The rails appear to converge in the distance but we know that the rails are parallel We know that they are the same distance apart a mile down the track as they are where we are standing so the brain says The tracks only appear to converge because t ID: 70857
Download Pdf The PPT/PDF document "Perceptual Constancy Jonathan Cohen Our ..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
[PUTFIGURE2ABOUTHERE.][PUTFIGURE3ABOUTHERE.]Thekindofperceptualconstancyexempliedinthesecases,andotherslikethem,isubiquitous,ordinary,andcentraltothewayperceptiontellsusabouttheworldinwhichwelive.Withoutthiskindofconstancy,wewouldexperiencetheworldasaJamesianblooming,buzzingconfusion—aconstantuxofcolors,shapes,andsoundswithnoapparentorganization.For,unavoidably,theperceptualsignalsincidentonourtransducersaretheresultsofnotonlythekindsofdistalindividualsthereareandpropertiestheyexemplify,butalsotheconstantlychangingdetailsofthecircumstancesunderwhichweperceive(theangleanddistancefromtheperceivedobject,thelightingconditions,theambientnoise,ourowncognitiveandperceptualhistoriesandfutures,ourexpectations,andsoon).Ifperceptionwereincapableofrepresentingtheworldasinsomewaysconstantdespitevariouschangesinourperceptualcircumstances,itwouldradicallymisrepresentthedistalworld:itwouldfailtorevealwaysinwhichtheworldisstable.Andsincethesewaysunderpinourengagementwiththatworld,thiswould(disastrously)underminethepossibilityofeffectiveactionandempiricalknowledge.However,despiteitsrecognizedubiquityandimportance,thereareseveralrespectsinwhichthephenomenonofperceptualconstancyispoorlyunder-stood.Asidefromtheindependentinterestingettingclearonthesematters,perceptualconstancyhasguredprominentlyinrecentdebatesabouttheontologyofcolorsandothersensiblequalities,knowledge,attention,mentalmodularity,thecontentsofmentalrepresentation,andtheobjectivityofourrepresentationsoftheworld.1Therefore,inthisessayI'llreviewsomeofwhatisandisnotknownaboutperceptualconstancywithaneyetodrawingconnectionswithongoingcontroversiesinthephilosophyofperceptionandelsewhere.22PerceptualConstancyasPerceptualStabilityAsbothitsnameandtheinitialexamplesusedtointroducethephenomenonabovesuggest,perceptualconstancyis,insomesenseyettobeexplained,abouttheabsenceofchange.Indeed,thetextbookcharacterizationhasitthatperceptualconstancyisnothingmoreorlessthanastabilityinperceptual 1Recentlyanumberofphilosophershavereturnedtoissuesaboutconstancyanew;forexample,seeHilbert(2005);Thompson(2006);Cohen(2008);Bradley(2008);Hateld(2009);Gert(2010);Matthen(2010);Wright(2013).AlsoseeBurge(2010),forwhomperceptualconstancyisusedasatouchstonefortheobjectivityofintentionalrepresentationquitegenerally.2Becausethereisvastlymoreresearch,bybothphilosophersandpsychologists,onperceptualconstancyinvisionthaninothermodalities(and,evenmoreparticularly,oncolorconstancy),thisentryis,regrettably,unavoidablyvisuocentricinitschoiceofexamplesandtheoriesdiscussed.Thereremainsmuchworktobedoneinthisarea.2 ofgreydotsattheintersectionsofanachromaticgrid(theHermanngridillusion,gure5),theinterpretationofapairofopposedlightnessgradientsastwoconstantlightnessregionsseparatedbyanedge(theCornsweetillusion,gure6),andtheappearanceoflightordarkbandsnexttotheboundarybetweentwodifferentlightnessgradients,evenwhenthelightnessonbothsidesoftheboundaryisthesame(Machbands,gure7).5[PUTFIGURE4ABOUTHERE.][PUTFIGURE5ABOUTHERE.][PUTFIGURE6ABOUTHERE.][PUTFIGURE7ABOUTHERE.]Perceptualcontrastisbynomeansrestrictedtotheperceptionoflight-ness/brightness;withinvisiontherearealsosimultaneouscontrasteffectsforchromaticcolor,size,spatialfrequency,orientation,motion,andspeed,interalia.Forexample,gure8illustratesaninstanceofsimultaneoussizecontrast:althoughthecentralcirclesarethesamegeometricsize,theperceptualsystemrepresentsthemasdifferentinsizebecauseofthecontrastwiththedifferentelementssurroundingthem.Moreover,inadditiontosimultaneouscontrast—contrastbetweensimultaneouslyperceiveditems,therearealsoubiquitousinstancesofsuccessivecontrast—effectsofcontrastbetweensuccessivelyperceiveditemsforeachofthesedimensions.And,ofcourse,contrastoccursinnon-visualmodalitiesaswell(althoughthereismuchlesssystematicinvestigationofcontrastoutsidevision).Thus,ingustation,wecommonlyobservethatsweetwinesstrikeusasmarkedlylesssweetwhenconsumedwithdessertitems(whichcontainmuchmoresugarthanthewines)thanontheirown.Inaudition,wendthatitismucheasiertodetectvariationsinpitch(say,whiletuningaguitarstring)bycontrastingthetargetagainstother(simultaneouslyorsuccessivelyperceived)tones.Or,again,inkinaesthesia,Gibson(1933)reportsthatafterblindfoldedsubjectsruntheirngersoveracurvedsurfaceforthreeminutes,straightedgesseemtothemtobecurvedintheoppositedirection.[PUTFIGURE8ABOUTHERE.]Ineachofthesecases,theperceptualsystemreactsdifferentlytoobjectsdependingonhowtheycontrastwithotherperceiveditems.Perceptualcontrastoccursbecauseperceptualsystemstendtoberesponsivetomagnitudedifferences,asopposedtomagnitudesthemselves.6Forourpurposes,thephenomenonofcontrastisimportantbecauseitmakesforavividdemonstra-tionoftheobservationmadeabove:contrarytothetextbookcharacterization,ourperceptualresponsestoanobject/propertyarenotconstant,butinstead 5Foradiscussionoftheroleofcontrastinmanylightnessillusions,seeAdelson(2000).6Thestandardphysiologicalexplanationofthisgeneralizationturnsonlateralinhibitionbetweenneuronscarryingperceptualinformation(e.g.,retinalganglioncells,inthecaseoflightnessperception).Lateralinhibitionresultsinthesuppressionofallbutthemoststimulated/leastinhibitedneurons;consequently,theoverallringpatternishighestincellscorrespondingtopartsofthestimuluswherethereisasteepspatial/temporalgradient—whereasmallpopulationofmostactivecellsisleftrelativelyuninhibitedbytheringoftheirneighbors.4 samesubjectsrespondedtoinstructionsto“adjustthetestpatchtolookasifitwere`cutfromthesamepieceofpaper'asthestandard,i.e.,tomatchitssurfacecolor”(1744)bymakingverysmallchromaticitychanges(suggestingthattheirperceptualsystemsinitiallyrepresentedthetestandthestandardpatchasquitesimilar).94StabilityandInstabilityItseems,then,thattherightthingtosayisnot,ornotjust,thattheperceptualsystemrespondsinaconstantorunchangingwayinthefaceofvariationsintheperceptualconditions—eitherasageneralmatteroreveninthecasesthathavebeenputforwardasparadeinstancesofperceptualconstancy.Ontheotherhand,neitherdoesitseemthattheperceptualsystemrespondsbytreatingobjectsasmerelyapproximatelythesameindifferentperceptualconditions—thesimilaritiesanddissimilaritiesthatperceptionrecognizesarenotcollapsedintoasinglescalarvaluesomewherebetweentheextremesofperfectqualitativematchandperfectqualitativemismatch.Rather,whatweshouldsayisthatperceptionrepresentsbothsomeaspectsofsimilarityandsomeaspectsofdissimilarityinitsresponsestoobjectsacrosschangingperceptualcircumstances.Moreover,weshouldrecognizethatboththerespectsofsimilarityandtherespectsofdissimilarityareinmanycasesavailabletotheperceivingsubjectforthepurposeofmakingperceptualdiscriminations.10 9Thattheperceptualsystemdisplaysthissortofbimodalbehaviorhasbeenunderstoodforalongtime;seeEvans(1948,163–164);Beck(1972,66–67)foranoverviewofsomeoftheearlierwork.Formorerecentwork(mostlyoncasesofsimultaneouscolorconstancy),seeBlackwellandBuchsbaum(1988);ValbergandLange-Malecki(1990);Arendetal.(1991);TroostanddeWeert(1991);CornelissenandBrenner(1995);B¨auml(1999).Whiletherehasbeenfarlesssystematicinvestigationofthiseffectwithrespecttocasesofsuccessivecolorconstancy,investigatorshavefoundthesamesortofbimodalpatternofresultsheretoo(Delahunt(2001,114–117);DelahuntandBrainard(2004,71–74)).10Manyphilosophersandpsychologistsworkinginthisareahavetendedtobesoimpressedbytheconstantaspectsofourperceptualresponsesthattheyhaveplayeddown,dismissed,or,morefrequently,justignoredtheinconstantaspectsofourperceptualresponsestothesamescenarios.Thus,onesometimesseesassertionstotheeffectthattheinconstantaspectsofperceptionare“unnaturalandsophisticated...[and]difculttoattain”(Smith,2002,182;cf.178).Whateverelsewethinkofsuchclaims,Isuggestthatanadequatetheoryofperceptionmustaccountforallofthewaysinwhichperceptualsystemsrespondtotheworldratherthanonlysomeofthem—whethertheseresponsesarenaturalorunnatural,naiveorsophisticated,andeasilyattainedornot.Emphasisonconstantaspectsofourperceptualresponsesattheexpenseofinconstantaspectsalsoshowsupinaprominentlineofargumentfortheviewthatcolorsareillumination-independentfeaturesofobjects(IdiscusstheseargumentscriticallyinCohen,2008).Forexample,Tye(2000,147–148),Hilbert(1987,65),andByrneandHilbert(2003,9)explicitlyappealtoconstancyreactionsincolorperceptionascaseswheretheverysamefeaturecanbeextracteddespitevariationintheambientillumination,andinferfromthisclaimthatcolor(whichtheyreasonablyassumeisindeedrepresentedbycolorperception)isitselfillumination-independent.However,ifitisreasonabletotakeconstancyreactionstoshowthatperceptionrepresentsconstantfeatures,itisnoless(andnomore)reasonabletotakeinconstancyreactionstoshowthatperceptionrepresentsinconstantfeatures.Butifcolorperceptionrepresentsbothconstantandinconstantfeatures,thereisnosoundinferencefromthepremisethatcolorisrepresented6 resultant.Thus,forexample,considercolorconstancyonceagain,sincethatistheareainwhichthemostintenseresearchoncomputationalmethodshasbeencarriedout.11Incolorperceptiontheperceptualsystembeginswithanarrayoflightintensitiesontheretinawhichisthejointproductoftwofactors—thefeaturesoftheilluminationincidentonsurfacesandthoseofthesurfacesthatreectlighttooureyes.Theleadingapproachtocomputationalcolorconstancyhasinvolvedndingmethodsofestimatingthepropertiesoftheilluminantsothatthesystemcan,asitwere,subtractoffthisfactorfromthetotalsignal(inHelmholtz'sphrase,“discountingtheilluminant”),leavinganillumination-independentcharacterizationofthereectingsurface(MaloneyandWandell,1986;Brainardetal.,1997;Brainard,1998).Crucially,sincethischaracterizationisillumination-independent,thethoughtisthatitwillbesharedbydistinctregionsofauniformsurfacethathappentobeilluminateddifferently(e.g.,theregionsofthecupingure1).Therefore,aperceptualsystemthatperformedthissortofcomputationwouldbeabletotreatsuchregionsas(inthisonerespect)perceptuallysimilar,eventhoughtheyareclearlydiscriminablydifferent.Modellershavepursuedawidevarietyofstrategiesforestimatingtheseparatecontributionstotheretinalarraymadebyilluminantsandsurfaces.Forexample,Maloney(1986);MaloneyandWandell(1986)showhowasystemwithmoreclassesofreceptorsthantherearedegreesoffreedomin(thesystem'slinearmodelsof)surfacereectionprolescanexploititsmultiplereceptoralsignalstorecoverrepresentationsofsurfaces.Otherapproachessolvetheinverseproblembyaddingasconstraintsassumptionsaboutthekindsofscenesperceptualsystemswillencounter.Thus,Buchsbaum(1980)proposesamodelthatrestsontheassumptionthatthemedianlightnessvalueinascenecorrespondstoamiddlegreysurface,andcomputesfromthisassumptionwhattheincidentilluminationwouldhavetobetoresultintheobservedintensityarray.Arelatedbutdistinctstrategyproceedsfromtheassumptionthatanchorssomepartofthevisualimage(ratherthanamean)toanextremallightnessvalue—forexample,bytreatingthelightestvisiblesurfaceaswhite(LandandMcCann,1971;Gilchristetal.,1999).Othershaveproposedestimatingilluminantsfrominformationaboutmutualreectionsinthescene(Funtetal.,1991),theboundariesofregionsknowntobespecularreections(D'ZmuraandLennie,1986;Lee,1986),andshadows(D'Zmura,1992).Stillothersproposetosolvetheinverseproblembyappealtohigher-orderscenestatistics,suchasthecorrelationbetweenrednessandluminancewithinthescene(GolzandMacLeod,2002)orthestatisticaldistributionofcolorswithinthescene(MacLeod,2003;Brainardetal.,2006).Inrecentyears,manytheoristshaveadvocated“Bayesian”probabilisticmodelsassolutionstotheilluminantestimationproblem.AccordingtoBayesians,thevisualsystemrstselectsasitsestimatethathypothesisabouttheilluminantwiththehighest 11Muchoftheworkinthistraditionisrestrictedtotheperceptionofsurfacecolors(asopposedtothecolorsoflights,volumes,lms,andsoon).Moreover,many(butnotall)ofthemodelsdependonthesimplifyingassumptionsthatsurfacesareilluminatedbyconstantorsmoothlyvarying,andexclusivelydiffuse,illumination.8 ofconstancy.13And,thoughtheseareproposalsaboutcolorconstancyinparticular,thegenerallessonstheyteachmaywellbeapplicableforothervisualandnon-visualinstancesofperceptualconstancyaswell.6IsPerceptualConstancyPerceptual?Perceptualconstancyshowsthatperceiversarenotpassivereceiversofthearrayofenergyfallingontheirreceptors—foriftheywere,theycouldnotreactinsimilarways(insomerespects),astheysometimesdo,whentherearelargedifferencesinthatarray.Somethingmoremustbegoingon.Butisthatsomethingmoreaperceptualprocess?Orisitapost-perceptualprocessthatgetsitsstartatthepointwhereperceptionends?Itisclearthat,forexample,subjectswill(undersomeexperimentalinstructions)judgethatthepennyingure3isrelevantlyalikeinshapewhenpresentedfromtwodistinctviewpoints.Butwhatisnotcleariswhetherthatjudgmentisinformedbytheoutputofperceptualsystemsbythemselves,orbytheintegrationofperceptualsystemstogetherwithcertainkindsofcognitivecorrectivefactors(e.g.,memoriesaboutthecanonicalcolors,shapes,etc.ofsimilarobjects).14Anearlyinstanceofapost-perceptual/cognitiveviewaboutperceptualconstancyistheproposal,defendedbyvonHelmholtz(1962)andHer-ing(1964),thatcolorconstancyis(atleastinpart)drivenbyourmem-ory/knowledgeaboutthecolorsoffamiliarobjects.15This“memorytheory” 13Thereareseveralfurtherpiecesofevidencethatconrmthepredictionofsuchmodelsthatperceptualsystemsmaintainrepresentationsoftheilluminationratherthansimplydiscardingthem.Perhapsthemostdirectisjustthatsubjectscan,whenasked,makematchesofambientilluminationasopposedtosurfacelightness(Katz,1935;Gilchrist,1988;Hurlbert,1989;JamesonandHurvich,1989;Zaidi,1998).Itisworthnotingthatthepossibilityofcomputingconstancywithoutderivingspecicobject/object-propertyrepresentationsundercutsthe(oft-made)claimthatobjecttrackingandreidenticationdependonrepresentingcondition-independentobjectproperties.14Obviously,one'sapproachtothislastquestionwillbeshaped,inpart,byhowoneunderstandsthecognition/perceptiondistinction.Iwon'tattempttosettlethisvexedissuehere,butwillsimplytakeforgrantedthate.g.,memoryforthecolors/shapes/sizes/etc.ofobjectsandotherapparentinstancesofconceptdeploymentfallonthecognitivesideofthedivide,andthat,e.g.,receptoraladaptationeffectsareperceptual.Whatisatstakeis(ofcourse)notthelabels,butinsteadwhatkindsofcausalexplanatoryresourcesareinvokedtoexplainobservedinstancesofperceptualconstancy.15Anevenearlierpost-perceptualviewofconstancyemergesfromLocke'sdiscussionoftheroleofjudgmentinsensation:Whenwesetbeforeoureyesaroundglobeofanyuniformcolour...itiscertainthattheIdeatherebyimprintedinourMind,isofaatCirclevariouslyshadow'd,withseveraldegreesofLightandBrightnesscomingtoourEyes.Butwehavingbyusebeenaccustomedtoperceive,whatkindofappearanceconvexBodiesarewonttomakeinus;whatalterationsaremadeinthereectionsofLight,bythedifferenceofthesensibleFiguresofBodies,theJudgmentpresently,byanhabitualcustom,alterstheAppearancesintotheirCauses:Sothatfromthat,whichtrulyisvarietyofshadoworcolour,collectingtheFigure,itmakesitpassforamarkofFigure,andframestoitselftheperceptionofaconvexFigure,andanuniformColour;whentheIdeawereceivefromthence,isonlyaPlainvariouslycolour'd,asisevidentinPainting(Locke,1975,II.ix.8).10 sizeconstancythatsparegeneralcognitiveabilities.Allthatsaid,itisalsotruethathigher-level,cognitivecues—e.g.,memoryforthecanonicalshapeandsizeofrecognizedobjects,comparisontootherperceiveditemswhoseshapeandformareestablishedindependently,thesmoothedappearanceoftexturefromadistance—enhanceshapeandsizeconstancysubstantially(forausefuloverview,seePalmer,1999,ch.5,ch.7).Cumulatively,theseresultssuggeststronglythatperceptualconstancyisneitherexclusivelyperceptualnorexclusivelycognitive.Instead,itappearsthat“the”phenomenonofperceptualconstancy,evenconsideredasconstancyforasingledimensionofasinglequalitywithinasinglemodality(e.g.,justforlightness),isaninteractioneffectproducedbyseveraldifferentmechanismsoperatingacrossdifferentspatialandtemporalscales—somepossiblymoreandsomepossiblylesscognitivethanothers,dependingonhowonechoosestomarkthecognitive/non-cognitivedistinction.16Whetheranyoneofthesemechanismscontributestoperceptualconstancyonanyparticularoccasionwilldependonthedetailsofmanyfeaturesoftheperceptualcircumstance.7ConclusionWhileIhavearguedthattheperceptualstabilitiesemphasizedbytraditionalcharacterizationsofperceptualconstancycanonlybepartofthestory,itremainstrue,indisputable,andimportantthatsomeaspectsofourperceptualresponsesarestableeventhroughchangesinperceptualcircumstancesthatresultinchangesintransducedperceptualsignals.Itisnolessindisputablethatthereareimportantlessonstobelearnedfromthephenomenonofperceptualconstancy,althoughmanyunresolvedquestionsremain.Aswehaveseen,thereisnocompletelygeneralaccountofwhichdi-mensionsofperceptualresponsemustremainxed,andwhichmayvary,acrosswhichkindsofvariationinperceptualconditions,foraperceptualepisodetocountasaninstanceofperceptualconstancy.Moreover,thereisnogeneralunderstandingoftherelationbetweenperceptualconstancyandperceptualcontrast.And,partlybecausesomuchlessisunderstoodaboutbothconstancyandcontrastinnon-visualmodalities,itissofarunclearwhat(ifany)systematiccross-modalgeneralizationsholdforeach.Finally,therangeofcomputationalstrategiesthatperceptionusestoextractstabilities,ofthemechanismsunderlyingtheirimplementation,andofthewaysthesedistinctstrategiesandmechanismsarecombinedwithoneanotherinreal-timeperceptionremainincompletelyunderstood.Notwithstandingthesesubstantialgapsinourknowledge,itseemsclearthatconstancyisanabsolutelyfundamentalaspectofperception,andtherefore 16Cf.Foster(2003),whopointstotheheterogeneityofthefactorsinoperationasareasontobeskepticalabouttheveryexistenceofcolorconstancy.12 Brainard,D.H.,Delahunt,P.B.,Freeman,W.T.,Kraft,J.M.,andXiao,B.(2006).Bayesianmodelofhumancolorconstancy.JournalofVision,6,1267–1281.Buchsbaum,G.A.(1980).Aspatialprocessormodelforobjectcolourperception.JournaloftheFranklinInstitute,310,1–26.Burge,T.(2010).OriginsofObjectivity.OxfordUniversityPress,Oxford.Byrne,A.andHilbert,D.R.,editors(1997).ReadingsonColor,Volume2:TheScienceofColor.MITPress,Cambridge,Massachusetts.Byrne,A.andHilbert,D.R.(2003).Colorrealismandcolorscience.BehavioralandBrainSciences,26(1),3–64.Cohen,J.(2008).Colorconstancyascounterfactual.AustralasianJournalofPhilosophy,86(1),61–92.Cohen,L.,Gray,F.,Meyrignac,C.,Dehaene,S.,andDegos,J.-D.(1994).Selectivedecitofvisualsizeperception:twocasesofhemimicropsia.JournalofNeurology,Neurosurgery,andPsychiatry,57,73–78.Cornelissen,F.W.andBrenner,E.(1995).Simultaneouscolourconstancyrevisited:ananalysisofviewingstrategies.VisionResearch,35,2431–2448.Craven,B.J.andFoster,D.H.(1992).Anoperationalapproachtocolourconstancy.VisionResearch,32(7),1359–1366.Dannemiller,J.L.(1989).Atestofcolorconstancyin9-and20-week-oldhumaninfantsfollowingsimulatedilluminantchanges.DevelopmentalPsychology,25,171–184.Dannemiller,J.L.(1993).Rankorderingsofphotoreceptorphotoncatchesfromnaturalobjectsarenearlyilluminant-invariant.VisionResearch,33,131–140.Dannemiller,J.L.andHanko,S.A.(1987).Atestofcolorconstancyin4-month-oldhumaninfants.JournalofExperimentalChildPsychology,44,255–267.Delahunt,P.B.(2001).AnEvaluationofColorConstancyAcrossIlluminationandMutualReectionChanges.Ph.D.thesis,UniversityofCalifornia,SantaBarbara,SantaBarbara.Delahunt,P.B.andBrainard,D.H.(2004).Doeshumancolorconstancyincorporatethestatisticalregularityofnaturaldaylight?JournalofVision,4,57–81.Douglas,R.H.,Eva,J.,andGuttridge,N.(1988).Sizeconstancyingoldsh(carassiusauratus).BehaviouralBrainResearch,30,37–42.D'Zmura,M.(1992).Colorconstancy:surfacecolorfromchangingillumination.JournaloftheOpticalSocietyofAmericaA,9,490–493.14 vonHelmholtz,H.(1962).Helmholtz'sTreatiseonPhysiologicalOptics.Dover,NewYork.Originallypublishedin1867.Hering,E.(1964).OutlinesofaTheoryoftheLightSense.HarvardUniversityPress,Cambridge,Massachusetts.Originallypulishedin1878;translatedbyLeoM.HurvichandDorotheaJameson.Hilbert,D.R.(1987).ColorandColorPerception:AStudyinAnthropocentricRealism.CSLI,Stanford.Hilbert,D.R.(2005).Colorconstancyandthecomplexityofcolor.PhilosophicalTopics,33,141–158.Hunter,F.,Biver,S.,andFuqua,P.(2007).LightScienceandMagic:AnIntroductiontoPhotographicLighting.FocalPress,Oxford,thirdeditionedition.Hurlbert,A.C.(1989).Cuestothecoloroftheilluminant.Ph.D.thesis,MassachusettsInstituteofTechnology,Cambridge,Massachusetts.Hurlbert,A.C.andLing,Y.(2005).Ifit'sabanana,itmustbeyellow:Theroleofmemorycolorsincolorconstancy.JournalofVision,5(8),787.Hurvich,L.M.(1981).ColorVision.SinauerAssociates,Sunderland,Massachusetts.Ingle,D.(1998).Perceptualconstanciesinlowervertebrates.InV.WalshandJ.J.Kulikowski,editors,PerceptualConstancy:WhyThingsLookasTheyDo,pages173–191.CambridgeUniversityPress.Jameson,D.andHurvich,L.M.(1989).Essayconcerningcolorconstancy.AnnualReviewofPsychology,40,1–22.ReprintedinByrneandHilbert(1997),177–198.Katz,D.(1911).DieErscheinungsweisenderFarbenundihreBeeinussungdurchdieIndividueleErfahrung.Barth,Leipzig.Katz,D.(1935).TheworldofColor.Kegan,Paul,TrenchTruber&Co.,London.translatedbyR.B.MacLeodandC.W.Fox.Kraft,J.M.andBrainard,D.H.(1999).Mechanismsofcolorconstancyundernearlynaturalviewing.ProceedingsoftheNationalAcademyofSciences,U.S.A.,96,307–312.Land,E.H.andMcCann,J.J.(1971).Lightnessandretinextheory.JournaloftheOpticalSocietyofAmerica,61.Lee,H.C.(1986).Methodforcomputingthescene-illuminantchromaticityfromspecularhighlights.JournaloftheOpticalSocietyofAmericaA,3,1694–1699.16 Troost,J.M.anddeWeert,C.M.M.(1991).Namingversusmatchingincolorconstancy.Perception&Psychophysics,50,591–602.Tye,M.(2000).Consciousness,Color,andContent.MITPress,Cambridge,Massachusetts.Valberg,A.andLange-Malecki,B.(1990).`Colourconstancy'inMondrianpat-terns:Apartialcancellationofphysicalchromaticityshiftsbysimultaneouscontrast.VisionResearch,30(3),371–380.Whittle,P.(2003).Contrastcolours.InR.MausfeldandD.Heyer,editors,ColourPerception:MindandthePhysicalWorld,pages115–138.OxfordUniversityPress,NewYork.Wright,W.(2013).Colorconstancyreconsidered.ActaAnalytica,28(4),435–455.Wyszecki,G.andStiles,W.S.(1982).ColorScience:ConceptsandMethods,QuantitativeDataandFormulae.Wiley,NewYork.Secondedition.Zaidi,Q.(1998).Identicationofilluminanyandobjectcolors:Heuristic-basedalgorithms.JournaloftheOpticalSocietyofAmericaA,15,1767–1776.Zaidi,Q.(1999).Colorandbrightnessinduction:fromMachbandstothree-dimensionalcongurations.InK.R.GegenfurtnerandL.T.Sharpe,editors,ColorVision:FromGenestoPerception,pages317–343.CambridgeUniversityPress,Cambridge.Zaidi,Q.(2001).Colorconstancyinaroughworld.ColorResearchandApplication,26,S192–S200.18 Figure4:Aninstanceofsimultaneouslightnesscontrast:thecentralpatchesarequalitativelyidentical,butperceptionrepresentsthemdifferentlybecauseofthecontrastwithsurroundingitems. Figure5:TheHermanngridillusion.21 Figure8:TheEbbinghausillusionisaninstanceofperceptualsimultaneoussizecontrast.23