennari, Shi & Wang,1999; Papafragou, Massey & Gleitman, 2002). Instead,our results are more consistent with the Whorfian view,which argues for the influence of language on thought.

While two-thirds of bilinguals demonstrated conver-gence, the remaining one-third patterned with either L1-based or L2-based concepts when categorizing spatialmotions: three bilingual participants mirrored Koreancategorization patterns, and nine of them adhered toEnglish classification preferences. Although pattern (a)does not lend support for unique bilingual patternsof spatial concepts, the correlational analyses provideadditional insights into the process the bilingual mindmay undergo. The negative correlation between Englishproficiency and the Korean category, together with thepositive correlation observed between English proficiencyand the English category suggests that the shift fromL1-based conceptual patterns to L2-based conceptualpatterns is mediated by English proficiency. This reflectsthe dynamic nature of the bilingual mind, showing thatthe restructuring of bilingual systems is an ongoing,developmental phenomenon (Brown & Gullberg, 2013;Pavlenko, 2011) as opposed to a finite outcome (e.g.,Ameel, Malt, Storms & Van Assche, 2009). The L1–L2convergence observed among the bilinguals in the currentstudy, then, should not be viewed as a permanent status,

but rather as “intermediate stages in development” (Brown& Gullberg, 2013, p. 14). As the current findings suggest,spatial concepts in the bilingual mind do not seem to befixed once and for all, but rather seem to be influenced andeven restructured by the acquisition of novel conceptualcategories.

The final research question examined whether there isany relationship between bilinguals’ categorization andlinguistic and extra-linguistic variables. In both Tasks 1and 2, English proficiency was the most robust predictor ofbilinguals’ use of both Korean and English categorizationpatterns. In contrast, Korean proficiency was not found tobe predictive of bilinguals’ performance in either task. Itis possible that the observed lack of correlation betweenKorean proficiency and bilinguals’ categorization patternsis attributed to bilinguals’ inaccurate reporting of theirKorean proficiency. While reporting L2 proficiency occursfrequently with bilinguals living in the L2 context, theyare rarely asked to rate their L1 proficiency in thesecircumstances. In addition, self-report surveys have beenquestioned by scholars as a valid measuring tool (Dörnyei,2001; Schunk, Pintrich & Meece, 2008), as participantsmay under or over report. The inherent challenges associ-ated with self-report measures, as well as the insufficientexperience of self-assessing one’s own L1 proficiency,may have affected participants’ reporting of their Koreanproficiency, resulting in less precise information.

In addition to English proficiency, total Korean useappeared to be the second most effective predictorof bilinguals’ use of English conceptualizations inTask 1. However, it did not affect either Koreanconceptualizations in Task 1 or both Korean and Englishconceptualizations in Task 2. Since overall languageuse for both Korean and English rarely emerged asa significant predictor, the current findings suggestthat language use may not be a robust predictor ofbilinguals’ categorization. Considering that language usewas found to be predictive of bilinguals’ performance inAthanasopoulos et al. (2011), the current findings mayseem contradictory to their results at first. However, it isimportant to note that L2 proficiency in Athanasopouloset al. (2011) was kept constant while more variability wasobserved with L2 proficiency in the current study (seethe Participants section). Therefore, the difference in thevariability of L2 proficiency in the two studies makes itdifficult to claim that their results concerning language useare contradictory to one another. What this rather suggestsis that it is yet too early to make any conclusive remarks onthe role of language use in bilinguals’ conceptualizationbased on these two studies. Thus, more research on the roleof language use in various contexts is warranted beforeany definitive conclusions can be made about this variable.Another factor that must be taken under considerationregarding language use is the potential for inaccuratereporting. Since bilinguals’ language use was obtained




by self-report, there is always a possibility that they mighthave either over- or under-reported the amount of languageuse. The lack of correlation observed between total Koreanuse and Korean proficiency also raises concerns over thevalidity of the language use scores, suggesting that a morecautious approach must be taken when interpreting theresults concerning language use.

Another influential variable that has been frequentlydiscussed in relation to bilingualism is age of onset.Although age of onset in successive bilingualism hasbeen found to be the single most important factor (e.g.,Meisel, 2010), the effect of age of onset on bilingualconceptualization has been deemed rather controversialin the past. While some studies have found it tobe a strong predictor of bilinguals’ cognitive shift(Boroditsky, 2001), others have reported the oppositeoutcome (Athanasopoulos, 2009; Cook et al., 2006).The current results side with Athanasopoulos and hiscolleagues in that no significant effect for age of onsetwas found. However, we cannot completely eliminate thepredictive value of age of onset provided that it marginallyapproached significance in correlational analyses in bothtasks. More research is thus warranted to examine therole of age factors on bilingual conceptualization beforemaking any conclusive statements about its effect.

Overall, the current study contributes to the existingknowledge on the categorization of spatial motionsat several levels. First, we sought to extend previousL1 research on language-specific semantics and spatialcognition by examining the minds of adult monolingualspeakers. Confirming previous research (Choi & Hattrup,2012; McDonough et al., 2003), the current studyalso lends support to previous findings indicating thatlanguage-specific encoding of space acquired in infancyis maintained over time. Second, the issue of language-specific semantics and cognition was brought into thecontext of bilingualism to examine the role of bilingualismin spatial cognition. The results indicated that bilinguals inthe current study showed categorization patterns distinctfrom either monolingual group, suggesting convergenceof the two language systems. However, it is necessaryto acknowledge that there are several limitations to thestudy. As previously mentioned, utilizing self-reports tomeasure bilinguals’ language proficiency and languageuse may have yielded less accurate estimates of thesevariables. Moreover, the relatively small sample size ofthe bilingual population may have contributed to thelack of significance for the variables that marginallyapproached significance in correlational analyses (e.g.,age of onset and age of arrival). A larger sample sizemay provide opportunities for more fine-grained analysesof these potentially predictive variables. Future researchcan also benefit from investigating a wider range ofproficiency levels, including beginner and intermediatelevel L2 speakers, in order to obtain more information

regarding the role and impact of proficiency on bilinguals’cognitive restructuring. Another limitation of this studyconcerns the design of the triad-matching task used inTask 1. In Task 1, participants were presented with threetriad images at the same time, and they were asked toselect one image that did not match the other two imagesin terms of the motion represented. Critical item setswere carefully designed so that two monolingual groupswould yield different answers. That is, a participant relyingon the Korean tight fit relationship would choose Image1 as the odd motion, and a participant employing theEnglish “on–in” contrast would choose Image 2. However,it seems that some of the participants were not able toperceive the motions correctly and, as a result, subscribedto neither conceptual system by selecting Image 3 as theodd motion. The three groups’ mean scores for the Othercategory indicate that a moderate number of participantsselected Image 3 as their answer. As it has been foundthat having this third option for choice makes it difficultfor participants to perceive motions correctly and forresearchers to clearly examine participants’ categorizationpattern, future studies may consider using a differentsimilarity task to conduct research. One alternative isto require participants to determine which of two itemsmost resembles the target item (Athanasopoulos, 2006,2007; Athanasopoulos & Kasai, 2008). Using this typeof methodological design will eliminate the potentiallyconfounding variable of the unrelated third choice, andthus provide a clearer picture of bilinguals’ L1 andL2 conceptual patterns of space. The free sort task inthe study could also be improved by including motionsthat refer to “tight put-in” as well as “tight put-on”. Inorder to examine English monolinguals’ categorization,it would have been ideal to include critical complexitems that represented both types. However, since Task1 necessitated more items to create 40 triads, the limitednumber of “tight put-in” items were all included in Task1. Lastly, to ensure that the tasks employed are measuringbilinguals’ nonverbal spatial categorization, future studiesmay consider conducting a post-experiment survey to askparticipants about whether they were engaged in verbalthinking during the tasks (Choi & Hattrup, 2012). Suchself-report data will help the researchers to make strongerclaims about the nature of the tasks.

Conclusion

The current study sought to examine whether adultKorean–English bilinguals categorize spatial conceptsconcerning containment, support, and contact differentlyfrom Korean and English monolinguals. The findingsindicate that the bilingual participants performeddifferently from both Korean and English monolingualspeakers, showing distinct categorization patterns ofspatial concepts. This can be attributed to the convergence




of their L1 and L2 categories, which reflects the ongoingconceptual restructuring that is taking place in thebilingual mind. These results confirm previous researchthat has demonstrated the effect of novel languageacquisition on cognitive restructuring in bilinguals(Athanasopoulos, 2011; Athanasopoulos et al., 2011;Athanasopoulos & Kasai, 2008). Linguistic variables suchas L2 proficiency and L1 use were found to play significantroles in conceptual restructuring in the direction of theL2. These results also add to the growing body of

research (e.g., Athanasopoulos, 2006, 2007, 2009, 2011;Athanasopoulos & Kasai, 2008; Brown & Gullberg,2011, 2013) demonstrating the influence and effects oflanguage on cognition. However, continued investigationsinto the factors influencing the cognitive restructuring ofbilinguals in domains other than space are needed in orderto generalize findings of the current study. The effects oflinguistic categories on cognition may be domain-specific,and additional research is needed to further clarify theeffect of language on bilinguals’ cognition.

Appendix A. The full list of nine items used in Task 2




Appendix B. Task 2 debriefing questionnaire




References

Ameel, E., Malt, B. C., Storms, G., & Van Assche, F. (2009).Semantic convergence in the bilingual lexicon. Journal ofMemory and Language, 60, 270–290.

Athanasopoulos, P. (2006). Effects of the grammaticalrepresentation of number on cognition in bilinguals.Bilingualism: Language and Cognition, 9, 89–96.

Athanasopoulos, P. (2007). Interaction between grammaticalcategories and cognition in bilinguals: The roleof proficiency, cultural immersion, and language ofinstruction. Language and Cognitive Processes, 22, 689–699.

Athanasopoulos, P. (2009). Cognitive representation of colorin bilinguals: The case of Greek Blues. Bilingualism:Language and Cognition, 12, 83–95.

Athanasopoulos, P. (2011). Cognitive restructuring inbilingualism. In Pavlenko (ed.), pp. 29–65.

Athanasopoulos, P., Damjanovic, L., Krajciova, A., & Sasaki,M. (2011). Representation of colour concepts in bilingualcognition: The case of Japanese blues. Bilingualism:Language and Cognition, 14, 9–17.

Athanasopoulos, P., & Kasai, C. (2008). Language andthoughts in bilinguals: The case of grammaticalnumber and nonverbal classification preferences. AppliedPsycholinguistics, 29, 105–121.

Bassetti, B., & Cook, V. J. (2011). Relating language andcognition: The second language user. In V. J. Cook &B. Bassetti (eds.), Language and bilingual cognition,pp. 143–190. New York: Psychology Press.

Bialystok, E. (2006). Effect of bilingualism and computer videogame experience on the Simon task. Canadian Journal ofExperimental Psychology, 60, 68–79.

Bialystok, E., Craik, F. I. M., & Ruocco, A. C. (2006). Dual-modality monitoring in a classification task: The effectsof bilingualism and ageing. The Quarterly Journal ofExperimental Psychology, 59, 1968–1983.

Bloom, P., Peterson, M., Nadel, L., & Garrett, M. (eds.) (1996).Language and space. Cambridge MA: MIT Press.

Boroditsky, L. (2001). Does language shape thought? Englishand Mandarin speakers’ conceptions of time. CognitivePsychology, 43, 1–22.

Bowerman, M. (1996). Learning how to structure space forlanguage: A crosslinguistic perspective. In Bloom et al.(eds.), pp. 385–436.

Bowerman, M., & Choi, S. (2001). Shaping meaningsfor language: Universal and language-specific in theacquisition of spatial semantic categories. In Bowerman& Levinson (eds.), pp. 475–511.

Bowerman, M., & Choi, S. (2003). Space under construction:Language-specific spatial categorization in first languageacquisition. In Gentner & Goldin-Meadow (eds.), pp. 387–427.

Bowerman, M., & Levinson, S. C. (eds.) (2001). Languageacquisition and conceptual development. Cambridge:Cambridge University Press.

Brown, A., & Gullberg, M. (2011). Bidirectional cross-linguisticinfluence in event conceptualization? Expressions ofPath among Japanese learners of English. Bilingualism:Language and Cognition, 14, 79–94.

Brown, A., & Gullberg, M. (2013). L1–L2 convergence inclausal packaging in Japanese and English. Bilingualism:Language and Cognition, 16, 1–18.

Bylund, E., & Jarvis, S. (2011). L2 effects on L1 eventconceptualization. Bilingualism: Language and Cognition,14, 47–59.

Choi, S. (2006). Influence of language-specific input on spatialcognition: Categories of containment. First Language, 26,207–232.

Choi, S., & Bowerman, M. (1991). Learning to express motionevents in English and Korean: The influence of language-specific lexicalization patterns. Cognition, 41, 83–121.

Choi, S., & Hattrup, K. (2012). Relative contribution of cog-nition/perception and language on spatial categorization.Cognitive Science, 36, 102–129.

Choi, S., McDonough, L., Bowerman, M., & Mandler,J. M. (1999). Early sensitivity to language-specific spatialcategories in English and Korean. Cognitive Development,14, 241–268.

Cook, V. J., Bassetti, B., Kasai, C., Sasaki, M., & Takahashi,J. (2006). Do bilinguals have different concepts? The caseof shape and material in Japanese L2 users of English.International Journal of Bilingualism, 10, 137–152.

Davidoff, J., Davies, I., & Roberson, D. (1999). Color categoriesin a stone-age tribe. Nature, 398, 203–204.

Dörnyei, Z. (2001). Teaching and researching motivation.Harlow: Longman.

Gentner, D., & Goldin-Meadow, S. (eds.) (2003). Languagein mind: Advances in the study of language and thought.Cambridge, MA: MIT Press.

Hohenstein, J., Eisenberg, A., & Naigles, L. (2006). Is he floatingacross or crossing afloat? Cross-influence of L1 and L2 inSpanish–English bilingual adults. Bilingualism: Languageand Cognition, 9, 249–261.

Imai, M., & Gentner, D. (1997). A cross-linguistic study of earlyword meaning: Universal ontology and linguistic influence.Cognition, 62, 169–200.

Imai, M., & Mazuka, R. (2003). Re-evaluating linguisticrelativity: Language-specific categories and the roleof universal ontological knowledge in the construalof identification. In Gentner & Goldin-Meadow (eds.),pp. 429–464.

Jameson, K. A., & Alvarado, N. (2003). Differences in colornaming and color salience in Vietnamese and English.COLOR Research and Application, 28, 113–138.

Jarvis, S., & Pavlenko, A. (2008). Cross-linguistic influence inlanguage and cognition. London: Routledge.

Levinson, S. C. (1996). Frames of reference and Molyneux’squestion: Cross-linguistic evidence. In Bloom et al. (eds.),109–169.

Levinson, S. C. (2001). Covariation between spatial languageand cognition, and its implications for language learning.In Bowerman & Levinson (eds.), pp. 566–588.

Lucy, J. (1992). Grammatical categories and cognition: A casestudy of the linguistic relativity hypothesis. Cambridge:Cambridge University Press.

Lucy, J., & Gaskins, S. (2001). Grammatical categories and thedevelopment of classification preferences: A comparativeapproach. In Bowerman & Levinson (eds.), pp. 257–283.




Lucy, J., & Gaskins, S. (2003). Interaction of language type andreferent type in the development of nonverbal classificationpreferences. In Gentner & Goldin-Meadow (eds.), pp. 465–492.

Majid, A., Bowerman, M., Kita, S., Haun, D. B. M., & Levinson,S. C. (2004). Can language restructure cognition? The casefor space. Trends in Cognitive Science, 8, 108–114.

Malt, B., Sloman, S., Gennari, S., Shi, M., & Wang, Y.(1999). Knowing versus naming: Similarity and thelinguistic categorization of artifacts. Journal of Memoryand Language, 40, 230–262.

Mandler, J. M. (1992). How to build a baby: Conceptualprimitives. Psychological Review, 99, 587–604.

McDonough, L., Choi, S., & Mandler, J. M. (2003).Understanding spatial relations: Flexible infants, lexicaladults. Cognitive Psychology, 46, 229–259.

Meisel, J. M. (2010). Age of onset in successive acquisitionof bilingualism: Effects on grammatical development.In M. Kail & M. Hickman (eds.), Language acquisitionacross linguistic and cognitive systems, pp. 225–247.Amsterdam: John Benjamins.

Papafragou, A., Massey, C., & Gleitman, L. (2002). Shake,rattle, ’n’ roll: The representation of motion in languageand cognition. Cognition, 84, 189–219.

Pavlenko, A. (2005). Bilingualism and thought. In J. F. Kroll& A. M. B. de Groot (eds.), Handbook of bilingualism:Psycholinguistic approaches, pp. 433–453. Oxford: OxfordUniversity Press.

Pavlenko, A. (2011a). Thinking and speaking in two languages:Overview of the field. In Pavlenko (ed.), pp. 237–262.

Pavlenko, A. (ed.) (2011b). Thinking and speaking in twolanguages. Bristol: Multilingual Matters.

Pavlenko, A., & Malt, B. (2011). Kitchen Russian: Cross-linguistic differences and first language object naming in

Russian–English bilinguals. Bilingualism: Language andCognition, 14, 19–45.

Roberson, D. (2005). Color categories are culturally diverse incognition as well as in language. Cross-Cultural Research,39, 56–71.

Roberson, D., & Davidoff, J. (2000). The categoricalperception of colors and facial expressions: The effectof verbal interference. Memory & Cognition, 28, 977–986.

Romaine, S. (1995). Bilingualism. Oxford: Blackwell.Sachs, O. S., & Coley, J. D. (2006). Envy and jealousy in

Russian and English: Labeling and conceptualization ofemotions by monolinguals and bilinguals. In A. Pavlenko(ed.), Bilingual minds: Emotional experience, expression,and representation, pp. 209–231. Clevedon: MultilingualMatters.

Schunk, D. H., Pintrich, P. R., & Meece, J. L. (2008). Motivationin education: Theory, research, and applications. UpperSaddle River, NJ: Pearson Education.

Talmy, L. (1975). Semantics and syntax of motion. InJ. P. Kimball (ed.), Syntax and semantics (vol. 4), pp. 181–238. New York: Academic Press.

Talmy, L. (1985). Lexicalization patterns: Semantic structurein lexical forms. In T. Shopen (ed.), Language typologyand syntactic description: Grammatical categories and thelexicon, pp. 57–149. Cambridge: Cambridge UniversityPress.

Whorf, B. L. (1956). Language, thought, and reality: Selectedwritings of Benjamin Lee Whorf. Cambridge, MA: MITPress.

Winawer, J., Witthoft, N., Frank, M., Wu, L., Wade, A.,& Boroditsky, L. (2007). Russian blues reveal effectsof language on color discrimination. Proceedings of theNational Academy of Sciences, 104, 7780–7785.


Cognitive shift in the bilingual mind: Spatial concepts in Korean–English bilinguals

Documents

Transcript of Cognitive shift in the bilingual mind: Spatial concepts in Korean–English bilinguals