PAPER
Language is not necessary for color categories
Ozge Ozturk,1,5Shakila Shayan,1Ulf Liszkowski 2,4and Asifa Majid 3,4
1.Max Planck Institute for Psycholinguistics,Language Acquisition Department,Nijmegen,The Netherlands
2.Max Planck Institute for Psycholinguistics,Communication before Language Group,Nijmegen,The Netherlands
3.Max Planck Institute for Psycholinguistics,Language and Cognition Department,Nijmegen,The Netherlands
4.Donders Institute 描写山水的古诗名句
for Brain,Cognition,and Behaviour,Radboud University,The Netherlands
5.
Department of Psychology,Princeton University,USA
Abstract
The origin of color categories is under debate.Some rearchers argue that color categories are linguistically constructed,while others claim they have a pre-linguistic,and possibly even innate,basis.Although there is some evidence that 4–6-month-old infants respond categorically to color,the empirical results have been challenged in recent years.First,it has been claimed that previous demonstrations of color categories in infants may reflect color preferences instead.Second,and more riously,other labs have reported failing to replicate the basic findings at all.In the current study we ud eye-tracking to test 8-month-old infants ’categorical perception of a previously attested color boundary (green –blue)and an additional color boundary (blue –purple).Our results show that infants are faster and more accurate at fixating targets when they come from a different color category than when from the same category (even though the chromatic paration sizes were equated).This is the ca for both blue –green and blue –purple.Our findings provide independent evidence for the existence of color categories in pre-linguistic infants,and suggest that categorical perception of color can occur without color language.
Rearch Highlights
•Speech-evoked auditory brainstem respons are more similar among siblings than unrelated age-and x-matched children.
•Unrelated children matched on reading ability also have more similar respons than unrelated children matched only on age and x.
•Speech-evoked auditory brainstem respons reflect both siblingship and common literacy skills.
•
Speech-evoked auditory brainstem respons may be uniquely suited for asssing familial risk for reading impairments
Introduction
In the Welsh language,shades from darkest blue to lightest green are described with the word glas (Lazar-Meyn,2004).In English,of cour,we make a basic lexical distinction between blue and green ,while in languages such as Turkish a further distinction amongst shades of blue are made,distinguishing light blue from
dark blue (€Ozgen &Davies,1998).The differences in
naming may be thought to be mere peculiarities of
linguistic convention,but accumulating evidence from recent years suggests that language determines our color categories.So,while English speakers respond categor-ically to colors across the blue –green boundary,Berinmo speakers (who like the Welsh do not mark the distinction between blue and green)do not respond categorically across the same color boundary (Davidoff,Davies &Roberson,1999;Roberson,Davies &Davidoff,2000;cf.Kay &Kempton,1984).
Cross-cultural findings like the could be dismisd since many things vary between English speakers,on the one hand,and Berinmo speakers,on the other.But the notion that language plays a critical role in color cognition is bolstered by additional evidence.First,categorical perception of color is eliminated by verbal,but not visual interference,suggesting an online medi-atory role of language (Roberson &Davidoff,2000;Gilbert,Regier,Kay &Ivry,2006;Winawer,Witthoft,Frank,Wu,Wade &Boroditsky,2007).Second,cate-gorical perception is found to be stronger in the right visual field,which is preferentially procesd in the language-dominated left hemisphere for both adults,and toddlers who know their color terms (Franklin,Driv-onikou,Bevis,Davies,Kay &Regier,2008a;Gilbert et al .,2006;Roberson,Pak &Hanley,2008).Moreover,
Address for correspondence:Ozge Ozturk,Department of Psychology,New Y ork University,6Washington Place,New Y ork,NY 10003,USA;urk@nyu.edu
Developmental Science 16:1(2013),pp 111–115DOI:
10.1111/desc.12008
it is only found in the left hemisphere of patients with a corpus callosotomy,where the connection between the cerebral hemispheres is vered(Gilbert et al.,2006). Finally,recent ERP studies indicate that the effects are automatic and unconscious(Thierry,Athanasopoulos, Wiggett,Dering&Kuipers,2009).Taken together,the findings may lead us to conclude that language is esntial for color categorization.
Infant color categorization studies have the potential to challenge this conclusion.If pre-linguistic children show categorical effects of color perception then color vocabulary is not a necessary precondition for color categorization.There is some evidence that pre-linguistic infants are able to respond categorically to Bornstein,Kesn&Weiskopf,1976;Catherwood, Crassini&Freiber
g,1990).But some of the studies have been criticized for not adequately matching within-and between-category paration sizes.Previous studies also concluded that infants perceived color categorically even though the methods ud–novelty preference or habituation–cannot distinguish whether the differential responding is a result of perception or memory(cf. Huttenlocher,Hedges&Vevea,2000).So even if there are categorical effects,prior studies have not been able to pinpoint the origin of the effect.
More recently,in a ries of studies conducted by Franklin and colleagues,there is new evidence in favor of categorical perception of color by Clifford, Holmes,Davies&Franklin,2009;Franklin,Pilling,and Davies2005;Drivonikou,Kay,Regier,Ivry,Gilbert, Franklin&Davies,2007;Franklin et al.,2008a).For example,Franklin et al.(2005)ud a target detection task,in which they prented infants with a colored dot on a differently colored background.By measuring infants’eye-movements,they were able to show that infants fixated a target(green)dot faster when it was prented on a background from a different color blue)than when it was prented on a same color n)even when the distance between the target and background was equated across conditions.On the basis of this and related studies,Franklin luded that human categorical responding to color is not entirely lan
guage-mediated and there may be an innate precursor,or at least a pre-linguistic bias,in the perception of color.Moreover,this effect ems to stem from perception,not just memory. The results have not been accepted by everyone, however.In the Franklin et al.(2005)paradigm the target color was kept constant while the background color varied.Roberson and Hanley(2009)argue that Franklin et al.’s findings might,therefore,be a reflection of a simple color preference by the infants for the specific colors ud in this study(although e Franklin,2009).
A cond point of criticism has come from Davidoff, Goldstein and Roberson(2009).Although they acknowl-edge that Fr丹参酮胶囊的功效与作用
anklin and colleagues’(2005)study does not suffer from any of the early faults in the infant color categorization literature,they state that their own attempt to replicate the categorical perception of colors
in infants produced no evidence for color categories(e
also Gerhardstein,Renner&Rovee-Collier,1999).They
report their participants(4-month-olds)found the task
uninteresting and became restless and untestable.
Davidoff et al.(2009;e also Roberson&Hanley,
2009)also criticize Franklin et al.for having tested
infants only for the prence of a single category
boundary(the blue–green boundary).And,in fact,an
earlier study by Gerhardstein et al.(1999)which tested
the red–yellow boundary failed to find categorical
perception of color in infants(cf.Davies&Franklin,
2002).Hence,Davidoff lude,‘It is too soon to
abandon the debate about the simple contrast[nature vs.
nurture][bad on the infant studies]…Given the
significance of the theoretical conquences of innate
color categories,it is hoped that more infant laboratories
will pursue investigations of the issues’(2009,pp.246,
247).
With the current paper we aim to contribute to this
ongoing discussion about the nature of infant color
perception.In order to address the first criticism by
Roberson and Hanley(2009),we showed participants
green,blue,and purple targets prented on either the
same colored background or a different color back-
ground.This way,we test whether Franklin et al.’s
findings could be accounted for in terms of a color
preference for a specific target or background color.
Second,we test for an additional color boundary.While
previous eye-tracking studies with infants have been
limited to the blue–green boundary,we also test the blue –purple boundary.Finally,since other rearchers have reported failing to replicate the categorical perception of
colors in infants(Davidoff et al.,2009),we provide
independent evidence to this endeavor.Given the theo-
retical importance of infant color categories for the
debate about the origin of color categories,repeated
replication with multiple stimulus ts and category
boundaries is desirable.
Method
Participants
A total of43healthy infants took part in the study(22
girls,M=8months,22days;SD=15.5days).Of
the,five infants were excluded from the final sample
due to a reported family history of color vision
deficiency(two infants),due to failure to complete the
procedure as a result of general fussiness(one)and due
to excessive head movement of the infant that made eye-
tracking difficult(two).
Apparatus and experimental tup
During the experiment,infants were placed on an infant
at,on the lap of the parent,who was sitting blindfolded
112Ozge Ozturk et al.
on a chair facing the screen.This procedure allows for cloness between the parent and the infant while the infant is still independently ated and their mobility is restricted by the infant at.The distance between the infant’s eyes and the screen was around60cm.During the experiment,the gaze of both eyes was recorded using a corneal reflection eye-tracker(Tobii1750,Tobii Technology,Stockholm,Sweden).The Tobii eye-track-ing system is integrated in a17-inch TFT flat-screen monitor,on which the stimuli were shown,and the apparatus records eye gaze data at50Hz with an average accuracy of0.5visual angle.The monitor was attached to an adjustable arm,so that the screen could be positioned at the correct distance and kept parallel to the participant’s face.Before the experimental ssion began,the infant’s eye gaze was calibrated for ven out of nine calibration points.
Stimuli and design
Following Franklin et al.(2005),a colored circle (diameter=2cm)was shown on a colored background. The target appeared in one of eight locations positioned in a ring equidistant from the center of the di
splay.The chromatic paration size of the target and the back-ground was equated for all trials.The target and background colors varied only in hue,with saturation and lightness kept constant(for the Y,x,y chromaticity coordinates of the stimuli and the white point,e Table1).
The study consisted of two blocks of stimuli.In the first block participants were prented with stimuli that contrasted green and blue.In the cond block they were prented with stimuli that contrasted blue and purple. As a between-subjects variable,we varied the color of the dot.For the green–blue contrast,half the participants saw a green dot(prented on either a green or blue background),and the other half saw a blue dot (prented on a green or blue background).The same between-subjects manipulation was followed for the blue –purple contrast.Participants in Group1saw a green dot(prented on a green or blue background)and a blue dot(on a blue or purple background).Participants in Group2saw the opposite combinations(blue dot on green/blue background;purple dot on blue/purple back-ground).
We counterbalanced the prentation of within-or between-category target and background in the first trial so that half of the subjects saw a target and春秋五霸称霸顺序
background from the same category in the first trial and the other half saw a target and background from two different categories.
Each trial lasted4conds and participants saw a total of12trials.Before each trial,a flashing duck’s image accompanied by a sound was prented in the center of the screen to ensure that fixation was centered.When the child fixated the duck,a gray screen appeared for250mc before the experimental trial began.
Procedure
After the infant was ated in an infant car at on the lap of their parents,the eye-tracker was calibrated using the standard Tobii calibration program to nine distinc-tive screen locations.Once accurate calibration was achieved,the time code generator and video recording were started.Next,each participant was shown an attention getter(a moving duck that makes a ringing noi)to direct the infant’s attention to the center of the screen.Once the experimenter was sure that the partic-ipant was centrally fixated,the experimental trial was initiated.This procedure was repeated until all experi-mental trials were completed.
Results
Eye-movement data were analyzed using the point of gaze coordinates to determine the time the participant took to fixate the target.A fixation was defined as a ries of data points within a30-pixel r
adius(visual angle of about1.5)for a minimum duration of100ms.Of the 82%of trials included in the analysis(18%of trials were excluded becau the eye-movement signal was lost),the target was fixated on77.5%of the time(353trials).The average number of trials on which the target was fixated was9.29out of12trials.The percentage of trials on which the target was fixated was not significantly different for within-category trials(78.5%of the trials) and between-category trials(77.2%of the trials).All infants fixated the target on a minimum of two trials per condition.
An ANOV A with Color Contrast(blue–green/blue–purple)and Color Category(between-vs.within-Cate-gory)as within-subjects variables and Group(Group1 vs.Group2)as a between-subjects variable and fixation time(average time taken to locate the target)as the dependent variable revealed a significant effect of Cat-
Table1Y,x,y chromaticity coordinates of the stimuli and of
the intertrial gray.1
Stimulus Y x y
Blue-Green
G214.170.2670.411
G114.140.2330.344
B114.110.2110.289
B214.180.2100.255
Blue-Purple
P119.470.2510.257
P219.470.2430.263
B319.470.2330.276
B419.470.2310.283
Gray40.010.330.34
1The blue,green values in the blue-green contrast,and the intertrial
gray values are the same as tho reported in Franklin et al.(2005).
Language is not necessary for color categories113
egory F (1,35)=5.37,MSE =149,979,p <.05.Between-category targets that were prented on a background color from another category (e.g.a green target pre-nted on a blue background)were detected at a rate significantly faster than within-category targets (e.g.a green target prented on another shade of green background;M between =598.0,SE =37.7,M within =753.7,SE =72.7).There was no significant main effect of Color Contrast,no effe帮扶总结
ct of Group,and no interac-tions (all F s <1)(e Figure 1).
Discussion
The results show that 8-month-old infants can perceive color hey are faster at fixating a target when it comes from a different color category than when it comes from the same category.Critically,the chro-matic paration sizes were equated across conditions so the faster detection times for targets from a different category cannot be accounted for in terms of stimulus asymmetries.
We find faster detection of between-category targets regardless of the color of the target or background (blue,green,or purple),indicating that a simple preference for color cannot lie at the heart of this effect.Although there are previous studies that have reported categorical perception of purple in 4-month-old infan孔子登东山而小鲁
ts (Franklin &Davies,2004),it is controversial whether the novelty preference method really involves perception (Roberson &Hanley,2009).The results reported in this paper are the first to unequivocally show categorical perception of purple (in contrast to blue)in pre-linguistic children.Overall,then,the results demonstrate that language is not necessary for color categories in humans.Eight-month-olds clearly do not have productive knowledge of color words,nor do they have comprehension of color terms at this age (Sandhofer &Smith,2001,with 2-year-olds;O ’Hanlon &Roberson,2007,with 3-year-olds).So,categorical responding to color in this study cannot be the result of language.Note,we are not claiming that
color categories,therefore,have to be innate.Although our findings are consistent with an innate basis for color categories,they do not rule out a learning account (e also Clifford et al .,2010).By 8months,infants have a lot of perceptual experience with the world around them.The color 伤感的歌曲
categories we find in this study could equally be a result of perceptual learning,as hard-wired innate categories.In order to distinguish the accounts,one would have to test infants from different perceptual environments on a number of different color categories.
Finally,we believe that we are at an exciting new crossroads in color categorization rearch.It appears that there are strong pre-linguistic bias in color perception,which until recently had been dismisd for the most part.This opens up a number of interesting avenues for rearch.What are the color categories that infants begin with?And how do the pre-linguistic categories interact with language input?Given that languages across the world vary in the color categories they have,does this mean that some categories will be easier to learn than others?Franklin et al .(2008a;Franklin,Drivonikou,Clifford,Kay,Regier &Davies,2008b)have shown that pre-linguistic infants begin with categorical perception of colors in the right hemisphere only but as they learn the color categories of their language,categorical perception flips to the left hemi-sphere,where it also resides in the adult.In fact,it appears that the original right hemisphere categories disappear altogether (cf.Regier &Kay,2009).Future rearch will have to specify exactly how this radical perceptual reorganization occurs.The major challenge that lies ahead for infant rearchers is to specify exactly how language input can restructure perceptual catego-ries.
Acknowledgements
This rearch was funded by the Max Planck Gell-schaft.We would like to thank members of the Categories across Language and Cognition project and members of the Language Acquisition Depar
tment at the Max Planck Institute for Psycholinguistics for critical feedback and discussion of the prent work.Thanks also to Sabine Hunnius and Gudmundur Thorgrimsson for advice regarding the analysis of data,and Angela Khadar,Margret van Beuningen,Birgit Knudn,and Daniel Puccini for their help with collecting the data.
References
Bornstein,M.H.,Kesn,W .,&Weiskopf,S.(1976).Color vision and hue categorization in young human infants.Journal of Experimental Psychology:Human Perception and Performance ,2,115–
129.
Figure 1Average time taken to locate the target for between-and within-category targets in the two color contrasts,Blue –Green and Blue –Purple.Error bars reprent standard error.
114Ozge Ozturk et al .
Catherwood,B.,Crassini,B.,&Freiberg,K.(1990).The cour of infant memory for hue.Australian Journal of Psychology, 42,277–285.
Clifford,A.,Holmes,A.,Davies,I.R.L.,&Franklin,A.(2010). Color categories affect pre-attentive color perception.Bio-logical Psychology,85,275–282.
Davidoff,J.,Davies,I.,&Roberson, D.(1999).Colour categories in a stone-age tribe.Nature,398,203–204. Davidoff,J.,Goldstein,J.,&Roberson, D.(2009).Nature versus nature:the simple contrast.Journal of Experimental Child Psychology,102,246–250.
Davies,I.R.L.,&Franklin,A.(2002).Categorical similarity may affect colour pop-out in infants after-all.British Journal of Developmental Psychology,20,185–203.
Drivonikou,G.V.,Kay,P.,Regier,T.,Ivry,R.B.,Gilbert,A.L., Franklin,A.,&Davies,I.R.L.(2007).Further evidence that Whorfian effects are stronger in the right visual field than the left.Proceedings of the N
ational Academy of Sciences,USA, 104,1097–1102.
Franklin, A.(2009).Pre-linguistic categorical perception of colour cannot be explained by colour preference:respon to Roberson and Hanley.Trends in Cognitive Sciences,13,501–502.
Franklin,A.,&Davies,I.R.L.(2004).New evidence for infant color categories.British Journal of Developmental Psychology, 22,349–377.
Franklin,A.,Drivonikou,G.V.,Bevis,L.,Davies,I.R.L.,Kay, P.,&Regier,T.(2qq邮箱官方网站
008a).Categorical perception of color is lateralized to the right hemisphere in infants,but to the left hemisphere in adults.Proceedings of the National Academy of Sciences,USA,105,3221–3225.
Franklin,A.,Drivonikou,G.V.,Clifford,A.,Kay,P.,Regier,T., &Davies,I.R.L.(2008b).Lateralization of categorical perception of color changes with color term acquisition. Proceedings of the National Academy of Sciences,USA,47, 18221–18225.
Franklin,A.,Pilling,M.,&Davies,I.R.L.(2005).The nature of infant color categorization:evidence from eye movements on a target detection task.Journal of Experimental Child Psychology,91,227–248.
Gerhardstein,P.,Renner,P.,&Rovee-Collier,C.(1999).The roles of perceptual and categorical similarity
in colour pop-out in infants.British Journal of Developmental Psychology, 17,403–420.
Gilbert,A.L.,Regier,T.,Kay,.P.,&Ivry,R.B.(2006).Whorf hypothesis is supported in the right visual field but not the
left.Proceedings of the National Academy of Sciences,USA, 103,489–494.
Huttenlocher,J.,Hedges,L.V.,&Vevea,J.L.(2000).Why do categories affect stimulus judgment?Journal of Experimental Psychology,General,129,220–241.
Kay,P.,&Kempton,W.(1984).What is the Sapir-Whorf hypothesis?American Anthropologist,86,65–79.
Lazar-Meyn,H.A.(2004).Color naming:‘grue’in the Celtic languages of British Isles.Psychological Science,15,288. O’Hanlon, C.G.,&Roberson, D.(2007).What constrains children’s learning of novel shape terms?Journal of Exper-imental Child Psychology,97,138–148.
€Ozgen,E.,&Davies,I.R.L.(1998).Turkish color terms:tests of Berlin and Kay’s theory of color universals and linguistic relativity.Linguistics,36,919–956.
Regier,T.,&Kay,P.(2009).Language,thought and color: Whorf was half right.Trends in Cognitive Sciences,13,439–446.
Roberson,D.,&Davidoff,J.(2000).The categorical perception of colors and facial expressions:the effect of verbal interfer-ence.Memory&Cognition,28,977–986.
Roberson,D.,Davies,I.,&Davidoff,J.(2000).Color categories are not universal:replications and new evidence from a stone-age culture.Journal of Experimental Psychology:Gen-eral,129,369–398.
Roberson,D.,&Hanley,J.R.(2009).Only half right:comment on Regier&Kay.Trends in Cognitive Sciences,13,500. Roberson,D.,Pak,H.S.,&Hanley,J.R.(2008).Categorical perception of colour in the left and right hemisphere is verbally mediated:evidence from Korean.Cognition,107,752–762. S考虑用英语怎么说
andhofer,C.,&Smith,L.B.(2001).Why children learn color and size words so differently:evidence from adults’learning of artificial terms.Journal of Experimental Psychology: General,130,600–620.
Thierry,G.,Athanasopoulos,P.,Wiggett,A.,Dering,B.,& Kuipers,J.(2009).Unconscious effects of language-specific terminology on pre-attentive colour perception.Proceedings of the National Academy of Sciences,USA,106,4567–4570. Winawer,J.,Witthoft,N.,Frank,M.,Wu,L.,Wade,A.,& Boroditsky,L.(2007).The Russian Blues reveal effects of language o男人补肾
n color discrimination.Proceedings of the National Academy of Sciences,USA,104,7780–7785.
Received:6June2011
Accepted:23July2012
Language is not necessary for color categories115
