Top t-angDisordersVol.25, No. 3, pp. 207-21920()'i
LippmcotrWilliams & Wilkins, Inc.Stimulabilityand
TreatmentSuccessSusan
Rvacbew,PhDThisarticleaddresses2questionsofimportancetothetreatmentofspeechsounddisorders:(1)
When selecting treatmenttargets, is it bestto begin with themost or
the least stimulable po-tential phoneme targets? (2) When treating
imstimulable phonemes, which treatmentprocedureswill resultin the
best outcome?A summary of the findings from3
randomizedcontrolledtrials isprovided.In these studies, outcomes
were generally better whenstimulable targets were
treated;however,outcomesforunstimulahletargetswereimprovedby
includingphonemicperceptiontraining alongside
phoneticplacementproceduresin thetreatmentprogram. The
clinicianmusttake finalresponsibilityforjudging theapplicability of
these re,search findings to eachindividualcase. Clinical decisions
should be made after discussing the knownbenefits and ri,sks of
anygiventreatment practice with the client and/or the clients
family. Key v/ords: ei'ieience based practice,phonological
disorder,speech perception, speech sounddisorder,speech therapy,
stimulabilitySTIM U IJ V B IL ITYreflectsa child's
ability-tocorrectlyimitateagivenphonemewhenprovidedwiththeinstructionto"watchatidlisten"followedbymodelsofthephoneme,usually
in the contextof nonsensesyllables
orsimplerealwords(foranhistoricaloverviewoftheclinicalapplicationofstimtilability-testingprocedures,seePowell&Miccio,1996).Stimuiabilitytestingisrecommendedasanessentialpartofanassessment,evenwhenthe"clockis
running"
andtheclinicianisforcedtomakedeparturesfromidealas-sessmentpractices(e.g.,
Bleile,2002; Miccio,2002).Promthe School of Communication Sciences
andDisorders,Mcdill University.Montreal,Quebec,Canada.Theauthor
helped develop and has a financial interestin Speech Assessment and
Interactive Learning Systems,a treatment procedure that was
evaluated in severaloftbe studies summarizedin this article.This
article wasprepared with support fmmthe Cana-dian Language
andLiteracy Research Netu^ork(wivw.cllnet.ca). Portions of this
article were /nvsented as partofaseminaron"Evidence-Based
Practice:TheWhy andthe How"chaired fry AmyWeiss and Rehecca
McCavley,at the2004 annualconvention ofthe A merican
Speech-Langtiage-Hearing Association.Corresponding author:
.SusanRvachew,PhD,SchoolofCommunication Sciences and Disorders,
McGill Univer-sity,1266Pine Avenue West, Montreal,Quebec,CanadaH3G
IA8(e-mail:
susan.rvacheu>@mcgillca).Onereasonforincludingstimulabilitytest-ing
in anassessmentof a
child'spbonologicalskillsisthattheresultsmayhaveclearprog-nosticindications.Improvementsinarticula-tory
accuracy for an
unmasteredphonemearemorelikelytooccuroveragiventimepe-riodwhenthechildis
stimulable,ratherthanunstimulable,foraccurateproductionofthespeechsound
in question (e.g., Miccio,
Elbert,&Forrest,1999;Tyler,1996),Furthermore,childrenwhohavegenerallyhigherlevelsofstimulabilityshowbetterprogressovertimethanchildrenwhohavegenerallylowerlevelsofstimulability(eg.,Sommers,Leiss,Delp,Gerber,Fundrella,Smith,Revucky,Ellis,&Haley,1967),Tliesefindingsapplyregardlessofwhetherchildrenshowtyp-icalordelayedspeechdevelopmentandregardlessofwhethertheyarereceivingintervention.Tbereis
a
consensusthatstimulabilitytestrestiltsalsohaveimplicationsfortreatmentplanningin
generalandfortargetselectioninparticular.Thereisconsiderablecontroversyabouttheclinicalimplicationsofstimulabil-ityoralackthereofforthedesignofanet-fectivephonologyintervention,however(foran
historicaloverviewof this
controversy,seePowell&Miccio,1996),Basicresearchonstimulabilitypresentsaproblemforclini-caldecisionmakingbecausetheapparent207208TOPICS
IN LANGUAGEDISORDERS/JULY-SEPTEMBER
2005implicationsofthebasicresearchfindingsare contradictory: if
stimulable phonemes
im-proveevenwithouttreatment,thelogicalcotirseofactionmightbetoavoidthesephonemesas
treatmenttargetsandfocusin-tervention on the unstimulahle phonemes;
onthe other hand, if unstimulablephonemes
donotimproveevenwhentheyaretreated,anequallylogicalcourseofactionmightbetoavoid
themas treatmenttargets and focus
in-terventiononthestimulablephonemes. Re-solvingtheissue
requirestheclinicianto re-view the relevant literature while
applying theprinciples of evidence-basedpractice to
iden-tifythecourseof actionthatis likelytoleadto the best clinical
outcome for a given client.Thepurposeof thisarticleis
toexaminetbeempiricalevidencerelatingtotreatmentofunstimulable and
stimulable
phonemes.THEORIESABOUTTHENATUREOFSTIMULABIUTYTraditionally,stimulabilit>'wasviewedasevidenceofthestructuralandfunctionalin-tegrityoftbechild'sspeechmechanism.
Amorerecenttheorypositsthattheabilityto imitate a speech sound
reflectsunderlyingphonologicalknowledge aboutthe pboneme(Powell
& Miccio, 1996), Phonological
knowl-edgeincludesanunderstandingofthelin-guisticfeaturestbatforma
givenpboneme.thecontrastiverelationshipsbetweenthephonemeandotberphonemes,andthepbonotacticenvironmentsinwbicba
givenpbonememayoccur.Underlyingphonolog-icalknowledgeis,inturn,seenastheba-sistorgeneralizationofarticulatoryaccu-racybothwithinandacrossphonemesandphoneme
classes. Tliis theoretical perspectiveis
linkedtotbesuggestionthatunstimulablephonemes should be the focus
of
interventionbecauseteachingtbesesoundsprovidesanopportunitytointroducenewphonologicalknowledgethatwillreorganizethechild'sphonologicalsystemandresultinenhancedgeneralizationof
learning fromtreatedto un-treated phonemes.Lof (1996) examined the
correlates of stim-ulabilityinanefforttoilluminatetherela-tionship
between stimulability and
phonolog-icalknowledge.Thirtychildren'sunderlyingphonologicalknowledgeforeachofseveralcommonlymisarticulatedphonemeswas
de-terminedbyassessingphonemicperceptionusingaprocedurerecommendedbyLocke(1980).Specificallythechildwasaskedtojudgethecorrectnessoftargetwordspro-ducedby
an adult,eitbercorrectlyorwith
asimulatederror(i.e.,iftbechildhascorrectphonemicperceptionforthe/I/phoneme,[lif]
shouldbe perceivedas a
correctproduc-tionoftbeword"leafbut[wil]shouldbejudged to be
incorrect), Lof found that all
pos-siblecombinationsofphonemicperceptionperformanceandstimulabilityoccurred.Forexample,tbesamechildmightperceive/k/and
be stimulable for tbis pboneme,
perceive/r/andnotbestimulableforthisphoneme,misperceive/I/andbestimulablefortbisphoneme,andmisperceive/s/butnotbestimulable
for this pboneme. Rvachew, Rafaat,and Martin (1999) replicated this
finding withtwoadditionalsamplesofcbildren,usingasimilar
testprocedure exceptthatrecordingsof authenticrather
thansimulatederror pro-ductions were used for tbe phonemic
percep-tion task. These data, summarizedin Table
1,demonstratethatunderlyingknowledgeof aphonological contrast and
stimulability are in-dependentconstructs.Lof (1996) found that
stimulability was pos-itively associatedwith greater visibility of
thetarget sound, older age of the child, higher
so-cioeconomicstatus of the family,and
greateroverallspeech-imitationabilitiesofthechildasreportedbytheparent,Lofspeculatedthatstimulabilitymay
also reflectchild focus(Kwiatkowski & Shriberg, 1993), meaning
tbepropensityofthechildtoattendtothe rele-vant aspects of the model
combined with thechild'smotivationtochangehisorberpro-duction
accuracy.Theseresearcb findings
suggestthatstim-ulabilityandphonemicperceptionareinde-pendentabilitiesthatareeachrequiredforphoneme
acquisition as illustrated in Figure
1.StimulabilityandTreatmentSuccess209Table1.Relationship
betweenphonemic perceptionand stimulability across three studies
and53 children*StudyLf(1996)Rvachew ctRvachew
etTotalalal(1999),(1999),StudyStudy12Perceiveandstimulable6101026Notperceiveandnotstimulable103215Perceivebut
notstimulable91616Notperceivebutstimulable57820*The values
represeni numbers of phoneme,s Lorresponding to each combination of
perceptual abiliw and stimulability.The totals sum to greater
thantbenumber of subjectsbecause multiplephonemes were testt-d
persubject.The totalnumber of phonemes assessed, summed across the
53 children, was
77.Stimulabilityreflectsthestructuralandfunc-tionalintegrityof
thespeechmechanism,thechildsaccesstovisual,tactile,andkines-thetic
informationabout the requiredarticula-tory
gestures,imitationskills, and
childfocus.Phonemicperceptionarisesfromthestruc-turalandfunctionalintegrityofthechildsauditoryandspeechperceptionmechanisms,thechild'saccesstoappropriatespeechin-put,andthecognitive-linguisticskillsre-quiredforprocessingandlearningfromthatinput.TheframeworkillustratedinFigure1
alsoreflectsthefindingthatstimulabilitydoesnotbyitselfleaddirectlyandinevitablytophonemeacquisition,
even whenthe child re-ceivestherapyforremediationofthespeechsound
errors (Miccio etal.,1999;
Tyler,1996).Rvachewetal.(1999)showedthatimprove-mentsinspeechsoundaccuracyforagivenstructural
and functional integrity ofthe auditoryand
speechperceptionmechanismsAppropriateinput(many and
variedexemplarsof
contrastingphonemecategories).Cognitive-linguisticprocessingabilitiesPhonemicperceptionPhonemeacquisitionStructural
and functional integrity of'the speech
mechanism.Appropriateinput(visual, tactile,
andkinaestheticinformationabout therequired articulatory
gestures).Imitation skills.Focus and
motivation.StimulabilityFigure1. An
illustrationoftheproposedrelationshipsamongphonemicperception,stimulability,andphoneme
acquisition.210TOPICSIN LANGUAGE DisoRDERsyJiiLY-SEPTEMBER
2005phoneme were most likely to occur when
thechilddemonstratedstimulabilityandphone-mic perception prior to
the onset of a 12-weekphonologyintervention (improvements
wereobserved for 64% of phonemes that were
per-ceivedandstimulableand0% ofphonemesthat were not perceived and
not stimulable).A theoryoftheprocessbywhichphone-mic perceptionand
stimulabilitymightinter-acttoproducephonemeacquisitionisin-stantiatedinaneuralnetworkmodelcalledDirections
inauditoryspace toVelocities inArticulatorspace (DIVA;
Guenther.1995).Acentralfeatureof thismodelis
thatthegoalofarticulationis assumedtobetheachieve-mentof a
particular acoustic-phoneticprod-uct(as opposedto a
specificconstellation ofarticulatorygestures). Thelearneruses
audi-toryfeedbacktoidentifydeviations
fromthedesiredacoustic-phonetietargetspaceandthenadaptsthe
articulatory gestures
tobothachieveandmaintaitiarticulatoryaccuracy,evenasthevocaltractisundergoingdevel-opmentalchangesinsizeandshape(Callan,Kent,Guenther,&
Vorperian,2000). For ex-ample,the /JV phonemeis
characterizednotbyaspecificpeakfrequencyofnoiseen-ergy, but rather
by a range of
frequenciesthatarelowrelativetothosepeakfrequenciesthatcharacterizethe
/s/ phoneme.
Feedbackoftheacoustic-phoneticproductofspeechmovementsallowsthechildtolearnto
pro-duce the varying constellations of articulatorygestures that
will result in this phoneme in dif-ferentphonetic contexts (e.g.,
greater tongueretractionisrequiredinthecontextofun-rounded than
rounded vowels).Figure1 alsosuggestssomeinterventionproceduresthat
may beeffective.Theemer-genceofphonemicperceptionandstimula-biiity
eachrequire certaininputs to the
childthatarefullyunderthecontroloftheclini-cian.Manyempiricalinvestigationshavere-vealedthetypeofinputthatisrequiredfortheinductionofcategoricalperceptionofphonemes(Guenther,Husain,Cohen,&Shinn-Cunningham,1999;Maye,Werker,&Gerken,2002;Rvachew,1994).Specifically,the
learner must hear a broadly vaiying distri-butionof exemplars of
contrastingphonemecategories.Forexample,whenteachingthechild to
perceive the contrast between /s/ and/J/, it is bestto presenta
large number of
ex-emplarsofwordscontaining/s/andwordscontaining/J/,ideallyproducedbydifferenttalkers.
Most of the words that the child
hearsshouldrepresenttheprototypicalexemplarsofthetwosoundcategories(i.e.,lowerfre-quency
fricative noise for /J/ and much higherfrequency fricative noise
of/s/). However, it isalso importantto presentsome exemplars
ofthese phonemesthat are less than ideal, eventhose approaching the
category boundary be-tweenthe two
sounds.Theinputsrequiredforachievingstimula-bilityshouldprovideinformationabouttheappropriatepositioningofthearticulatorsandhelpthechildfocusattentionontac-tile
andkinesthetic feedbackassociated
withtheproductionoftheappropriatearticula-torygestures.Modelling,shaping,andpho-neticplacementtechniquesare
suggestedinSecord(1981)andhavebeenshowntobeusefulforremediatingspeechsounderrors(Powell,Elbert,Miccio,Strike-Roussos,&Brasseur,1998).EVALUATING
THEEVTOENCEAlthoughmanytreatmentefficacystudiesthatarerelevanttothetopicofstimula-bilityhavebeenpublished,mostofthesehaveemployedexperimental(e.g.,multiplebaseline)ornonexperimental(e.g.,multipleprobe)
single-subject designs and thus are
notparticularlywellsuitedtothetworesearchquestionsconsideredhere(Gierut,Elbert,
&Dinnsen, 1987; Gierut, Morrisette, Hughes, &Rowlands,2001;
Powell,Elbert,& Dinnsen,1991;Tyler& Figurski,1994). A
fullreviewandcritiqueofalltherelevanttreatmentef-ficacy studiesis
beyondthescopeof tbis ar-ticle. However, a fewstudies involving a
ran-domizedcontrolled design will bediscussed.A
randomizedcontrolledtrialis essentialforstudying
thistopicbecauseitis theonly
de-signthatcanhelpdifferentiatematurationfrom generalization and
treatment effects(forStimulabiiityandTreatment
Success211furtherdiscussionofresearchdesignissues,see Rvachew &
Nowak,
2001).STIMULABILITYANDTARGETSELECTIONSTRATEGYRvachewandNowak(2001)investigatedthe
relative effectivenessof two distinct
targetselectionstrategiesontreatmentoutcomes,using a
randomizedcontrolleddesign.Forty-eight 4-year-(3ld children witb
moderate or se-verespeechsounddisordersreceiveda
pre-treatmentassessment (Al), a (>week block oftreatment
targeting two consonantphonemes(bUx:k1), a second assessment(A2),
another6-weekblockoftreatmenttargetingtwoad-ditionalconsonantphonemes(block2),
andaposttreatmentasse.ssment(A3),All assess-ments were conducted by
a speech-langtiagepathologist who was blind to the
childsgroupassignmentandtreatmenttargets.Thetreat-mentapproachduringeachblockwastra-ditional,withphoneticplacement,shaping,andmodellingtechniquesusedtoestablishstimulabilityatthesyllablelevel.Oncestim-ulabilityinsyllableswasachieved,thechildwasprovidedwithopportunitiesforartic-ulationpracticeinimitatedwords,sponta-neouswords,imitatedpatternedsentences,spontaneouspatternedsentences,imitatedsentences,andspontaneoussentences.Allchildrenbeganthetreatmentprogramatthestimulationphaseandprogressedfromonetreatmentsteptothenextuponachieving80%
correctresponding.Thephonemesthatweretargeteddur-ingthetwotreatmentblocksdependedonwhetherthechildwasassignedatrandomtoreceivetreatmentforthemoststimula-ble,
early developingsounds(ME group)ortheleast stimulable,late
developing sounds(LL group). At each of the three
assessments,thechildwas askedtoimitateall itemsfromthe
ProductivePhonologicalKnowledge Pro-file (Gierutetal.,1987),
whichprovides19opportunities to producemost of the
Englishconsonants. The child's responses were usedto rankthe
child's unmasteredphonemes
ac-ct)txlingtotypeofProductivePhonologicalKnowledge. Gierutetal.
describedsixtypesofproductivephonologicalknowledgethatreflecttheconsistencywithwhichthechildarticulates
a givenphonemeacross word po-sitions. Potential treatment targets
were
thosephonemesthatwereattype4(masteredinatleastonebutnotallthreewordposi-tions),type5
(inconsistentlycorrectbutnotmasteredinanywordposition),ortype6(never
correct). Rvachew and Nowak consid-ered type 6 pbonemes in their
study to be un-stimulable because the child failed to producethe
phoneme correctly even after15 opportu-nities to imitate it in
simple words. These po-tential treatmenttargetswere
furtherrankedwithineachtypeofProductivePhonologicalKnowledgeaccordingtothe90thpercentileageofacquisition,accordingtoSmit,Hand,Freilinger,Bernthal,andBird
(1990). In eachtreatmentblock, children assignedto the MEgroup
received treatmentfor the two soundsthatweremoststimulable
andearliest
devel-oping,withtheprovisothatthetwosoundsnotsharethesamemannerclass(e.g.,/k/and
/f/would be selected as treatment
targetsratherthan/k/and/g/).ChildrenassignedtotheLL
groupreceivedtreatmentforthetwo sounds that were least stimulable
and lat-est developing, again with the
constraintthatthetwosoundsnotsharethesamemannerclass(e.g.,/r/and/
s/wouldbeselectedastreatmenttargets rather than / s/ and /J/),
ThecriteriaforchoosingtargetsforchildrenintheLL
conditioneffectivelyensuredthatthisgroup would receive treatmentfor
unstimula-ble sounds, at least during the first
treatmentblock.Themostfrequentlytargetedsoundsduringthefirsttreatmentblockforthe
MEgroup were /k/and /f/.Themostfrequentlytargeted consonants during
the first treatmentblock for the LL group were / s/ and /r/,Figtire
2 shows mean group response accu-racy for three sets of consonants
from the
Pro-ductivePhonologicalKnowledgeProfileforthethreeassessments,withtheME
group'sperfonnanceshown in the leftpanel (2a) andtheLL
group'sperformanceshownintheright panel (2h). The three sets of
consonantsillustrated in Figure 2a are the most
stimulable212a.14cQ.10(ATOPICSINLANGUAGEDISORDERS>(JULY-SEPTEMBER2005b.TreatmentprovidedTreatmentnotprovidedBlock1
targetBlock2 targetNevertreatedAl
A2AssessmentA3140)1^VIcoTreatmentprovidedTreatmentnotprovidedNevertreatedBlock2
targetBlock1 targetAl
A2AssessmentA3Figure2.MeanntimberofcorrectresponsesforcertainconsonantsontheProductivePhonologicalKnowledgeProfile,administeredatthreeassessmentpoints:priortotreatment(Al),afterthefirsttreat-mentblock(A2),
andafterthesecondtreatmentblock{A3).
Tliepanelontheleft(a)showstheresultsforchildrenintheME
group,andthepanelontheri^ht(b)showstheresultsforthechildreninthe
LLgroup. (BasedondatafromRvachew& Nowak,2001).phonemes (treated
in the first block), the
sec-ondmoststimulablephonemes(treatedinthesecondblock),andtheleaststimulablephonemes(nevertreatedintheME
group).ThethreesetsofconsonantsillustratedinFigure2b
aretheleaststimulablephonemes(treatedinthefirstblock),thesecondleaststimulablephonemes(treatedinthesecondblock),andthemoststimulabiephonemes(never
treated in the LL group).Expressingthechangeinscores
fromoneassessmenttothenextintheformofmeanresidualizedgainscoresallows
fora
numberofdifferentcomparisonsofrelativerateofchange,aftertakingintoaccountthechild'spretreatmentlevelofproductiveaccuracyforthesephonemes(specifically,aresidual-ized
gain score is the differencebetweenthechild s
actualposttreatmentprobescore
andthescorethatwouldbepredictedfromthechild
spretreatmentprobescore,asdeter-mined by a regression equation
derivedfromthepretreatmentandposttreatmentprobescores for all 48
children).First,therateofchangeduringblockswhenthephonemesweretreatedwas
com-paredwiththerateof
changeobserveddur-ingblockswhenthephonemeswerenottreated.Theseanalyseswereconductedsep-arately
withineach group and
cotifirmedthedifferencesinrateofchangethatareread-ily observablein
Figure 2. For
theMEgroup,therateofchangefortbemoststimulablephonemes was higher
during the block
wbenthesepbonemesweretreated,thanduringtheblockwhenthesephonemeswerenottreated;0.44versus0.05:K95)
=2.60, p=.011;seesolidlinesversusdashedlinesforblocks1and 2 targets
in Figure 2a. For the LLgroup, the rate of changefortheleast
stimu-lablephonemes was higher during the blockwhen these phonemes
were treated, than dur-ingtheblockwhenthesephonemeswerenottreated;
0.08 versus-0.56; /(95) =4.58,StimulabilityandTreatment
Success213p=,000; seesolidlinesversusdashedlinesfor blocks 1and 2
targets in Figure 2b, In otherwords, gains were greater
forblock1targetsduring block1thanduringblock2. whereasgains
weregreater forblock2 targets duringblock 2 than during block1,
This was true forboth
groups.Ihesecondanalysisconsideredrateofchangefortreatedphonemesacrossthe
MEand LL groups. The rate of change in produc-tion accuracyfrom A1
to A3 for treated stim-ulable phonemes was greater than the rate
ofchangefortreatedtinstimtilablephonemes;0,25versus-0.51;K9i)=3-78,
p=.000;highestline in Figure 2a versus lowest line inFigure
2b,Tbetbirdanalysisconsideredtherelativerate of change for
stimulable pbonemes acrossthe ME and LL groups. From Al to A3, tbe
rateof change shown by the ME group for
treatedstimulablephonemeswasnotsignificantlygreater thanthe rate of
change sbownby theLL group for untreated stimulable phonemes;0,25
verstis 0,21; rt:94) =0.22,/? =,829;
high-estlineinFigure2aversusbigbestlineinFigure
2b.Thefinalanalysisconsideredtherateofchange for unstimulable
phonemes across theME andtheLL groups.FromAltoA3, therate of change
shown by the ME group for un-treated unstimulablepbonemeswas
actuallysignificantlygreaterthantherateofchangeshownbytbeLLgroupfortreated
unstimu-lablephonemes;0,05versus-0,51;?(94)=2,97, p=,000;
lowestlineinFigure2a
ver-suslowestlineonFigure2b,Overall,treat-mentprogresswasverypoorforunstimu-lablepbonemesbuttreatingthestimulablephonemes
first appeared to facilitate
improve-mentsforunstimulablephonemesinthe
MEgroup.Othermoreglobalmeasuresofcbangeinthisstudy,includingPercentageofConso-nantsCorrect,didnotrevealanysignificantdifferencesinoutcomesbetweentbetwogroups
of participants. Therefore, it is nol
sur-prisingtbattbemostappropriateinteqire-tationoftbesefindingsisamatterofcon-tinuingdebate(Morrisette&Gierut,2003:Rvaehew
&Nowak, 2003), However, there isno doubt that, in this study,
whichinvolved atraditional treatment approach, children
weremostlikely tomaster tbetargetedconsonantwben it was stimulable
(fiirther studies are
re-quiredtodeterminewhetherthissameout-comewouldbe acbievedin the
contextof aphonologicalinterventioninvolvingminimalpairs
procedures).ParentsandcliniciansweremostsatisfiedwiththeoutcomeoftheME
targetselectionstrategy. Targeting stimuiable sounds first
didnotpreventtbechildrenfrommakingspon-taneousimprovementstowardcorrectpro-ductionof
other more difficultphonemes; infact,tbeME
groupshowedgreaterprogressforuntreatedunstimulablepbonemesthandidtheLLgroupfortreatedunstimulablephonemes.Absolutelynoimprovementwasobservedfor45%
of treatedunstimulable tar-gets. For these reasons, Rvacbew and
Nowak(2001) concluded that the most prudent
strat-egyformostchildrenistotargetstimulablephonemesfirst.Ontheotherhand,10%
ofdiildrenintheLLgroupactuallymasteredatreated unstimulable target
sound,
Stimulabil-ityshouldneverbetheonlyvariableunderconsiderationwhen
selecting tbe most appro-priate pbonemes to
remediate.EFFECTIVETREATMENTOFUNSTIMUIABLEPHONEMESThere are many
cases whenit is necessaryto target a phoneme for wbicb tbe child is
un-stimulable,Tbephonemethathasthemostimpactonintelligibilit>-
may beunstimulabiefor example. If an entire class of phonemes
ismissingfromthechild'srepertoire,youwillbe forcedto introduce one
or more membersof thisclass evenif thechildis
unstimulableforalloftherelevantphonemes.Therefore,itisimportanttoidentifytreatmentproce-duresthatwillresultinbetteroutcomesforunstimulablepbonemesthanthoseobservedby
Rvacbew et al. (1999) and by Rvacbew andNowak (2001). To follow is
a summary of two.studiessbowingthatphonemicperception214TOPICS IN
LANGUAGEDISORDERS/JULY-SEPTEMBER
2005interventioncanfacilitatetheacquisitionofstimulabilityandeventualmasteryoftargetsounds.The
frameworkshownin Figure1suggeststhatacquisitionof a given
phonemerequiresaccuratephonemicperceptionofthetar-geted
phonemecontrastand stimulabilityforthe target phone. Rvachew (1994)
conducteda randomized control trial to test this hypoth-esis. In
this study, six
once-weeklytreatmentsessionswereprovidedto27preschoolerswithspeechsounddisorders,allofwhomwerecompletelyunstimulablefor/J/.Eachtreatmentsessionforallchildrenconsistedof10minofphonemicperceptiontrainingimplementedinacomputer-gameformat,followedby20minofarticulationtherapydirectedatthe/J/phoneme.Thetreatmentapproach
during the articulation therapy partof the sessions was
traditional, withphoneticplacement, shaping and modelling
techniquesusedtoestablish.stimulabilityattheisola-tionandsyllablelevels.Ifstimulahilitywasachieved,thetreatmentprogramprogressedinstepsfromimitatedwordsthnnighprogressively
longer units of speech. All
childrenbeganthetreatmentprogramatthestimula-tion phase and
progressed from one treatmentstep to thenextuponachieving 90%
correctresponding.Thephonemicperceptiontrain-ing procedurewas
alsothesame forall chil-dren: the child was asked to listen to
recordedwordsandidentifythosewordsthatweregoodexemplarsofaparticulartarget.Thestimuli
that the children listened to dependedonthechild
sgroupassignment,howeverThe experimentalgrouplistenedto a
varietyofexemplarsoftheword"shoe"recordedfromadults and children,
with half the wordsbeing correctexemplars and the other half ofthe
words representing commonmisarticula-tionsofthisword (e.g., [tu],
[SU],[^U]). Whenthe 'i:hild accurately identifieda word as
being"shoe"or "not shoe," interesting cartoon
char-actersappearedonthecomputerscreentoreinforce the children for
correctresponding.Thisstudyinvolvedtwocontrolgroups.Thefirstcontrolgrouplistenedtorepeatedpresentationsof
a
singleprototypicalrecord-ingoftheword"shoe"andasingleprototypicalrecordingoftheword"moo"
andrespondedas describedabove
fortheexper-imentalgroup.Thiscontrolgroupwasin-cluded to confirmthe
importance of
present-ingchildrenwithavarietyofexemplarsofthetargetphoneme,as
opposedtopresent-ing only perfector prototypical exemplars ofthe
target
sound.Thesecondcontrolgrouplistenedtothewords"cat"and"Pete"producedcorrectlybya
singleadulttalker.Thesechildrenwereexpectedtoidentifycorrectversionsoftheword"cat."Theclinicianwho
conductedthearticulationtherapywasblindtothechild'sgroup
assignment. This control condition wasnot expectedto impact the
children's speechperception skills. Rather, it provided a
controlfor other aspects of the
phonemicperceptiontrainingprocedure(e.g.,extratherapytimeandindividualattention,theopportunitytoplay
a funcomputer game).Outcome assessments were also conductedby a
blind observerOn average, the
childrenintheexperimentalgroupachievedstimula-bilitj'andthenlearnedtoproducethe/J/phonemeatthelevelof
spontaneous words.In contrast, only one child in the second
con-trolgroupachievedstimulabilityattheiso-latedsoundlevel.
Theremainingchildreninthisgroupexperiencednogainevenafter6weeksoftherapy.Theperformanceofthegroupthatheard"shoe"
andmoo" wasin-termediatebetweentheexperimentalgroupand the second
control group. Percent correctrespondingona
spontaneousobjectnamingprobe also confirmedthat the phonemic
per-ception training procedure facilitated the chil-dren s
acquisitionofthisdifficultphoneme.Furthermore,presentingthechildrenwith
avarietyofgoodqualityandlesserquality ex-emplars of the target word
producedby
mul-tiplechildandadulttalkerswasparticularlyeffeetive.Subsequently,a
commercialversionofthephonemicperceptiontrainingprogramthatwas
administeredtotheexperimentalgroupinRvachew(1994)wasdevelopedandex-pandedtocover
additionalphonemesintheStimulahilityandTreatment
Success215wordinitial and word final positions
(SpeechAssessmentandInteractiveLearningSystem[SAII^];AVAAZ
Innovations,Inc.,1994).
Arecentmndomizedcontrolledtrialdemon-strated that this program
enhanced sound pro-duction learning by children receiving
phono-logicalinterventions(Rvachew,Nowak,&Cloutier,2004). In
thisstudy,34 pre-school-age childrenwith moderate or severe
speechsounddisorderswererandomlyassignedtoan
experimentaltreatmentprogramor a
con-trolprogramthatwasofferedasasupple-ment to their regular speech
therapy
sessions.Eachchildreceived16once-weeklyspeechtherapysessionsinwhichthechild'sclini-cianselectedtreatmenttargetsandprovidedthe
interventions thatshefelt were
appropri-ate.Aftereachofthesesessions,thechildreceivedanadditional10-mininterventionthatwas
administeredhya
studentresearchassistant.Thechildrenassignedtotheex-perimentalconditionreceivedthe
SAILSpro-gram targeting eight consonants. The
childrenassignedtothecontrolconditionlistenedtocomputerizedbooks.Again,treating
clini-cianswereblindtothechild s groupassign-ment as was the
speech-language
pathologistwhoconductedtheoutcomeassessments.Theposttreatmentassessmentrevealedthattheexperimentalgroupdemonstratedsignif-icantly
greater articulation accuracyin
single-wordnamingandinconversationthandidthecontrolgroup.Furthermore,afollow-upassessment1yearlater,whenthechildrenwere
approaching the end of kindergarten, re-vealedthat50%
ofexperimentalgroupchil-drenhadachievedage-appropriatearticula-tion
skills whereas only19% of
controlgroupchildrenhadachievedarticulationskills thatwould be
consideredtj'pical.One of the outcome measures was Percent-age of
Consonants Correct in conversation forspecific phonemes that were
not mastered bythe children prior to treatment, calculated
be-foreandaftertreatment.Ihechildrenwhoreceived the
SAILSprogrammade greater im-provements in articulation accuracy for
thesephonemesthanchildrenwhowereassignedtothecontrolcondition.This
benefittotheSAILS
groupwasobservedfor11ofthe13difficultsoundsthatweremonitoredinthisstudy
(see Figure 1of Rvachew et al.,
2004).AlthoughthestudybyRvachewetal.(2004) was nt)t directly
concernedwith stim-ulability, details of the pretreatmentand
post-treatmentconversationsamplessuggestedthatthiscombinationof
articulationtherapyandphonemicperceptiontrainingenhancedoutcomesforunstimulablephonemes.Figure
3 shows the change in
percentcorrectarticulationofcertainconsonantsfromthepretreatmenttotheposttreatmentassess-ment,asdescribedabove.However,inthiscase,
the changes in articulation accuracy
areshownasafunctionofpretreatmentinven-torystatus.In otherwords,
the figure showschange in articulation accuracy separately
forphonemesthatwerepresentinthechild'sinventor}' prior to
treatment(and thus clearlystimulable)andforphonemesthatwerenever
producedin the child s inventory priorto treatment (and thus not
likely to have beenstimulable). Figure3a showsthatchildren
intheSAILS
groupmadea27%improvementin/J/accuracyregardlessofwhetherthisphoneme
was presentin the child's
pretreat-mentconversationalsample.Childreninthecontrolgroupwhowerestimulablefor/J/made
more than twice the gain in
productionaccuracyforthisphonemethandidcontrolgroupchildrenwhoneverproducedthissoundpriortotreatment.Theresultsfor/e/andA/indicatedthatchangefromnocorrectproductionstoatleastsomecorrectproductionsoccurredonlyforchildrenwhoreceived
the SAILS
intervention.CONCLUSIONSl\voquestionswereaskedabouttbetreatmentofstimulableandunstimulabiephonemetargets:(1)Whenselectingtreat-ment
targets, is it best to begin with the
mostortheleaststimulablepotentialphonemetargets?(2)Whentreatingunstimulablephonemes,whichtreatmentproceduresareassociatedwiththehighestprobabilityof216TOPICS
IN LANGUAGEDISORDERS/JULY-SEPTEMBER2005a.504540353015105027271///
in inventory|D Ijlnot in inventory12^^M5SAILS
Control50454035?30n025120I15105038^ ^ ^ 114IQl in inventoryD IQl
not in inventory12^ ^ ^ 10SAILS Control50454035o?3020151025Ai/in
inventoryn/j/notin inventorySAILS ControlFigure3.Frf
treatmenttoposttreatmentimprovementsinpercentcorrectproductionofcertainconso-nantstorchildrenwhoreceivedtheSAILSphonemicperceptiontrainingprogramversuschildrenwhowereassignedtothecontn>lgroup,asafunctionofpretreatmentinventorystati:s.(BasedondatafromRvachewetaL,
2(M)4).StimulabilityandTreatmentSuccess217achievingaccuratearticulationofthenewphoncmt?Withregardtothefirstquestion,anindoniizedcontrolledtrialledtothecon-clusionthattreatmentofthemoststimulahlepotentialtargetsis
likelytoresultinagreaterrateofchangeforthetreatedphonemethanistreatmentoftheleaststimulablepotentialtargets.
The results of this studyalsosuggestedthata
targetselectionstrategythatbeginswiththemoststimulableandearliestdevelopingphonemeswillfacilitatethespontaneousemergenceofunstimulablephonemes.However,thesefindingsreflectonlyasinglestudy.Greaterconfidenceintheseconclusionscouldcomefromreplica-tions
and extensions of this study in
additionalrandomizedcontrolledtrials.Studiesthatex-amineothercombinationsoftargetselectioncriteriawouldbevaluable.Itispossiblethatthemosteffectivetargets,intermsofpromotingsystemwidechange,mightbethosephonemesthatarestimulablebutrarelyproducedcorrectly(inotherwords,phonemesrepresentingType5ProductivePhonologicalKnowledge).Otherpossibleidealtargetsmightbeunstimulablebutearlydevelopingphonemesorstimulablelatede-velopingpbonemes.Thesehypotheseshavenotyetbeeninvestigatedexperimentally.Withregardto
thesecondquestion,a
num-berofstudieshavesbownthatacombina-tionotphonemicperceptionandstimtila-bilitytrainingenhanceschildren'sresponsetotreatmentClamieson&Rvachew,1992;Rvacbew,1989,1994;Rvachewetal,,1999,2004),Thisenhancementofsoundproduc-tionlearningisatleastasdramaticforun-stimulablepbonemesasitisforstimulablephonemes.TheDIVAmodel(Callanetal,.2000)suggeststhattbisenhancementoc-cursbecausetbeSAILSphonemicpercep-tiontrainingprocedureprovidesthechildwithastableacoustic-phonetictargetforthephonemebeinglearned.Thisunderlyingacoustic-phonetictargetallowsthechildtodiscoverthe
mappingbetweentbephoneme,theappropriatearticulatorygestures,andtheacoustic-phoneticoutcomeof
thosearticula-torygestures.Thisspecifichypothesisabouttbemecha-nismbywhichtheSAILS
programenhancessoundproductionlearninghasnotbeenex-perimentallyconfirmed,however.Otherex-planationsarepossible.Thetreatmenteffectmaybeduetoamoregeneralizedprocesswherebythe
childlearns toattendmore
care-fullytospeechinput.Alternatively,tbepri-marylessonthattbechildlearnsmaybetomakejudgmentsabouttbeaccuracyofhisorberownspeech.Treatmentapproacliesthatrequirethechildtoself-monitormaybemoreeffectivethantraditionalclinician-monitoring
procedures,regardless of whetherthe SAILS programis a componentof
tbether-apy. Furthermore, these studies of tbeefficacyoftheSAILS
programinvolvedeitheratradi-tional behaviorist approach or a cycles
phono-logicalprocessapproachtotreatment(e,g,,Hodson&Paden,1983).TbeefficacyoftheSAILS
phonemicperceptiontrainingprogramhasnotbeenevaluatedincombinationwitba
morelitiguisticapproachinvolvingminimalpairs procedures (eg,,Dean
fit Howell,1986),Neither has the SAILS
treatmentprogrambeenevaluatedin relationto tbe
multiplephonemeprogramdesignedbyMicciotofacilitateacquisitionofstimulability(Miccio,2005;Miccio&Elbert,1996),Furtberstudiesre-gardingtheefficacyofphonemicperceptiontraining
to
facilitatetheacquisitionofunstim-ulablepbonemesarerequired.Itmaybefrustratingtocliniciansthatre-searchersalwaysconcltidetbatmorere-searchisrequired,"Thisinevitableconclu-sion
does not meanthat tbe available
researchcannotsupportclinicaldecisionmaking.
Ran-domizedcontrolledtrialsdoprovidegoodinformationabouttbelikelyoutcomeofagiventreatmentpracticetbragivenpopula-tionofclients.Tbespeech-languagepathol-ogistmustmakeajudgmentabouttheex-tenttowhichtheresearchfindingsapplytoaspecificchildinaspecificclinicalcontext.The
clinicianshould also engage tbe child
andthefamilyinadiscussionaboutthepoten-tialbenefitsandrisksassociatedwiththeap-plicationofagivenclinicalpractice.There-searchreviewedheresuggeststhatthereis218TOPICSINLANGUAGEDISORDERS/^ULY-SEPTEMBER2005averystrongtiskofachievingnogainaftersixormoreweeksoftherapywhenunstim-ulabletargetsaretreatedusingabehavioristapproach.However,there
may bemanysitua-tions in which the potential
benefitsoutweightheknownrisk.Thechoiceoftreatmentap-proachfor a
given child is ultimatelyup to theclinician.REFERENCESAVAAZ
Innovations,Inc. (1994).
SpeecbAssessmentamiInteractii'et.eamingSystem(Version1.2).London,Ontario,Canada:Author.Bleile,K.{2(M>2).Evaluatinganiculationandphonolog-icaldisorderswhentheclockisrunning,Ameri-can
Journalof Speech-LanguagePathology.11,
243-249.Callan,D.E,,Kent,R.D.,(iuenther,EH.,&Vorperian,H.K.
(2(XX)). Anauditory-feedback-basedneuralnet-workmodel of
speechproductionthatis robustto
de-velopmentalchangesinthesizeandshapeofthear-ticulatorysystem,/oHfTJtf/ofSpeech,Language,andHearingResearch.4X721-738,Dean,E.,&Howell,J.(1986),Developinglinguisticawareness:A
theoreticallybasedapproachtophono-logicaldisorders.BritishfoumalofDisordersofCommunication,31,225-238.Gierut,
J,A.,Elbert,M,, & Dinnsen,D, A. (1987). A
func-tionalanalysisofphonologicalknowledgeandgener-alizationlearninginmisarticulatingchildren.
/ouma/of SpeechandHearingResearch,JO, 462-479.Gierut, J. A.,
Morrisette, M. L, Hughes, M. T, & Rowlands.S.
(2()()l).Phonologicaltreatmentefficacyanddevel-opmental norms,
language.Speech, andHearing Ser-t'ices
inSchools,27.215-230.Guenther,EH.(1995).Speechsoundacquisition,coar-ticulation,andrateeffectsina
neuralnetworkmodelof
speechproduction.PsychologicalReview,I02Q>),594-621.Guenther,EH.,Husain,ET.,Cohen,M.A.,&ShinnCunningham,B.G.(1999)-Effectsofcategorizationanddiscriminationtrainingunauditoryperceptualspace,
foumaloftheAcousticalSocietyofAmerica,106,2900-2912.Hodson,B. W.,
& Faden,E, P(1983).TargetinginteUigi-ble speech: A
pbonologicalapfnoacbtoremediation.Boston:
CollegeHill,Jamieson,D,G,,&Rvachew,S,(1992).Remediationofspeechproductioncmirswithsoundidentificationtraining,
you rruj/of
Speech-LanguagePathologyandAudiology,16,201-210.Kwiatkowski,J,,&Shriberg,LD.(1993).Speechnor-malizationindevetopmenUilphonologicaldisoRlers:A
retnjspectivestudyofcapability-focustlieory.Lan-guage,Speech,andHearingServicesinSdmols,24,10-18.Locke,
J.L. (1980). Theiniierenceofspeechperceptionin the phonologicalty
disordered child. Some clinicallynovel pniccdurcs,
theirust,.somefindings,yoni