Explaining Sonority Projection Effects
|
|
- Angel Holt
- 5 years ago
- Views:
Transcription
1 ExplainingSonorityProjectionEffects RobertDaland a,brucehayes a,marcgarellek a, JamesWhite a,andreadavis b,ingridnorrmann c a DepartmentofLinguistics,UCLA; b DepartmentofLinguistics,Universityof Arizona; c DepartmentofSpanish&Portugese,UCLA Abstract Sonorityprojectionreferstobehavioraldistinctionsspeakersmakebetween unattestedphonologicalsequencesonthebasisofsonority.forexample,among onsetclusters,thewell formednessrelation[bn]>[lb]isobservedinspeech perception,speechproduction,andnonwordacceptability(albright,inpreparation; Berent,Steriade,Lenertz,&Vaknin,2007;Davidson2006,2007).Webeginby replicatingthesonorityprojectioneffectsinanonwordacceptabilitystudy.thenwe evaluatetheextenttowhichsonorityprojectionispredictedbyexisting computationalmodelsofphonotactics(coleman&pierrehumbert1997;hayes& Wilson2008;etalia).Weshowthatamodelbasedonlyonlexicalstatisticscan explainsonorityprojectioninenglishwithoutapre existingsonoritysequencing principle.todothis,amodelmustpossess(i)afeaturalsystemsupporting sonority basedgeneralizationsand(ii)acontextrepresentationincluding syllabificationorequivalentinformation.
2 1 Introduction TheSonoritySequencingPrinciple(SSP)isthecross linguisticgeneralization thatthemostwell formedsyllablesarecharacterizedbyasonorityrisethroughout theonsettothenucleus,andafallfromthenucleusthroughoutthecoda(sievers 1881;Jesperson1904;Hooper1976;Steriade1982;Selkirk1984).Forexample,the onset[bn]ismorewell formedthantheonset[lb]becausetheformercontainsa smallsonorityrise(obstruenttonasal)andthelattercontainsalargesonorityfall (liquidtoobstruent).afundamentalgoalofphonologicaltheoryistounderstand broadgeneralizationslikethessp. Acompleteunderstandinginvolvesanswerstothefollowingquestions.Isthe SSPsynchronicallyactiveinspeakers grammars,oradiachronicbyproductof physicalfactorsgoverningspeechperceptionandproduction,orsomecombination ofboth?ifthesspisapartofspeaker sgrammars,isitinnate,orlearned,orsome combinationofboth?andiflearned,fromwhat?howisknowledgeofthessptobe formallycharacterized?andhowisitdeployedduringspeechproductionand speechperception? WhatisknownatpresentisthattheSSPissynchronicallyactiveinspeakers grammars(althoughthisdoesnotruleoutdiachronicfactorsinaddition).themost recentformofevidencetosupportthisconclusionistheexistenceofsonority projectioneffects responsestonovelstimulithatvarydependingontheextentof thesonorityviolation.inparticular,stronglyssp violatingclustersaremorelikely tobeproducedandperceivedwithvowelepenthesis,e.g.[lb] [ləb]ismorelikely than[bn] [bən](davidson2006,2007;berentetal.2007). 1 Thesearetermed projectioneffects(inthesenseofbaker,1979)becausetheoffendingclustersare systematicallyandequallyabsentfromspeaker sinput,andyetspeakersappearto differentiatesomeclustersaslesswell formedthanothers. WhatisnotknownishowtheSSPcomestobeapartofspeakers grammars; infact,thisiscontroversialintheliterature,andaprincipalgoalofthispaperisto contributetothedebate.thispaperfocusesonthefollowingquestions: 1. Whatpropertiesmustanyphonotacticmodelhaveinordertopredict sonorityprojectioneffects? 2. HowdospeakerscometopossessknowledgeoftheSSP?Ifitisnotinnate, uponwhatkindofexperienceisitbased? 1 LisaDavidson(p.c.)andDoncaSteriade(p.c.)suggestthat sonority maynotbeatruephonological primitive,butratherconsistsofahostofphoneticfactors.forexample,davidson(2010)proposesan accountofher(2006)productioneffectsintermsofarticulatory(mis)coordination.weimaginethat perceptualepenthesis(dupouxetal.,1999;berentetal.,2007)canbeaccountedforsimilarlytothe productionaccountoutlinedinwilson&davidson(inpress):thephonotacticprobabilityofthe epentheticparseissomuchhigherthantheintendedparsethatthelistenersimplymisconstruesthe tokenascontainingthephonotacticallyacceptablesequence.theseareallimportantquestionsthat deservefurtherresearch,butforthepurposesofthepresentpaper,whatiscrucialisthatbehavioral variationcanbepredictedforclustersspeakershavenoexperiencewith,andthisvariationgenerally linesupwithtraditionaldefinitionsofsonority.webelievetheexperimentalandmodelingresults presentedherearejustascompellingwhethersonorityisinterpretedasatruephonological primitiveorasacovertermforavarietyofphoneticproperties. 2
3 3 Wedeferthequestionofmodelpropertiesforthemoment.Asforhowsomething likethesspcomestobeknown,wedistinguishlexicalisttheories inwhichthessp isprojectedfromthelexicon fromuniversalisttheories.thelexicalisthypothesisis consistentwithabodyofworkdemonstratingotherphonotacticgeneralizations thatareprojectedfromthelexicon(e.g.frisch&zawaydeh2001;hay, Pierrehumbert,&Beckman2003).Theuniversalisthypothesiscomesintwoforms. ThemostdirectformistopositthattheSSPisinnate.Theinnatistapproachis commoncurrencyinlinguistictheory,althoughformanyspecificaspectsof grammaritisdifficulttofindatheoristwhoadvocatesaninnateexplanation.the otheruniversalapproachthathasbeenproposedisthatthesspisphonetically grounded learnedfromexperienceinproducingandcomprehendingspeech,and universalbecause certainbasicconditionsgoverningspeechperceptionand productionarenecessarilysharedbyalllanguages,experiencedbyallspeakersand implicitlyknownbyall (Hayes&Steriade2004).Thesepossibilitiesare schematizedintable1. Hypothesis Projectedfrom lexicalist lexicon innatist UniversalGrammar phoneticallygrounded speechperception/productionexperience Table1.ExplanationsfortheSonoritySequencingPrinciple. Atpresent,thelexicalisthypothesisisthedominantexplanationfor phonotacticknowledge:evidencefromavarietyofmethodologiesconvergesonthe conclusionthatthelexiconisanimportantseatofphonotacticgeneralizations.for example,thestrengthofgradientocp Placeeffectsinnonwordacceptability judgementsispredictablefromlexicaltypestatistics(frisch&zawaydeh2001;see alsocoleman&pierrehumbert1997).asanotherexample,nonwordrepetition accuracyisbelievedtoindexphonotacticproficiency(coady&evans2008)andis stronglypredictedinchildrenbytheirvocabularysize,asconsistentwiththeview thatthephonotacticgrammarisprojectedfromthelexicon(edwards,beckman,& Munson2004;seealsoHayetal.2003).Thequestionisnotwhetherthelexiconisa sourceforphonotacticgeneralizations,butwhetheritisthesolesource. Toshowthatthereissomeothersource,itwouldbenecessarytofinda particularphonotacticgeneralizationanddemonstratethatitcannotbeprojected fromthelexicon.justsuchanargumenthasbeenmadeforthessp,instrongeror weakerforms,byseveralauthors.theargumentgoesasfollows:lexicalistmodels assignwell formednessonthebasisoflexicalfrequency.unattestedclustershavea frequencyof0.therefore,lexicalistmodelsshouldclassifyallunattestedclustersas ungrammatical,andcrucially,equallyungrammatical.inotherwords,theyshould failtopickoutsome(stronglyssp violating)clustersasmoreungrammaticalthan other(weaklyssp violating)ones.sonorityprojectioneffectsoccur,andsolexicalist modelsareunabletoaccountforthem.thisargumentismadeexplicitlybyren, Gao,&Morgan(2010,abstract):
4 ThesensitivitytotheSSPcanhardlybeaccountedforbylexicalstatisticfactors becausemandarinsyllableshavenoonsetclustersandnocodaconsonantswiththe exceptionof[n]and[ŋ],soallthestimuliinourexperimentswerealientothem. Thesensitivitycannotbeexplainedbyphoneticconfusionseither,becausesimilar sensitivityhasalsobeenfoundinreadingtasks(berent2009).thetwofindings shedlighton basicquestionsofgenerativegrammarbyindicatingthatthessp,as auniversalprinciple,mayconstituteapartofhumanlinguisticknowledge. Berentetal.(2007)arguesimilarly.Theyshowthataparticularlexicalmodel,the Vitevitch&Luce(2004)PhonotacticProbabilityCalculator,hasnostatistically significantcorrelationwiththeresultsoftheirsonorityprojectionstudy.they conclude(pp ): OurfindingsdemonstratethatEnglishspeakersmanifestsonority related preferencesdespitethelackoflexicalevidence,eitherdirect(i.e.,theexistenceof therelevantonsetsintheenglishlexicon)orindirect(thestatisticalco occurrence ofsegmentsinenglishwords). Experimentalresultsalongtheselines(seealsoBerent,Lennertz,Jun,Moreno,& Smolensky2008;Albright2009)constituteintriguingevidenceforthehypothesis thatthesspisnotprojectedfromthelexicon.inthetheoreticaltaxonomyoftable 1,theymaybetakenassupportingeithertheinnateorthephonetically grounded hypotheses.however,asberentetal.(2007,p.624)pointout,theargumentrelies onthefailureofparticularstatisticalmodelstopredicttheresult,andthereisno guaranteethatothermodelswillsimilarlyfail.itisthispointthatwepursuehere. Lexicalistmodelsassignwell formednessonthebasisoflexicalfrequency. Thekeyquestion,however,isfrequencyofwhat?Segmentsareanaturalstarting pointforphonologicalanalysis,andthereisabundantevidencethattheyrepresent apsychologicallyimportantlevelofrepresentation.however,segmentsarenotthe onlyrepresentationavailableforanalysis,andfromaphonologicalstandpoint,they arenotnecessarilyeventhebestone.analternative,notedbyberentetal.,isto considermodelsthatemployfeatures,i.e.acousticand/orarticulatoryproperties thataresharedbynaturalclassesofsegments. Ifamodelislimitedtocountingsegments,thenitistruethat,forexample, theonsets[tl]and[lt]areequallyunattested.however,fromafeaturalperspective, theonset[tl]receivesmorelexicalsupportthantheonset[lt].therearemany attestedonsetclustersthatarefeaturallysimilarto[tl],e.g.,[pl],[kl],[tr],[tw],[sl]. Incontrast,therearenoattestedonsetclustersthatareequallysimilarto[lt].A lexicalistmodelthatgeneralizesacrossmultiplefeaturallevelsofabstractionmight distinguishdegreesofwell formednessbetweentheseclustersonthisbasis,even thoughthesegmentalfrequencyofeachclusteris0.indeed,atleasttwolexicalist modelshavebeenproposedthatdogeneralizeonthebasisoffeatures:thehayes& Wilson(2008)PhonotacticLearnerandAlbright s(2009)featuralbigrammodel. However,thereisasyetnopublishedworkassessingfeature basedcomputational modelsforsonorityprojection(thoughseealbright,inpreparation). 4
5 Thus,thegoalofthispaperistotestavarietyofpublishedcomputational modelsofphonotacticsonthiscaseofsonorityprojectioneffects.thevalueofa directcomparisononthesamestimuliisthatwemaygainclearinsightonwhat modelpropertiesareresponsibleforsuccessandfailureonthisparticular phonotacticdomain whichmayinformourunderstandingastowhatcollectionof propertiesthenextgenerationofmodelsshouldhave. Inordertoassessthepredictiveutilityofamodel,itisnecessarytohave humanbehavioraldataforthemodeltoexplain.inthiscase,thefocusissonority projectioneffects,andsowebeginthepaperbycollectingnonwordacceptability ratingswithnonwordswhoseonsetclustersvaryintheextentofssp violation.asa matterofgeneralinterest,wealsoincludednonwordswithfrequentlyattested onsets(like[bl])andmarginallyattestedonsets(like[bw]). Withnonwordacceptabilitydatainhand,thepaperwillproceedtothe modelingstage.weimplementanumberofcomputationalmodelsofphonotactics describedintheliterature,specifically: classicalbigrammodel(jurafsky&martin2009) featuralbigrammodel(albright2009) syllabicparser(coleman&pierrehumbert1997) PhonotacticLearner(Hayes&Wilson2008) PhonotacticProbabilityCalculator(Vitevitch&Luce2004) GeneralizedNeighborhoodModel(Bailey&Hahn2001) Theadequacyofthemodelsisassessedbylinearregressionagainstthenonword acceptabilitydata. Toanticipatebriefly,wefindthatsomepublishedmodelsexhibit considerablesuccessinpredictingsonorityprojectioneffects.thekeyfindingsare discussedindepthlater;fornowtheymaybesummarizedasfollows:alexicalist modelcananddoespredictsonorityprojectioneffectsifithas(a)thecapacityto representsonority,and(b)arepresentationofphonologicalcontextthatisrich enoughtorepresenttheexpectedsonoritylevel.inotherwords,lexicalistmodels exhibitsonorityprojectionwhentheyareequippedwiththerepresentationsand architecturenecessarytodoso.thisworksupportsalexicalistaccountofthessp. Thepaperisstructuredasfollows.InSection2,wedescribetwoexperiments collectingnonwordacceptabilityjudgementsfromthemechanicalturk,anonline laborforum.insection3,wegivebriefdescriptionsofthecomputationalmodels testedhere,nonecontainingthesspasabias.insection4,wedescribetheresults ofcomputationalmodeling;eachmodelwastrainedonthesameenglishlexiconand thenassessedonitsabilitytopredicthumanjudgementsforunattestedclusters varyingintheirdegreeofssp violation.insection5,wediscusstheempirical findingsofthisworkandtheirtheoreticalimplications. 2 SonorityProjectioninAcceptabilityRatings Inthissectionwedescribeanonwordacceptabilityjudgementexperiment withnonwordsthatweredesignedtovaryinthelevelofsspviolation.webegin withasummaryofsonorityscales,followedbyabriefdescriptionofthemechanical 5
6 Turk.Thenonwordsarethendescribed,followedbytheacceptabilityexperiment. Theexperimenthadtwoconditions:inthefirstcondition,participantsratedforms onalikertscale;inthesecondcondition,participantscomparedtwoformsand selectedthebetterchoice.thesectionconcludeswithatheoreticaldiscussionof sonorityprojection,andamethodologicalcomparisonofthesensitivitiesoflikert ratingversuscomparison. 2.1 Sonorityscale Todeterminewhetherparticipantsexhibitsonorityprojection(andwhether phonotacticmodelscanexplainit),itisnecessarytohaveanindependentmeasure ofsonority.anumberofsonorityscaleshavebeenproposedintheliterature(e.g. Steriade1982;Selkirk1984;Clements1988;Parker2002),generally 2 havingthe followingproperties: eachsegmenthasasonorityvaluerepresentedbyaninteger segmentsaregroupedintosonorityclassessharingthesamesonorityvalue theminimallysonorousclasshasasonorityvalueof0 sonorityincrementsby1betweenclasses TheriseofasequenceXYisdefinedassonority(Y) sonority(x).thenthesspcanbe formalizedbydefiningathresholdforacceptablerises,e.g. onsetsmusthavearise ofatleast2 impliesthat[bl]isacceptablesolongassonority(l) sonority(b) 2.This typeofformulationhasprovenremarkablysuccessfulindelimitingonset inventoriescross linguistically(seereferencesabove),andiswhatjustifiesthe assignmentofparticularintegervaluestoparticularsegmentclasses. Scalesproposedintheliteraturedifferchieflyingranularity.Elaborated scalessuchasselkirk(1982)distinguishobstruentvoicingandmanner,vowel height,andrhoticity.weselectedthecoarse grainedscaleinclements(1988): obstruents(0)<<nasals(1)<<liquids(2)<<glides(3)<<vowels(4).thisscale makesonlyuncontroversialdistinctionsrepresentingtheconsensusofthe phonologicalcommunity TheMechanicalTurk TheMechanicalTurk( providedbyamazon.com.itwasusedbecauseitoffersaquickandeasywayto conductwordacceptabilityandsimilarstudies thetotaltimetocompletedata collectionwasabout1hourforeachratingmethod,withacostof$3/participant+ 10%commissionforAmazon.com,whichcomparesfavorablywith2 3weeksand $5 $10/participantfortheequivalentlaboratorystudy.Qualityismaintainedinthe MechanicalTurkbytheapprove/rejectoption,andtheapprovalthreshold. Researchersmayrejecttheworkofanyindividualworker(andrefusetopay);they mayalsopre screenbyselectingworkerswhoseapprovalrateisaboveathreshold; therecommendedapprovalthresholdis95%.asaresult,workersandthewebsite arebothdirectlyincentivizedtoensureanoverallhighqualityofwork. 2 Selkirk sscalestartsat0.5forvoicelessstops.theremainderofthescalehastheseproperties. 3 Theanalysesreportedin 2.5werealsocomputedwiththerichlyelaboratedsonorityscaleof Selkirk(1982).Thegeneraleffectswerethesame:attestedness,andsonorityintheunattesteds. 6
7 7 AllparticipantswererecruitedfromtheMechanicalTurkusingthe recommended95%approvalthreshold.participantsgaveonlineconsentand completedabrieflanguagebackgroundsurveysurveyingenglishproficiency, dialect,andotherlanguagesspoken.resultswereretainedfromparticipants reporting high Englishproficiency(Likertrating:n=2;comparisonrating:n=12)or native proficiency(likertrating:n=17;comparisonrating:n=36).theresearch teaminspectednon nativeresultsandfoundthattheyexhibitedthesame qualitativepatternsasnatives,i.e.attesteds>>marginals>>unattesteds(seenext sectionfordetails).participantsreporting intermediate proficiencywerepaid,but theirresultswerediscardedandreplaced.one(native)participantwasexcluded fromthelikertconditionforratingover80%oftheitemsas Stimuli Thestimuliconsistedof96stress initialccvcvcnonwords,generatedby concatenatingacconsetwithavcvctail(e.g.,pr + eebid=preebid).therewere 48onsetsand6tails.Thus,eachonsetwaspairedwith2tails,andeachtailwas pairedwith16onsets(48*2=96=16*6).eighteenclustersthatneveroccuras Englishonsets(unattesteds)werechosentovaryacrossthewholerangeofsonority (e.g.[tl]involvesalargesonorityrisewhereas[rg]involvesalargesonorityfall). Alsoincludedwere18clustersthatoccurfrequentlyasEnglishonsets(attesteds) and12clustersthatoccuronlyrarelyorinloanwords(marginals,e.g.[gw]in Gwendolyn,[ʃl]inschlep).Attestedandmarginalclusterswereincludedtovalidate thetask(participantsshouldexhibitthepreferenceattesteds>>marginals>> unattesteds)andtoincreaseecologicalvaliditybyprovidingatleastsometestitems thatareplausibleenglishwords.sixvcvctailswereselectedtoyieldalmostno lexicalneighborsandtoavoidviolatinganymajorphonotacticconstraintsof English. ThelistofonsetsandtailsisshowninTable2,withsonorityvaluesin parenthesesfortheunattesteds: Attested Onsets Marginal Onsets UnattestedOnsets (sonority) Tails tw,tr,sw, shr,pr,pl, kw,kr,kl, gr,gl,fr, fl,dr,br, bl,sn,sm gw,shl, vw,shw, shn,shm, vl,bw, dw,fw, vr,thw pw(3),zr(3),mr(2), tl(2),dn(1),km(1), fn(1),ml(1),nl(1), dg(0),pk(0),lm( 1), ln( 1),rl( 1),lt( 2), rn( 2),rd( 3),rg( 3) ottiff[ ɑtɪf] eebid[ ibɪd] ossip[ ɑsɪp] eppid[ ɛpɪd] eegiff[ iɡɪf] ezzig[ ɛzɪg] Table2.Listofonsetsandtails. Thestimuliwerecounterbalancedinanumberofways.Eachtailappeared approximatelythesamenumberoftimesforeachsonorityrange,sothat,e.g. ottiff wouldnotappearmoreoftenwithrelativelywell formedunattestedonsets.thecooccurrenceoftailswithonsetphonemeswascounterbalanced;forexample, ottiff
8 wouldnotappearmoreoftenwithanonsetcontaining/p/.repeatedsegments(e.g. dgeegiff)wouldbeindependentlydispreferredbytheocp,sotheseitemswere avoidedasmuchaspossible.onsetsandtailswerecombinedsoastoensurethatno nonwordhadmorethanonelexicalneighbor(neighbor=anexistingword obtainablefromthenonwordbyinserting,deleting,orsubstitutingonephoneme). Tocontrolforembeddedwords,weavoidednonwordswhoseC 1 C 2 VC 3 partsformed arealwordwithattestedandmarginalonsets;forunattestedonsets,nonwords whoseembeddedc 2 VC 3 partsformedarealwordweredistributedacrossthe sonorityrange.allofthenon wordswerepresentedinenglishorthographyinall capitalletters.toensurethatthestimuliwerephonologicallyunambiguous,they werepresentedtofivenaïveenglishspeakers;allnon wordswerepronouncedas intended,suggestingthatthespellingsarelargelyunambiguous. 2.4 DesignandProcedure Aftergivingconsentandfillingoutthelanguagebackgroundsurvey, participantscompleted6practiceitems,andthenperformedthemaintask.all itemswerepresentedonasinglepage,withradiobuttonsfortheanswers. FortheLikertratingcondition,participantswereinstructedthattheywould beratingpotentialnewwordsofenglish,thattheywouldseemultiplepotential wordsandthattheyshouldratethembasedonhowlikelyitwasthatthewords couldbecomenewwordsofenglishinthe21 st century.thepracticeitemswere STALLOP,SKEPPICK,THRISHAL,SHMERNAL,LBOBBIB,SHTHOKKITH,andwere intendedtoexposeparticipantstoawiderangeofwell formedness.eachitem consistedofasinglenonword,andtheresponseswere 1 (unlikely)to 6 (likely). Eachparticipantratedall96items;fourdifferentrandomizationswereusedto controlorder of presentationeffects. Forthecomparisonratingcondition,participantswereinstructedtochoose thenonwordthatseemedmorelikeatypicalenglishword.thepracticeitemswere STALLOPvs.THMEFFLE,LBOBBIBvs.PRIFFIN,THRISHALvs.FTEMMICK,SKEPPICK vs.mzibbus,shmernalvs.dwiffert,andshthokkithvs.thpellop.each uniquenonwordpairwaspresentedtoexactly1participant,andeachparticipant wasassignedalistof95items,sotherewere48participants(96*95/2pairs=4560 pairs=48participants*95pairs/participant).nonwordposition(leftorright)was counterbalanced,andparticipantlistswereconstrainedtonotcontainanynonword morethantwice. 2.5 Results Allregressionsweredoneusingthelmerfunctionfromthelme4package (Bates&Sarkar2006)inR(RDevelopmentCoreTeam2006).Asacheckonthe task,theentiredatasetwasregressed,usingtheorderedfactorofattestedness (unattested<<marginal<<attested)asthefixedeffect.linearregressionswith ratingasthedependentvariablewereusedforthelikertconditionbecausethe responsevariableisscalar;onset,tail,andparticipantwereincludedasrandom effects.logisticregressionwasusedforthehead to headcondition,witheachtrial splitintotwoobservationscorrespondingtoeachofthenonwords;thedependent variableindicatedwhetherthenonwordwaschosen(notethatthissplittingwas 8
9 necessarybecauseunlikenormallogisticregression,thetwochoiceschangefrom trialtotrial 4 );onset,tail,andparticipantpreferenceforleft/rightresponsewere includedasrandomfactors.todeterminewhethersonorityinfluencedlistener judgements,thedatasetswererestrictedtotrialscontainingonlyunattested clusters.sonoritywasusedasthefixedeffect,butotherwisetheregressionswere thesameasabove(linearforlikert,logisticforhead to head,samerandomeffects). Attestednesswasasignificantpredictorofwell formednessinboth conditions.marginals(nonwordscontainingmarginalonsets)wererated significantlyhigherthanunattesteds(likert:t= 7.4,p<1e 4; 5 head to head:z= 6.4, p<1e 9)andattestedswereratedsignificantlyhigherthanmarginals(Likert:t=10.1, p<1e 4;head to head:z=6.2,p<1e 9).Tovisuallyinspectwhetherthereisasonority effectintheunattestedclusters,fig.1plotsthisregression sunattestedcluster randominterceptsagainstsonority.theplotshowsthatsonorityisanexcellent predictorofthevarianceremaininginunattestestedclusters. 4 Thisanalysisseparatesatrialintotwoobservations,oneforeachnonwordofthepair.The statisticalmodelassumestheseobservationsareindependent,whichisfalsebecauseifonewordis chosentheothermustnotbe.thiscodingchoicereducesthepowerofthemethod,andhencecanbe regardedasconservative.onealternativemethodwasspecificallydesignedforsuchcircumstances, andisknownas alternative specificconditionlogisticregression (McFadden,1974)becausethe choicebetweenthealternativesisconditionedonpropertiesthatarespecifictoeachalternative,e.g. sonorityoftheonsetcluster.however,theredoesnotyetappeartobeanimplementationthatallows forrandomeffects.anotheralternativewouldhavebeentomodelthechoicebetweenleftandright, andtoincludeboththeleftandrightnonwords propertiesasfixedorrandomeffects;however,this ignoresthereal worldstructureoftheproblemsinceitassignsnumericallydistinctcoefficientsfor itemsthatoccurontheleftversustheright.suchamodelisincorrectbecause,forexample,[bl]isthe sameonsetwhetheritoccursontheleftortheright.inshort,thecurrentlyavailablestatistical methodsallhaveminorflaws.theanalysismethodweselectedisimplemented,interpretablebecause thereisonlyonesetofcoefficients,andconservativebecauseignoringperfectanti correlations withinapairshouldreducepower. 5 Degrees of freedom(df)areunreportedbecausedfisill definedforlinearmixed effectsmodels (Bates&Sarkar,2006;seealsoBates commentsathttps://stat.ethz.ch/pipermail/r help/2006 May/ html).Accordingly,thosep valueswerecalculatedwithmontecarlosamplingusing pvals.fncinthelanguagerpackage(baayen.davidson,&bates2008). 9
10 cluster coefficient pw mr tl zr fn rl ml dn km rn nl lmln pk rd dg lt rg sonority Fig.1.Unattestedcluster(random)coefficientsplottedagainstsonority. Thesonorityregressionconfirmedthatsonoritywasasignificantpredictorofwell formednessforunattestedonsets(likert:t=6.2,p<1e 4;head to head:z=7.4,p<1e 12). Formodelingpurposes,itwillproveusefultohaveacanonical acceptability score assignedtoeachnonword.wedefinethisastheproportionofcomparisons trialsinwhichanonwordwasselectedasbetterthanitscompetitor;thisvalueis usedinpreferencetorandominterceptsfromaregressionmodelforconceptual transparencyandforgreatercomparabilitytopreviousstudies(coleman& Pierrehumbert,1997),thoughthetwoarehighlycorrelated. 2.6 Discussion Theresultsofthenonwordacceptabilityexperimentdemonstratedseveral importantpatterns.first,boththelikertratingandcomparisonconditions exhibitedtheexpectedeffectofattestedness,withthewell formednessscale attested>>marginal>>unattested;thisshowsthatparticipantsrecruitedfromthe MechanicalTurkexhibitthesamecoarsebehavioraslaboratoryparticipantsin previousstudies.second,sonoritywasasignificantpredictorofacceptabilityfor unattestedonsets;thisresultisconsistentwiththehypothesisthatspeakershave internalizedknowledgeofthessp,butishardtootherwiseexplain.finally,as 10
11 11 discussedbelow,whilebothconditionsexhibitedthesamepatternofsignificant differences,thecomparisonconditionwasmoresensitivefortheunattesteditems ofinterest.thesepointsarediscussedinturn Inclusionofnon nativespeakers Theresultsofthepresentstudyshowthatatacoarselevel,participants recruitedviatheinternetexhibitthesamebehaviorasparticipantsrecruited throughsubjectpoolsorcampusflyer.internetrecruitmentarguablyrepresentsa moreecologicallyvalidsampleofenglishspeakersthanastudywithmonolinguals, becauseanon trivialpercentageofspeakerssorecruitedareearlyorlatebilinguals. Thisisapotentialcauseforconcern,asevenhighlyproficientlatebilingualsmay exhibitsubtledifferencesinjudgementfromnativespeakers(coppieters,1987). Notehoweverthatinthisexperimenttheresearchgoalisnottoisolatecompetence oftheidealizedmonolingualenglishspeaker hearer,butrathertodetermine whethersonorityprojectionoccursinenglishnonwordacceptabilityjudgements. TheotherlanguagesourparticipantsreportspeakingincludeDutch,French,Hindi, Mandarin,Marathi,Punjabi,andafewothers;theselanguagesaregenerallyequally ormorerestrictivethanenglishwithrespecttoonsetsonorityprofiles.thus,the sonority violatingclustersinthepresentstudyareequallynoveltoallparticipants Sonorityprojection Thestatisticalmodelingresultsshowedthatsonorityisasignificant predictorofparticipants well formednessratingsforunattestedclusters.themost naturalexplanationforthisfindingisthatparticipantshaveinternalizedknowledge ofthessp.howeveritisworthconsideringthealternativehypothesisthatthese resultsreflectsomesortoforthotacticknowledge. Theorthotacticaccountcanexplainthecoarsedifferenceinratingbetween attested,marginal,andunattestedonsets,butitfailstoexplaintheeffectofinterest: sonorityprojectioninunattesteds.thefrequencyofallunattestedonsetclustersis (bydefinition)0,sotheyarecruciallynotdifferentiatedbyfrequency.moreover,the visualstructureoftheenglishalphabetdoesnotreflectitsphonology,e.g.rismore visuallysimilartopthantol,butmorephonologicallysimilartol.theprincipled relationshipbetweensonorityandwell formednesscannotbeexplainedbyenglish orthotactics Sensitivityatthebottomofthescale ThepatternofsignificantdifferenceswasthesameacrosstheLikertrating conditionandthehead to headcomparisoncondition.however,thecomparison taskwasevidentlymoresensitivefortheitemsofinterest,theunattestedonsets. Onebitofevidenceforthisclaimisthatthezstatisticforthehead to headsonority comparisonisgreaterthanthetstatisticforthelikertcomparison,witha correspondingdifferenceinsignificance(likert:t=6.2,p<1e 4;head to head:z=7.4, p<1e 12).ThepointcanbeappreciatedmoreclearlyinFig.2,whichplotsrawonset averagesfromthecomparisonconditionagainstthelikertcondition.(theonset average forthecomparisonconditionisdefinedastheproportionofcompetitions wonbynonwordscontainingthecluster.)
12 raw comparison average attested marginal unattested raw Likert average Figure2.ScatterplotofcomparisonscoresagainstLikertratings Thecomparisonaveragedifferentiatestheunattestedonsetclustersmuchbetter thantherawlikertaveragedoes.presumablythisoccursbecausethetarget, unattesteditemsareconcentratedatthebottomendofthewell formedness spectrum,yieldingnear floorratingsforallofthem.thisfactsuggeststhefollowing methodologicalpoint:innonwordacceptabilitystudies,head to headcomparisonis preferabletolikertratingwheneverthestimuliofinterestareconcentratedatone endofthewell formednessscale,owingtoceiling/flooreffectsinlikertratings. SimilarconclusionswerereachedinCoetzee s(2004)unpublisheddissertation,and bykager&pater(underrevision);anddifferentbutrelatedpointsareaddressedin Kawahara(ms);wementionthismethodologicalpointhereinthehopeofaverting unnecessaryreplication of effortinthefuture. 3 ComputationalModelsofPhonotactics Havingestablishedthehumanbehavioraldataofinterestonsonority projection,weturnnowtothequestionofexplainingthejudgementsthathumans make.inthissection,wegiveabriefoverviewofsixcomputationalmodelsthathave beenproposedtoexplainnonwordwell formednessjudgements.the trainingdata thatwillbeusedhereisthelexicondescribedin 4.1,thoughinprinciplethese modelscantrainonanylexicon. 3.1 Bigram 12
13 Classicalbigrammodelsassignprobabilitiescompositionally:theprobability ofthewholeistheproductoftheprobabilityofthesub partsandthewaytheyare combined.inclassicalbigrammodels,thesub partsarebigrams,andthewhole wordprobabilityistheproductofthetransitionalprobabilities.forexample, cat canbeexpressedas#kæt#(where#areboundarysymbols);itsbigramsare#k,kæ, æt,andt#(fordetailedexpositionseejurafsky&martin2009;forarecentlinguistic studyseegoldsmith&riggle,toappear).fromthese,theprobabilityof[kæt]is calculatedasin(1): (1)Calculatingprobabilityof[kæt]inaclassicalbigrammodel # k æ t # p(# k) * p(k æ) * p(æ t) * p(t #) Thetransitionalprobabilitiesareestimatedfromtrainingdatausingrelative frequency;forexamplep(k æ)isestimatedbydividingthefrequencyof[kæ]by thefrequencyof[k].thismodelistermedalexicalmodelbecausebigram frequenciesarecalculatedbytheirtypefrequencyinthetrainingdata inthe presentstudy,anenglishlexicon. Inthenaturallanguageprocessingliterature,wherebigramandrelated modelsareheavilyemployed,itisconsideredbestpracticetosmooththe transitionalprobabilities(manning&schutze1999;jurafsky&martin2009). Smoothingassignsamodestamountofprobabilitytounseenitems,soastoavoid assigningzeroprobabilitytoitemsthathappentobeabsentfromthetrainingset. 6 Inourimplementationoftheclassicalbigrammodel,weusedGood Turing smoothing(gale&sampson1995). 3.2 Coleman&Pierrehumbert(1997) Coleman&Pierrehumbert s(1997)modelissimilartoabigrammodelin thatitassignswordprobabilitiescompositionallybymultiplyingtheprobabilitiesof sub parts.however,itdiffersinthesub parts:wordsareparsednotintobigrams butintoaphonologically motivatedhierarchyconsistingofsyllables,onsets,and rhymes.separatecountsaremaintainedforstressedvs.unstressed,initialvs.noninitial,andfinalvs.non finalsyllables,foratotalofeightonsetandeightrhyme distributions.justasinthebigrammodel,thecountsareestimatedfromthelexicon. Ourimplementationdiffersslightlyfromwhatisdescribedintheoriginalpaper, becauseourtraininglexiconincludesmorethanjustthebinaryfeetonwhichthe originalimplementationisbased.therefore,ratherthanparsingintobinaryfeet, ourimplementationusesthedistributionoverallattestedstresspatterns. 6 SmoothingisappropriateforZipfiandistributions,inwhichnoveleventscontinuetobeobserved forarbitrarilylargesamples.segmentalbigramsinenglishfollowsuchadistribution(daland& Pierrehumbert,2011). 13
14 Hereisanillustrationofhowthemodelcomputestheprobabilityoftheword agenda: (2)Calculatingprobabilityof[әˈd ʒɛndә]inthecoleman/pierrehumbertmodel ω σ [ s,+i, f] σ [+s i f] σ [ s i+f] O [s,+i, f] R [ s,+i, f] O [+s i f] R [+s i f] O [ s i+f] R [ s i+f] ә d ʒ ɛ n d ә P(agenda)=theproductof P(ω=σ [ s] σ [+s] σ [ s] ) probabilityofamedial stressedtrisyllable P(O [ s,+i, f] = ) probabilitythatonsetofinitialstresslesssyllableisnull P(R [ s,+i, f] =[ә]) probabilitythatrhymeofinitialstresslesssyllableis[ә] P(O [+s, i, f] =[d ʒ]) probabilitythatonsetofmedialstressedsyllableis[d ʒ] P(R [+s, i, f] =[ɛn]) probabilitythatrhymeofmedialsyllableis[ɛn] P(O [ s, i,+f] =[d]) probabilitythatonsetoffinalstresslesssyllableis[d] P(R [ s, i,+f] =ә]) probabilitythatrhymeoffinalstresslesssyllableis[ә] Aswiththebigrammodel,Good Turingsmoothingwasused;aseparatesmooth wasdoneforeachonsetandrhymedistribution. 3.3 Albright(2009) Thefeaturalbigrammodelisbroadlysimilartotheclassicalbigrammodel describedabove.itdiffersinhowthetransitionprobabilitiesarecalculated.rather thantreatingeachsegmentasadistinct,uniquetype,itdeploysphonological features,sothateachsegmentmaybecharacterizedbyanyofthenaturalclassesto whichitbelongs.forexample,thesegment[b]canbeconstruedas[+labial], [+consonantal],[+labial,+consonantal],[+labial, nasal],andsoon.thelikelihoodof abigramiscalculatedfromits best natural classfeaturaldescription,accordingto (3): (3) FormulaforselectingfeaturalbigramsintheAlbright(2009)model 7 l(xy)=max A,B p(ab)*p(x A)*p(y B) 7 Notethatthisformulaassignsalikelihooddistributionratherthanatrueprobability distribution,becausethevaluesdonotsumto1.thisiswhyl(xy)isusedinsteadofpr(xy). 14
15 where AandBrepresentnaturalclassestowhichxandyrespectivelybelong p(ab)isthetypefrequencyofnaturalclassbigramabinthetraininglexicon p(x A)=1/ A (1overthenumberofsegmentsinA) p(y B)=1/ B Theoverallrationaleofthemodelisthatawordcontainingpopulousnaturalclass bigramsarelikelytobeparticularlywell formed,especiallywhenthesegmentsthat instantiatethenaturalclassformalargeshareofthatclass spopulation.therestof thecomputationworksanalogouslytotheclassicalbigrammodel. Weranourownimplementationofthemodel,meanttofunctionidentically toalbright sbutfacilitatetheuseofourownfeaturesetandtrainingdata. 3.4 Hayes&Wilson(2008) TheHayes&Wilson(2008)PhonotacticLearnerisaconstraint based learningmodel.constraintsarestatedinthephonologicalvocabularymade standardbychomsky&halle(1968)andsubsequentwork.forexample,the constraint*#[+sonorant, syllabic][+consonantal]militatesagainstword initial[lb] clustersandsimilarssp violatingforms.justaswiththefeaturalbigrammodel,the featuresallowthehayes/wilsonmodeltomakegeneralizationsoversegments, includinggeneralizationsbasedonsonority. Toassesswell formedness,thehayes/wilsonmodelemploysthemaximum entropyvariant(dellapietra,dellapietra,&lafferty1997;goldwater&johnson 2003)ofHarmonicGrammar(Legendre,Miyata,&Smolensky1990;Smolensky& Legendre2006;Pater2009;Potts,Pater,Jesney,Bhatt,&Becker2010).Each constraintc i hasanonnegativeweightw i.awordxisevaluatedbyfindingits constraintviolationcountsc i (x),multiplyingeachviolationcountbythe correspondingweight,andtakingthesum.thenegativeofthissumisknownasthe harmonyofx,andthelikelihoodofxistheexponentialofitsharmony.toensure thisisatrueprobabilitydistribution,likelihoodisdividedbyanormalization constantthatguaranteesitsumsto1: (4)ProbabilityofwordxinHayes/Wilsonmodel Pr(x)=e harmony(x) /Z harmony(x)= i w i C i (x) Z= x Ω *harmony(x) (Ω * isthesetofallpossiblewords) Theconstraintsdeployedinthegrammararefoundbyasearchalgorithm thatattemptstoidentifyconstraintsthatbestexplainthetraininglexicon.the algorithmprivilegesconstraintsthatarebrief(fewfeaturematrices),accurate(low expected/observedviolations),andgeneral(coveringlargenumbersofpossible forms).thenumberofconstraintsthatthealgorithmincludesinagrammarcanbe setbytheuser.wecausedthemodeltoterminateat400constraints,andexplored theeffectofconstraintnumberbyconsideringsub grammarsincludingonlythe 15
16 first100,150,200,250,300,and350constraints.weranthealgorithmusingthe softwarepostedatwww.linguistics.ucla.edu/people/hayes/phonotactics/. 3.5 Vitevitch&Luce(2004) TheVitevitch&Luce(2004)PhonotacticProbabilityCalculatoriswidely usedinpsycholinguisticresearch.itresemblesseveralmodelsdescribedalreadyin thatitassignsascoretoawordbydividingitintopartsandcombiningtheir probabilities.themodelissimilartoabigrammodelinthatitusesbigramsaswell theirsimplercousinunigrams. Themodelemploysapositionalrepresentation,basedonleft to rightserial positionofsegments.separatecountsaremaintainedforeachposition.forexample, theprobabilityof[b]asthefirstsegmentofawordisbasedonwhatfractionofall word initialsegmentsare[b];theprobabilityof[b]asthefourthsegmentofaword isbasedonwhatfractionofallfourth positionsegments(inwordswithatleastfour segments)are[b],andsoon.inthebigramversion,analogouscomputationsare carriedoutonbigrams. Themodelusesaweightingsystemevidentlyintendedtoprovidea compromisebetweentypeandtokenfrequency.itweightsunigramsandbigrams bythelogoftheirtokenfrequencies,whicharerescaledbythetotallogfrequency weighttogetunigramandbigramprobabilities. Ourimplementationreflectsthestandardpracticethathasevolvedin experimentalworkmakinguseofthismodel:aunigramscoreiscalculatedasthe sumoftheunigramprobabilities,andabigramscoreiscalculatedanalogously(see thesub partprobabilitiesarenotmutuallyexclusive,summinginthiswayimplies thatnonwordscorescannotbeinterpretedasprobabilities. 3.6 Bailey&Hahn(2001) TheGeneralizedNeighborhoodModel(Bailey&Hahn2001,hereafter BH2001)isanexemplarmodelinwhichthewell formednessscoreofanitemis determineddirectlyfromthelexicon,bythesumofitssimilaritiestoexistingwords. Thisisincontrasttotheothermodelsdiscussedabove,inwhichagrammarisfirst projectedfromthelexicon,andthenwell formednessisevaluatedbythegrammar. Thesimilarityofanoncewordω i toanexistingwordω j iscalculatedfrom thestring editdistanced ij.string editdistanceiscalculatedfromthenumberof insertions,deletions,andsubstitutionsneedtochangeω i toω j.asintheoriginal paper,weusedaninsertionanddeletioncostof0.7,andtheproportionofshared naturalclasses(frisch,broe,&pierrehumbert1997)asthesubstitutioncost.the similarityofω i toω j isgivenbyexp( D d ij ),wheredisascalingfactor.similarlyto Vitevitch&Luce(2004),logtokenfrequencyweightingwasincluded,althoughthis modeladoptsamorecomplicatedquadraticweightingscheme.thetotalscorefora formiscalculatedbysummingsimilarities;wedifferslightlyfrombh2001by summingovertheentirelexicon,anoperationthatwasnotcomputationallyfeasible in2001.thefullformulaisgivenin(5): 16
17 (5)NonwordacceptabilityscoreinBaileyandHahn(2001) score i = j (Alogf j2 +Blogf j +C) exp( D d ij ) where logf j =log(tokenfrequencyofω j +2) D=5.5; 8 d ij isthestring editdistance Owingtodifferencesbetweenthetraininglexiconandtestitemshereversusin BH2001,weconsideredseveralsetsoffreeparameters: Label A B C basis fig estimatedfromfigureinbh2001 oral oraltaskinbh2001(bailey,pc) writ writtentaskinbh2001(bailey,pc) lin nofrequencyweighting 3.7 Summary Eachofthemodelsdiscussedintheprecedingsectionisalearningmodel.It istrainedonalexiconofalanguageandassignsscalarwell formednessvalues basedonagrammarprojectedfromthelexicon(orbasedonthelexiconitself).a summaryofmodels propertiesisgiventable3. model output basedon from bigram probability segmentalbigrams lexicon grammar syllabicparser probability syllabicconstituents lexicon grammar featuralbigram likelihood featuralbigrams lexicon grammar PhonotacticLearner likelihood featuralconstraints lexicon grammar PhonotacticProbability scalar positionalbigrams lexicon grammar Calculator Generalized NeighborhoodModel scalar string editdistance lexicon Table3.Summaryofmodelproperties Forthemodelswhoseoutputshaveaprobabilisticinterpretation(thefirstfourin Table3),theoutputswerelog transformed.thiswaspartlydoneforcomparison withwell formednessratingssincecoleman&pierrehumbert(1997)foundthat nonwordlog likelihoodswerelinearlyrelatedtohumanacceptabilityjudgements;it wasalsosimpler,astheunderlyingcomputationsareactuallyperformedinthelog domain.nosuchlogtransformwasappliedtothescoresforvitevitch&luce(2004) andbailey&hahn(2001),bothforgreatercomparabilitytoexistingstudies,and sincethesevaluesarescalarsthatdonothaveaprobabilisticinterpretation. 8 TheapproximatevalueofDwaskindlysharedbyToddBailey,asweretheA/B/Cvaluesforthe oralandwritset.thelinsettingwasrecommendedbyareviewer. 17
18 4 Modelingnonwordacceptabilityjudgements 4.1 TrainingonanEnglishlexicon Themodelsdescribedinsection3arelexicalistlearningmodels,meaning thatthewell formednessscorestheyassignaredirectlyorindirectlyprojectedfrom thelexicon.forafaircomparison,itisnecessarytotrainallmodelsonthesame lexicon.thissubsectiondescribesthetraininglexicon. Ourgoalwastocreatearepresentativedictionaryofthewordslikelytobe knowntotheparticipants.weusedthecmupronouncingdictionary ( bin/cmudict)transcriptions,selectingonly thosewordsthathaveafrequencyofatleast1inthecelexwordformdatabase (Baayen,Piepenbrock,&Gulikers1995).Fromthisset,weremovedcompounds, residualinflectedforms,andformscreatedbyhighlytransparentprocessesof morphologicalderivation,yieldingasetof18,612wordsinphonemictranscription. Twoversionsofthetrainingsetwereused.Inone,syllabificationwas lexicallyspecifiedbyannotatingconsonantsasbelongingtothecodaornot. 9 Inthe other,codapositionwasnotdistinguishedinthelexicalform.thephonemesofthe trainingsetweresupplementedbyafeaturechart.(thelexiconandfeaturescharts areavailablefromthecorrespondingauthoruponrequest.)thefeaturechartwas usedbythefeaturalbigrammodel,thephonotacticlearner,andthegeneralized NeighborhoodModel.Forthesefeaturalmodels,annotatedcodaconsonantswere featurallyidenticaltotheironsetcousinsexcepttheyweremarked[+rhyme];for thenon featuralmodels,annotatedcodaconsonantswerecountedasdistinct atomicsymbols,i.e.onset[b]wasjustasdistinctfromcoda[b]asitwasfrom[l]. 4.2 Method Eachmodelwastrainedonthetraininglexiconandwasthentestedonthe setofnonwordsusedinexperiments1and2.trainingconsistedofestimating modelparametersasdescribedin Testingconsistedofassigningawellformednessvaluetoeachnonwordstimulus. 4.3 Results Togetabroadoverviewofmodelperformance,wecalculatedforeachmodel thecorrelationofitswell formednessscorewiththeempiricallyderivedwellformednessscorefromtheexperimentalhead to headdatain 2.5.These correlationsareshownintable4.thefocusofthispaperisonsonorityprojection, sowhatisofmostinterestisamodel sabilitytopredictvariationamongthesubset ofunattesteditems.however,forcompletenessandgeneralintellectualinterest,we alsocomputedcorrelationsfortheattestedandmarginalsubsets,aswellasthe entiredataset.thesearereportedintable4. syllabification no syllabification 9 Syllabificationwasassignedusingthemaximumonsetprinciple(Selkirk,1982):medialconsonant sequenceswereparsedwiththelongestonsetthatoccursword initially.giventhattheseare learningmodels,itisreasonabletowonderhowthehiddenstructureofsyllabificationislearned.we leavethisissueforfutureresearch. 18
19 model attested marginal unattested overall attested marginal unattested overall albright bigram coleman gnm.fig gnm.oral gnm.writ gnm.lin hw hw hw hw hw hw hw vl.uni vl.bi Table4.CorrelationsofmodelratingswithExperiment2scores.Key:albright=featural bigram;bigram=classicalbigram;coleman=coleman&pierrehumbert(1997);gnm.set= GeneralizedNeighborhoodModelwithparameterset;HW[n]=PhonotacticLearnerwithn constraints;vl.uni/bi=phonotacticprobabilitycalculator,unigramandbigrammodels, respectively. Good modelcorrelationsarebolded(seetextfordetails). Modelsarearrangedintherows,withmembersofthesame family adjacenttoone another.thecolumnsaredividedintotwogroups,withsyllabifiedtraining/testing ontheleft,andunsyllabifiedinputontheright.thecolumnsrepresentthesubsetof thedatabeingregressed,andtheentriesineachcellrepresentthecorrelation.for example,thetopleftmostnumericalcellindicatesthatthesyllabifiedfeatural bigrammodelratingshada27%correlationwithhumanjudgementsonthe nonwordswithattestedonsets.thesecorrelationsprovideaconvenientmacrolevelsummaryofthemodels predictions. Tosimplifyfurtheranalysis,fromeachfamilyweselecteda best model whichinourjudgementrepresentedthebestornear bestperformanceofthat family.forexample,hw100(syllabified)wasselectedfromthehayes/wilson familybecauseithadthe(near )highestattested,unattested,andoverall correlations.theintentionistofocusinonthemostinformativecomparisons thoseinwhichwecanbesurerelativelypoorperformanceisnotsimplytheresult ofanunfortunatechoiceofparametersforamodel.putanotherway,itiseasierto understand6dataseriesthan30,andsincemanyofthedataseriesareparametric 10The overall scoreincludesvariationwithinandacrosssubsets.forexample,thebigrammodel doesnotdowellatdistinguishingunattesteditemsfromoneanother(low unattested correlation) butitdoesdistinguishunattestedsasaclassfromattestedsasaclass(high overall correlation). 19
20 variants,itisbettertojustfocusonthe6 best ones.fig.3a fplotsmodelnonword predictionsagainstthecomparisonjudgementsfromtheexperiment. albright bigram attested marginal unattested coleman gnm.lin hw vl.bi Figure3Modelnonwordpredictionsversushumanjudgements.x axis:well formednessscore asdeterminedbyhead to headcomparisondata(z transformed);y axis:modelscore(ztransformed).eachpointrepresentsanonword.each best modelisplottedinadifferent pane:albright=featuralbigrammodel;bigram=classicalbigrammodel;coleman=coleman &Pierrehumbert(1997);gnm.lin=GeneralizedNeighborhoodModelwithnofrequency weighting;hw100=phonotacticlearner;vl.bi=phonotacticprobabilitycalculator 4.4 Discussion Severalpointsemergedfromtheresultsofthemodelingstudy.Themost significantfindingforlinguistictheoryasawholewasthatthereexistlexicalmodels thatexplainsonorityprojection,i.e.predictsonority relatedvariationinhuman behaviorforunattestedphonologicalsequences.amongthemodelstestedhere,the modelswhichweremosteffectiveatmodelingsonorityprojectioneffectswerethe PhonotacticLearner(Hayes&Wilson,2008)andAlbright s(2009)featuralbigram 20
21 model;wewillarguethatsonorityprojectionowestoafeaturalrepresentationof sonorityandarichenoughrepresentationofcontexttotracktheexpectedsonority profile.anotherfindingofinterestwasthatnocurrentmodelexcelledacrossthe well formednessspectrum,i.e.themodelsthatwerebestonunattestedonsetswere notbestonattestedonsets.thesepointsarediscussedindetailbelow Sonorityprojectionispossiblefromthelexiconalone Themosttheoreticallysignificantfindingofthepresentstudyisthatsonority projectionisachievedbyanumberofpublishedlexicalistphonotacticmodels.this findingdirectlycontravenespreviousclaimsintheliterature,suchasthefollowing passagefromberentetal.(2007,pp ): Our findings demonstrate that English speakers manifest sonority related preferences despite the lack of lexical evidence, either direct(i.e., the existence of therelevantonsetsintheenglishlexicon)orindirect(thestatisticalco occurrence ofsegmentsinenglishwords). Berentandcolleaguesfoundsonorityprojectioneffectsinperception,whichisa substantialcontributiontothefield,becauseitunequivocallydemonstratesthatthe SSPisapartofspeakers synchronicknowledge.whatwedisagreewithistheclaim thatthereisnolexicalevidenceforthesonority basedpreferences.itistruethat, forexample,lbandtlareequallyunattestedasenglishonsets,buttherearemany onsetclustersthatarefeaturallysimilartotlwhereastherearenonethatareso featurallysimilartolb.alexicalistmodelthatisequippedtomakegeneralizations onthebasisoffeaturesshouldinprinciplebeabletoexplainsonorityprojection, andwehaveshownherethatthisisexactlywhathappens. Fromthebeliefthatthereisnolexicalsupportforsonorityprojection,Berent andcolleaguesdrawtheinferencethatlistenersmustpossesssomeformof universalknowledge,whetheritarisesfrom inherentpreferencesofthelanguage system (p.593)orknowledgethatis inducedfromphoneticexperience (p.625). Ourresultsshowthattheinferenceofuniversal,non lexicalphonotacticknowledge doesnotfollowasalogicalnecessity althoughitmaystillbecorrect.inshort,the abilityofalexicalmodeltoexplainsonorityprojectioneffectbearsonfoundational issuesofourfield,becauseitrefutesapowerfulargumentfortheexistenceof universalphonotacticknowledge Modelpropertiesneededforsonorityprojection Beyondthesheerfactthatsonorityprojectionoccurs,itisofinteresttoknow whysomemodelsexhibititandothersdonot.wewillarguethatwhatisneededis theabilitytocapitalizeontworepresentationalproperties:asufficientlyrich representationofphonologicalcontext(e.g.syllabification),andasufficientlyrich representationofsonorityitself(e.g.features).thesyllabifiedfeaturalbigram modelexhibitssonorityprojection.howevertheunsyllabifiedfeaturalbigram modelsdoesnot,soremovingsyllabificationinhibitssonorityprojection.similarly, thesyllabifiedclassicalbigrammodeldoesnot,soremovingfeaturalgeneralization alsoinhibitssonorityprojection.thus,sonorityprojectionrequiresbothproperties. 21
22 Phonologicalcontext.Toexpresssonorityrestrictions,amodelneedstobe abletodistinguishcontextsthatconstrainthesonorityprofile,e.g.itshouldbe risingword initially.modelstrainedonsyllabifieddatacandothis,sincetheyare toldwhetheraconsonantsequenceisparsedasanonset,arhyme,orasa heterosyllabiccluster.withthisinformation,suchmodelsareinapositionto inductivelytrackthesonorityprofilescharacteristicofthesethreecontexts,and characterizewell formednessoftheseconfigurationswhentheyarefilledby particularsegments. Weillustrateusingthespecificexampleofthefeaturalbigrammodel.A clusterlike[lt]isperfectlyacceptableinenglishwhenitisnotanonsetcluster,e.g. halt,elton.whensyllabificationismadeavailabletothemodel,itshouldbeableto distinguishtheunacceptableonsetclusterfromtheacceptablecodaand heterosyllabicclusters.indeed,whenitistrainedwithsyllabifieddata,thealbright modelachievesacorrelationofr=.55withhumanjudgementsforunattestedonsets. Thecorrelationdropstor=.18whenthesamemodelistrainedonunsyllabifieddata (wewillshowlaterthatthislevelofcorrelationarisesmerelyfrommodelingtails). Sincetheonlydifferencebetweenthesetwocasesisthepresenceofsyllabification, itfollowsthatthecontextualinformationrepresentedbysyllabificationcausedthe difference.inotherwords,syllabificationprovidesasufficientlyrichrepresentation ofthecontextastoallowalbright smodeltorepresenttheexpectedsonority contour. Itispossibleforamodeltosucceedwithoutanexplicitrepresentationof syllabification.inparticular,withasufficientlylargenumberofconstraintsthe PhonotacticLearnerachievesroughlyequivalentperformancesonsyllabifiedor unsyllabifieddata.webelievethisowestothefactthatthephonotacticlearner allowstrigramconstraints.englishtrigramsprovidealevelofphonologicalcontext thatismorespecificthansyllabification;forexampleatrigrammodelcanuse structuraldescriptionsoftheform[xyc]and[xy#]inplaceofx coda y coda,aswas doneinchomskyandhalle(1968).atthesametime,scalingupamodeltotrigrams hasitsowncostsintermsofsparsenessofdataandcomputationalcomplexity (Jurafsky&Martin2009). Insummary,whatamodelneedsissomerepresentationofphonological contextthatissufficientlyrichastotracktheexpectedsonoritycontour(seekager &Pater,underrevision,foranotherstudyconcludingthatphonotacticmodelsmust representsyllabification).explicitsyllabificationisanespeciallysimpleand effectivemeansofdoingthis,asevidentfromthefactthatnearlyeverymodeldoes betteronnearlyeverysubsetofthedatawhenithasaccesstosyllabification. Phonologicalfeatures.Inadditiontophonologicalcontext,amodelneedsa systemofphonologicalfeatures.therationaleforthisclaimisverysimple:inorder tomakegeneralizationsonthebasisofsonority,amodelmustbeabletomake generalizations,anditmusthaveanexplicitrepresentationofsegments sonority. Phonologicalfeaturesperformbothofthesefunctions.Featuresrepresentinherent generalizations,becausethepresenceorabsenceofafeaturerepresentsan underlyingacousticorarticulatorypropertysharedbyanaturalclassofsegments. 22
CHAPTER 1 CLUSTER PHONOTACTICS AND THE SONORITY SEQUENCING PRINCIPLE. organized into well-formed sequences according to universal principles of
CHAPTER 1 CLUSTER PHONOTACTICS AND THE SONORITY SEQUENCING PRINCIPLE 1.1 Introduction Languages of the world differ in their syllable phonotactics. Some languages are extremely restrictive and only allow
More informationPEOPLE S KNOWLEDGE OF PHONOLOGICAL UNIVERSALS: EVIDENCE FROM FRICATIVES AND STOPS. A dissertation presented. Tracy Jordan Lennertz
1 PEOPLE S KNOWLEDGE OF PHONOLOGICAL UNIVERSALS: EVIDENCE FROM FRICATIVES AND STOPS A dissertation presented by Tracy Jordan Lennertz to The Department of Psychology In partial fulfillment of the requirements
More informationNeural evidence for a single lexicogrammatical processing system. Jennifer Hughes
Neural evidence for a single lexicogrammatical processing system Jennifer Hughes j.j.hughes@lancaster.ac.uk Background Approaches to collocation Background Association measures Background EEG, ERPs, and
More informationSonority as a Primitive: Evidence from Phonological Inventories Ivy Hauser University of North Carolina
Sonority as a Primitive: Evidence from Phonological Inventories Ivy Hauser (ihauser@live.unc.edu, www.unc.edu/~ihauser/) University of North Carolina at Chapel Hill West Coast Conference on Formal Linguistics,
More informationA Framework for Advanced Video Traces: Evaluating Visual Quality for Video Transmission Over Lossy Networks
Hindawi Publishing Corporation EURASIP Journal on Applied Signal Processing Volume, Article ID 3, Pages DOI.55/ASP//3 A Framework for Advanced Video Traces: Evaluating Visual Quality for Video Transmission
More informationNKPZ.E Motor Controllers, Float- and Pressure-operated. Motor Controllers, Float- and Pressure-operated
NKPZ.E174189 Pressure-operated Page Bottom Pressure-operated See General Information for Pressure-operated IFM ELECTRONIC GMBH FRIEDRICHSTRASSE 1 45128 ESSEN, GERMANY E174189 Trademark and/or Tradename:
More informationOptimum Frame Synchronization for Preamble-less Packet Transmission of Turbo Codes
! Optimum Frame Synchronization for Preamble-less Packet Transmission of Turbo Codes Jian Sun and Matthew C. Valenti Wireless Communications Research Laboratory Lane Dept. of Comp. Sci. & Elect. Eng. West
More informationSonority as a Primitive: Evidence from Phonological Inventories
Sonority as a Primitive: Evidence from Phonological Inventories 1. Introduction Ivy Hauser University of North Carolina at Chapel Hill The nature of sonority remains a controversial subject in both phonology
More informationStudent Guide to the Publication Manual of the American Psychological Association Vol. 5
APA Short Guide 1 Student Guide to the Publication Manual of the American Psychological Association Vol. 5 1. Use margins of 1 inch (2.54 cm) on all sides and a line length of no more than 6.5 in (16.51
More informationUniversity of Bristol - Explore Bristol Research. Peer reviewed version. Link to published version (if available): /ISCAS.2005.
Wang, D., Canagarajah, CN., & Bull, DR. (2005). S frame design for multiple description video coding. In IEEE International Symposium on Circuits and Systems (ISCAS) Kobe, Japan (Vol. 3, pp. 19 - ). Institute
More informationAnalysis of Packet Loss for Compressed Video: Does Burst-Length Matter?
Analysis of Packet Loss for Compressed Video: Does Burst-Length Matter? Yi J. Liang 1, John G. Apostolopoulos, Bernd Girod 1 Mobile and Media Systems Laboratory HP Laboratories Palo Alto HPL-22-331 November
More informationThe Bias-Variance Tradeoff
CS 2750: Machine Learning The Bias-Variance Tradeoff Prof. Adriana Kovashka University of Pittsburgh January 13, 2016 Plan for Today More Matlab Measuring performance The bias-variance trade-off Matlab
More informationProblem Points Score USE YOUR TIME WISELY USE CLOSEST DF AVAILABLE IN TABLE SHOW YOUR WORK TO RECEIVE PARTIAL CREDIT
Stat 514 EXAM I Stat 514 Name (6 pts) Problem Points Score 1 32 2 30 3 32 USE YOUR TIME WISELY USE CLOSEST DF AVAILABLE IN TABLE SHOW YOUR WORK TO RECEIVE PARTIAL CREDIT WRITE LEGIBLY. ANYTHING UNREADABLE
More informationStatistical Consulting Topics. RCBD with a covariate
Statistical Consulting Topics RCBD with a covariate Goal: to determine the optimal level of feed additive to maximize the average daily gain of steers. VARIABLES Y = Average Daily Gain of steers for 160
More informationMixed Effects Models Yan Wang, Bristol-Myers Squibb, Wallingford, CT
PharmaSUG 2016 - Paper PO06 Mixed Effects Models Yan Wang, Bristol-Myers Squibb, Wallingford, CT ABSTRACT The MIXED procedure has been commonly used at the Bristol-Myers Squibb Company for quality of life
More informationPredictability of Music Descriptor Time Series and its Application to Cover Song Detection
Predictability of Music Descriptor Time Series and its Application to Cover Song Detection Joan Serrà, Holger Kantz, Xavier Serra and Ralph G. Andrzejak Abstract Intuitively, music has both predictable
More informationIntroduction to Natural Language Processing This week & next week: Classification Sentiment Lexicons
Introduction to Natural Language Processing This week & next week: Classification Sentiment Lexicons Center for Games and Playable Media http://games.soe.ucsc.edu Kendall review of HW 2 Next two weeks
More informationAir Navigation Safety Assessment Methodology for ATS
Air Navigation Safety Assessment Methodology for ATS Cualquier copia impresa o en soporte informático, total o parcial de este documento se considera como copia no controlada y siempre debe ser contrastada
More informationA Discriminative Approach to Topic-based Citation Recommendation
A Discriminative Approach to Topic-based Citation Recommendation Jie Tang and Jing Zhang Department of Computer Science and Technology, Tsinghua University, Beijing, 100084. China jietang@tsinghua.edu.cn,zhangjing@keg.cs.tsinghua.edu.cn
More informationAcoustic and musical foundations of the speech/song illusion
Acoustic and musical foundations of the speech/song illusion Adam Tierney, *1 Aniruddh Patel #2, Mara Breen^3 * Department of Psychological Sciences, Birkbeck, University of London, United Kingdom # Department
More informationNavigating on Handheld Displays: Dynamic versus Static Peephole Navigation
Navigating on Handheld Displays: Dynamic versus Static Peephole Navigation SUMIT MEHRA, PETER WERKHOVEN, and MARCEL WORRING University of Amsterdam Handheld displays leave little space for the visualization
More informationA sensitive period for musical training: contributions of age of onset and cognitive abilities
Ann. N.Y. Acad. Sci. ISSN 0077-8923 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES Issue: The Neurosciences and Music IV: Learning and Memory A sensitive period for musical training: contributions of age of
More informationFor the SIA. Applications of Propagation Delay & Skew tool. Introduction. Theory of Operation. Propagation Delay & Skew Tool
For the SIA Applications of Propagation Delay & Skew tool Determine signal propagation delay time Detect skewing between channels on rising or falling edges Create histograms of different edge relationships
More informationBasic Natural Language Processing
Basic Natural Language Processing Why NLP? Understanding Intent Search Engines Question Answering Azure QnA, Bots, Watson Digital Assistants Cortana, Siri, Alexa Translation Systems Azure Language Translation,
More informationBCM Calibration for E Abstract
Jefferson Lab E8-4 Analysis Report July 22 BCM Calibration for E8-4 Patricia Solvignon Jefferson Lab E-mail solvigno@jlab.org Abstract In this note, the calibration procedure of the Beam Current Monitors
More informationRCBD with Sampling Pooling Experimental and Sampling Error
RCBD with Sampling Pooling Experimental and Sampling Error As we had with the CRD with sampling, we will have a source of variation for sampling error. Calculation of the Experimental Error df is done
More informationDiscipline of Economics, University of Sydney, Sydney, NSW, Australia PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by: [University of Sydney] On: 30 March 2010 Access details: Access Details: [subscription number 777157963] Publisher Routledge Informa Ltd Registered in England and Wales
More information[1]. S" = main stress, S = secondary stress, s = unstressed. Proto-Germanic: S s s s s s S s s s s s s S s s. Pintupi: S s S s S s S s S s S s s S s s
24.961 Stress-2 Trochaic typology (QI) [1]. S" = main stress, S = secondary stress, s = unstressed Proto-Germanic: S s s s s s S s s s s s s S s s Pintupi: S s S s S s S s S s S s s S s s Maranungku: S
More informationSTAT 113: Statistics and Society Ellen Gundlach, Purdue University. (Chapters refer to Moore and Notz, Statistics: Concepts and Controversies, 8e)
STAT 113: Statistics and Society Ellen Gundlach, Purdue University (Chapters refer to Moore and Notz, Statistics: Concepts and Controversies, 8e) Learning Objectives for Exam 1: Unit 1, Part 1: Population
More informationLife Domain: Income, Standard of Living, and Consumption Patterns Goal Dimension: Objective Living Conditions. Income Level
Life Domain: Income, Standard of Living, and Consumption Patterns Goal Dimension: Objective Living Conditions Measurement Dimension: Subdimension: Indicator: Definition: Population: Income Level I1113
More informationLINGUISTICS 321 Lecture #8. BETWEEN THE SEGMENT AND THE SYLLABLE (Part 2) 4. SYLLABLE-TEMPLATES AND THE SONORITY HIERARCHY
LINGUISTICS 321 Lecture #8 BETWEEN THE SEGMENT AND THE SYLLABLE (Part 2) 4. SYLLABLE-TEMPLATES AND THE SONORITY HIERARCHY Syllable-template for English: [21] Only the N position is obligatory. Study [22]
More informationin the Howard County Public School System and Rocketship Education
Technical Appendix May 2016 DREAMBOX LEARNING ACHIEVEMENT GROWTH in the Howard County Public School System and Rocketship Education Abstract In this technical appendix, we present analyses of the relationship
More informationAudio Feature Extraction for Corpus Analysis
Audio Feature Extraction for Corpus Analysis Anja Volk Sound and Music Technology 5 Dec 2017 1 Corpus analysis What is corpus analysis study a large corpus of music for gaining insights on general trends
More informationA NEW LOOK AT FREQUENCY RESOLUTION IN POWER SPECTRAL DENSITY ESTIMATION. Sudeshna Pal, Soosan Beheshti
A NEW LOOK AT FREQUENCY RESOLUTION IN POWER SPECTRAL DENSITY ESTIMATION Sudeshna Pal, Soosan Beheshti Electrical and Computer Engineering Department, Ryerson University, Toronto, Canada spal@ee.ryerson.ca
More informationOriginal citation: Yu, A. C. (2004) Efficient intra- and inter-mode selection algorithms for H.264/AVC. University of Warwick. Department of Computer Science. (Department of Computer Science Research Report).
More informationStudent Guide to the Publication Manual of the American Psychological Association Vol. 5
APA Short Guide 1 Student Guide to the Publication Manual of the American Psychological Association Vol. 5 I. Page Setup 1. Use margins of 1 inch (2.54 cm) on all sides and a line length of no more than
More informationTHE MASTER PLAN STUDY ON FLOOD FORECASTING AND WARNING SYSTEM FOR ATLAS REGION IN THE KINGDOM OF MOROCCO
No. JAPAN INTERNATIONAL COOPERATION AGENCY KINGDOM OF MOROCCO MINISTRY OF COUNTRY PLANNING, WATER AND ENVIRONMENT DIRECTORATE GENERAL OF HYDRAULICS THE MASTER PLAN STUDY ON FLOOD FORECASTING AND WARNING
More informationReliability. What We Will Cover. What Is It? An estimate of the consistency of a test score.
Reliability 4/8/2003 PSY 721 Reliability 1 What We Will Cover What reliability is. How a test s reliability is estimated. How to interpret and use reliability estimates. How to enhance reliability. 4/8/2003
More informationSoftware Package WW 9038 for the Sound Intensity Analysing System Type 3360 or the Digital Frequency Analyzer Type 2131
Software Package WW 9038 for the Sound Intensity Analysing System Type 3360 or the Digital Frequency Analyzer Type 2131 BO 0065-11 Software Package WW 9038 for the Sound Intensity Analysing System Type
More informationReconstruction of Ca 2+ dynamics from low frame rate Ca 2+ imaging data CS229 final project. Submitted by: Limor Bursztyn
Reconstruction of Ca 2+ dynamics from low frame rate Ca 2+ imaging data CS229 final project. Submitted by: Limor Bursztyn Introduction Active neurons communicate by action potential firing (spikes), accompanied
More informationQuick Start Function Summary Instructions for ASHCROFT GC52 Differential Pressure Transmitter Version 6.03 Rev. B
Quick Start Function Summary Instructions for ASHCROFT GC52 Differential Pressure Transmitter Version 6.03 Rev. B (See Complete I&M Manual for Further Detail) LOOK FOR THIS AGENCY MARK ON OUR PRODUCTS
More informationExercises. ASReml Tutorial: B4 Bivariate Analysis p. 55
Exercises Coopworth data set - see Reference manual Five traits with varying amounts of data. No depth of pedigree (dams not linked to sires) Do univariate analyses Do bivariate analyses. Use COOP data
More informationMrs. Norman s 2017 Unit Focus For Fahrenheit 451
Mrs. Norman s 2017 Unit Focus For Fahrenheit 451 Over the course of this novel, many of the Louisiana State Standards will be covered and assessed. Our focus standards, however, are RL.2 and RL.3. These
More informationQuantitative methods
Quantitative methods Week #7 Gergely Daróczi Corvinus University of Budapest, Hungary 23 March 2012 Outline 1 Sample-bias 2 Sampling theory 3 Probability sampling Simple Random Sampling Stratified Sampling
More informationTechnical report on validation of error models for n.
Technical report on validation of error models for 802.11n. Rohan Patidar, Sumit Roy, Thomas R. Henderson Department of Electrical Engineering, University of Washington Seattle Abstract This technical
More informationWHAT MAKES FOR A HIT POP SONG? WHAT MAKES FOR A POP SONG?
WHAT MAKES FOR A HIT POP SONG? WHAT MAKES FOR A POP SONG? NICHOLAS BORG AND GEORGE HOKKANEN Abstract. The possibility of a hit song prediction algorithm is both academically interesting and industry motivated.
More informationMoving on from MSTAT. March The University of Reading Statistical Services Centre Biometrics Advisory and Support Service to DFID
Moving on from MSTAT March 2000 The University of Reading Statistical Services Centre Biometrics Advisory and Support Service to DFID Contents 1. Introduction 3 2. Moving from MSTAT to Genstat 4 2.1 Analysis
More informationAutocorrelation in meter induction: The role of accent structure a)
Autocorrelation in meter induction: The role of accent structure a) Petri Toiviainen and Tuomas Eerola Department of Music, P.O. Box 35(M), 40014 University of Jyväskylä, Jyväskylä, Finland Received 16
More information10GBASE-R Test Patterns
John Ewen jfewen@us.ibm.com Test Pattern Want to evaluate pathological events that occur on average once per day At 1Gb/s once per day is equivalent to a probability of 1.1 1 15 ~ 1/2 5 Equivalent to 7.9σ
More informationBi-Modal Music Emotion Recognition: Novel Lyrical Features and Dataset
Bi-Modal Music Emotion Recognition: Novel Lyrical Features and Dataset Ricardo Malheiro, Renato Panda, Paulo Gomes, Rui Paiva CISUC Centre for Informatics and Systems of the University of Coimbra {rsmal,
More informationThe ACL Anthology Network Corpus. University of Michigan
The ACL Anthology Corpus Dragomir R. Radev 1,2, Pradeep Muthukrishnan 1, Vahed Qazvinian 1 1 Department of Electrical Engineering and Computer Science 2 School of Information University of Michigan {radev,mpradeep,vahed}@umich.edu
More informationN12/5/MATSD/SP2/ENG/TZ0/XX. mathematical STUDIES. Wednesday 7 November 2012 (morning) 1 hour 30 minutes. instructions to candidates
88127402 mathematical STUDIES STANDARD level Paper 2 Wednesday 7 November 2012 (morning) 1 hour 30 minutes instructions to candidates Do not open this examination paper until instructed to do so. A graphic
More informationMiles vs Trane. a is i al aris n n l rane s an Miles avis s i r visa i nal s les. Klaus Frieler
Miles vs Trane a is i al aris n n l rane s an Miles avis s i r visa i nal s les Klaus Frieler Institute for Musicology University of Music Franz Liszt Weimar AIM Compare Miles s and Trane s styles of improvisation
More informationFault Analysis of Stream Ciphers
Fault Analysis of Stream Ciphers M.Sc. Thesis Ya akov Hoch yaakov.hoch@weizmann.ac.il Advisor: Adi Shamir Weizmann Institute of Science Rehovot 76100, Israel Abstract A fault attack is a powerful cryptanalytic
More informationRectangle. Worksurface Ordering Guide. Width (W) Depth (D) esiergo.com. LT=Call for order lead time. Specify laminate (2 letter code)
Width (W) Depth (D) Specify laminate (2 letter code) Compatible table bases (As indicated by the ) Dimensions Model # Standard Price Triumph-LX Victory-LX 2-leg Victory-LX 3-leg All-Flex 2-leg All-Flex
More informationDeep Aesthetic Quality Assessment with Semantic Information
1 Deep Aesthetic Quality Assessment with Semantic Information Yueying Kao, Ran He, Kaiqi Huang arxiv:1604.04970v3 [cs.cv] 21 Oct 2016 Abstract Human beings often assess the aesthetic quality of an image
More informationarxiv: v1 [cs.sd] 13 Sep 2017
On the Complex Network Structure of Musical Pieces: Analysis of Some Use Cases from Different Music Genres arxiv:1709.09708v1 [cs.sd] 13 Sep 2017 Stefano Ferretti Department of Computer Science and Engineering,
More informationEXAMPLE: All-Flex 2-Leg ordering options. Dimensions Base model # WL WR D
Specifying your laminate choice Complete the worksurface model # with the 2-letter laminate code. 2R4824KM = 48" x 24" Rectangle Worksurface in Kensington Maple Worksurfaces - LT Typical lead time is 15-20
More informationApplication Note AN39
AN39 9380 Carroll Park Drive San Diego, CA 92121, USA Tel: 858-731-9400 Fax: 858-731-9499 www.psemi.com Vector De-embedding of the PE42542 and PE42543 SP4T RF Switches Introduction Obtaining accurate measurement
More informationSources of Error in Time Interval Measurements
Sources of Error in Time Interval Measurements Application Note Some timer/counters available today offer resolution of below one nanosecond in their time interval measurements. Of course, high resolution
More informationKlee or Kid? The subjective experience of drawings from children and Paul Klee Pronk, T.
UvA-DARE (Digital Academic Repository) Klee or Kid? The subjective experience of drawings from children and Paul Klee Pronk, T. Link to publication Citation for published version (APA): Pronk, T. (Author).
More informationTechnical Specifications
1 Contents INTRODUCTION...3 ABOUT THIS LAB...3 IMPORTANCE OF THE MODULE...3 APPLYING IMAGE ENHANCEMENTS...4 Adjusting Toolbar Enhancement...4 EDITING A LOOKUP TABLE...5 Trace-editing the LUT...6 Comparing
More informationWidth Right (WR) Victory-LX 2-leg. Standard Price. Triumph-LX
Width Right (WR) Depth (D) Width Left (WL) Depth (D) Specify laminate (2 letter code) Compatible table bases (As indicated by the ) Dimensions Model # Standard Price Triumph-LX Victory-LX 2-leg Victory-LX
More informationChapter 2 Introduction to
Chapter 2 Introduction to H.264/AVC H.264/AVC [1] is the newest video coding standard of the ITU-T Video Coding Experts Group (VCEG) and the ISO/IEC Moving Picture Experts Group (MPEG). The main improvements
More informationSEGMENTATION, CLUSTERING, AND DISPLAY IN A PERSONAL AUDIO DATABASE FOR MUSICIANS
12th International Society for Music Information Retrieval Conference (ISMIR 2011) SEGMENTATION, CLUSTERING, AND DISPLAY IN A PERSONAL AUDIO DATABASE FOR MUSICIANS Guangyu Xia Dawen Liang Roger B. Dannenberg
More informationThe Lowest Form of Wit: Identifying Sarcasm in Social Media
1 The Lowest Form of Wit: Identifying Sarcasm in Social Media Saachi Jain, Vivian Hsu Abstract Sarcasm detection is an important problem in text classification and has many applications in areas such as
More informationCommentary on David Huron s On the Role of Embellishment Tones in the Perceptual Segregation of Concurrent Musical Parts
Commentary on David Huron s On the Role of Embellishment Tones in the Perceptual Segregation of Concurrent Musical Parts JUDY EDWORTHY University of Plymouth, UK ALICJA KNAST University of Plymouth, UK
More informationNO-REFERENCE QUALITY ASSESSMENT OF HEVC VIDEOS IN LOSS-PRONE NETWORKS. Mohammed A. Aabed and Ghassan AlRegib
214 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) NO-REFERENCE QUALITY ASSESSMENT OF HEVC VIDEOS IN LOSS-PRONE NETWORKS Mohammed A. Aabed and Ghassan AlRegib School of
More informationImplementing sharpness using specific loudness calculated from the Procedure for the Computation of Loudness of Steady Sounds
Implementing sharpness using specific loudness calculated from the Procedure for the Computation of Loudness of Steady Sounds S. Hales Swift and, and Kent L. Gee Citation: Proc. Mtgs. Acoust. 3, 31 (17);
More informationRunning head: FACIAL SYMMETRY AND PHYSICAL ATTRACTIVENESS 1
Running head: FACIAL SYMMETRY AND PHYSICAL ATTRACTIVENESS 1 Effects of Facial Symmetry on Physical Attractiveness Ayelet Linden California State University, Northridge FACIAL SYMMETRY AND PHYSICAL ATTRACTIVENESS
More informationKatalin Tamási & Iris Berent
Sensitivity to Phonological Universals: The Case of Stops and Fricatives Katalin Tamási & Iris Berent Journal of Psycholinguistic Research ISSN 0090-6905 DOI 10.1007/s10936-014-9289-3 1 23 Your article
More informationA Categorical Approach for Recognizing Emotional Effects of Music
A Categorical Approach for Recognizing Emotional Effects of Music Mohsen Sahraei Ardakani 1 and Ehsan Arbabi School of Electrical and Computer Engineering, College of Engineering, University of Tehran,
More informationAbstract. Keywords Movie theaters, home viewing technology, audiences, uses and gratifications, planned behavior, theatrical distribution
Alec Tefertiller alect@ksu.edu Assistant professor. Kansas State University in Manhattan, Kansas, USA. Submitted January 23, 2017 Approved May 22, 2017 Abstract 2017 Communication & Society ISSN 0214-0039
More informationSupplementary Figures Supplementary Figure 1 Comparison of among-replicate variance in invasion dynamics
1 Supplementary Figures Supplementary Figure 1 Comparison of among-replicate variance in invasion dynamics Scaled posterior probability densities for among-replicate variances in invasion speed (nine replicates
More informationAdaptive decoding of convolutional codes
Adv. Radio Sci., 5, 29 214, 27 www.adv-radio-sci.net/5/29/27/ Author(s) 27. This work is licensed under a Creative Commons License. Advances in Radio Science Adaptive decoding of convolutional codes K.
More informationarxiv:cs/ v1 [cs.ir] 23 Sep 2005
Folksonomy as a Complex Network arxiv:cs/0509072v1 [cs.ir] 23 Sep 2005 Kaikai Shen, Lide Wu Department of Computer Science Fudan University Shanghai, 200433 Abstract Folksonomy is an emerging technology
More informationMATH 214 (NOTES) Math 214 Al Nosedal. Department of Mathematics Indiana University of Pennsylvania. MATH 214 (NOTES) p. 1/11
MATH 214 (NOTES) Math 214 Al Nosedal Department of Mathematics Indiana University of Pennsylvania MATH 214 (NOTES) p. 1/11 CHAPTER 6 CONTINUOUS PROBABILITY DISTRIBUTIONS MATH 214 (NOTES) p. 2/11 Simple
More informationLeveled Libraries K-8. Bookrooms. New! Oral reading records now available for all titles! Program Highlights
Leveled Libraries K-8 Leveled Libraries K-8 Great books at great savings with teacher support included! Program Highlights Books available from levels A - Z Wide range of fiction and informational text
More informationEE 109 Homework 6 State Machine Design Name: Score:
EE 9 Homework 6 State Machine esign Name: Score: ue: See Blackboard Blackboard ONLY Submission. While the Blackboard submission may not require you to go through all the design steps (such as drawing out
More informationCHORDAL-TONE DOUBLING AND THE ENHANCEMENT OF KEY PERCEPTION
Psychomusicology, 12, 73-83 1993 Psychomusicology CHORDAL-TONE DOUBLING AND THE ENHANCEMENT OF KEY PERCEPTION David Huron Conrad Grebel College University of Waterloo The choice of doubled pitches in the
More informationEnsemble LUT classification for degraded document enhancement
Ensemble LUT classification for degraded document enhancement Tayo Obafemi-Ajayi, Gady Agam, Ophir Frieder Department of Computer Science, Illinois Institute of Technology, Chicago, IL 60616 ABSTRACT The
More informationOpen Access Determinants and the Effect on Article Performance
International Journal of Business and Economics Research 2017; 6(6): 145-152 http://www.sciencepublishinggroup.com/j/ijber doi: 10.11648/j.ijber.20170606.11 ISSN: 2328-7543 (Print); ISSN: 2328-756X (Online)
More informationFPGA-BASED IMPLEMENTATION OF A REAL-TIME 5000-WORD CONTINUOUS SPEECH RECOGNIZER
FPGA-BASED IMPLEMENTATION OF A REAL-TIME 5000-WORD CONTINUOUS SPEECH RECOGNIZER Young-kyu Choi, Kisun You, and Wonyong Sung School of Electrical Engineering, Seoul National University San 56-1, Shillim-dong,
More informationWyner-Ziv Coding of Motion Video
Wyner-Ziv Coding of Motion Video Anne Aaron, Rui Zhang, and Bernd Girod Information Systems Laboratory, Department of Electrical Engineering Stanford University, Stanford, CA 94305 {amaaron, rui, bgirod}@stanford.edu
More informationArea-efficient high-throughput parallel scramblers using generalized algorithms
LETTER IEICE Electronics Express, Vol.10, No.23, 1 9 Area-efficient high-throughput parallel scramblers using generalized algorithms Yun-Ching Tang 1, 2, JianWei Chen 1, and Hongchin Lin 1a) 1 Department
More informationGuide for writing assignment reports
TELECOMMUNICATION ENGINEERING UNIVERSITY OF TWENTE University of Twente Department of Electrical Engineering Chair for Telecommunication Engineering Guide for writing assignment reports by A.B.C. Surname
More informationSCT Activities. Nick Bedford, Mateusz Dyndal, Alexander Madsen, Edoardo Rossi, Christian Sander. DESY ATLAS Weekly Meeting 03. Jun.
SCT Activities Nick Bedford, Mateusz Dyndal, Alexander Madsen, Edoardo Rossi, Christian Sander DESY ATLAS Weekly Meeting 03. Jun. 2016 1 Semi-Conductor Tracker Barrel 4 Layers 2112 identical modules Endcaps
More informationTechnical note: 5 micron fibres found in an ultrafine grower lot - implications for diameter distribution measurement By B. P.
TECHNOLOGY & STANDARDS COMMITTEE Raw Wool Group Chairman: A.C. BOTES (South Africa) SHANGHAI MEETING May 2001 Report No: RWG 02 Technical note: 5 micron fibres found in an ultrafine grower lot - implications
More informationUniversity of Bristol - Explore Bristol Research. Peer reviewed version. Link to published version (if available): /ICASSP.2016.
Hosking, B., Agrafiotis, D., Bull, D., & Easton, N. (2016). An adaptive resolution rate control method for intra coding in HEVC. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing
More informationERROR CONCEALMENT TECHNIQUES IN H.264
Final Report Multimedia Processing Term project on ERROR CONCEALMENT TECHNIQUES IN H.264 Spring 2016 Under Dr. K. R. Rao by Moiz Mustafa Zaveri (1001115920) moiz.mustafazaveri@mavs.uta.edu 1 Acknowledgement
More informationMUSIC transcription is one of the most fundamental and
1846 IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 25, NO. 9, SEPTEMBER 2017 Note Value Recognition for Piano Transcription Using Markov Random Fields Eita Nakamura, Member, IEEE,
More informationSupplemental Material for Gamma-band Synchronization in the Macaque Hippocampus and Memory Formation
Supplemental Material for Gamma-band Synchronization in the Macaque Hippocampus and Memory Formation Michael J. Jutras, Pascal Fries, Elizabeth A. Buffalo * *To whom correspondence should be addressed.
More informationOn the Complexity-Performance Trade-off in Code-Aided Frame Synchronization
On the Complexity-Performance Trade-off in Code-Aided Frame Synchronization Daniel Jakubisin and R. Michael Buehrer Mobile and Portable Radio Research Group (MPRG), Wireless@VT, Virginia Tech, Blacksburg,
More informationDoes Gujarati Stress Avoid the Lowest Sonority Vowel [ə]? Shu-hao Shih Rutgers University
Does Gujarati Stress Avoid the Lowest Sonority Vowel [ə]? Shu-hao Shih Rutgers University shuhao.shih@rutgers.edu CUNY 2016, 14 January 2016 Intro What is sonority-driven stress? [awːána ] coming [kójəldi]
More informationALF-200k: Towards Extensive Multimodal Analyses of Music Tracks and Playlists
ALF-200k: Towards Extensive Multimodal Analyses of Music Tracks and Playlists Eva Zangerle, Michael Tschuggnall, Stefan Wurzinger, Günther Specht Department of Computer Science Universität Innsbruck firstname.lastname@uibk.ac.at
More informationResearch Article Design and Analysis of a High Secure Video Encryption Algorithm with Integrated Compression and Denoising Block
Research Journal of Applied Sciences, Engineering and Technology 11(6): 603-609, 2015 DOI: 10.19026/rjaset.11.2019 ISSN: 2040-7459; e-issn: 2040-7467 2015 Maxwell Scientific Publication Corp. Submitted:
More informationAUTOREGRESSIVE MFCC MODELS FOR GENRE CLASSIFICATION IMPROVED BY HARMONIC-PERCUSSION SEPARATION
AUTOREGRESSIVE MFCC MODELS FOR GENRE CLASSIFICATION IMPROVED BY HARMONIC-PERCUSSION SEPARATION Halfdan Rump, Shigeki Miyabe, Emiru Tsunoo, Nobukata Ono, Shigeki Sagama The University of Tokyo, Graduate
More informationNETFLIX MOVIE RATING ANALYSIS
NETFLIX MOVIE RATING ANALYSIS Danny Dean EXECUTIVE SUMMARY Perhaps only a few us have wondered whether or not the number words in a movie s title could be linked to its success. You may question the relevance
More informationERROR CONCEALMENT TECHNIQUES IN H.264 VIDEO TRANSMISSION OVER WIRELESS NETWORKS
Multimedia Processing Term project on ERROR CONCEALMENT TECHNIQUES IN H.264 VIDEO TRANSMISSION OVER WIRELESS NETWORKS Interim Report Spring 2016 Under Dr. K. R. Rao by Moiz Mustafa Zaveri (1001115920)
More informationKing Fahd University of Petroleum and Minerals Electrical Engineering Department 1. Homework 5 - SOLUTION KEY
Electrical Engineering Department 1 Homework 5 - SOLUTION KEY EE-306 Electromechanical Devices - Semester 162 Electrical Engineering Department 2 Problem 1 Consider a Europeon city, it is necessary to
More information