www.bjorl.org
Brazilian
Journal
of
OTORHINOLARYNGOLOGY
REVIEW
ARTICLE
Biofeedback
in
dysphonia
---
progress
and
challenges
夽
Geová
Oliveira
de
Amorim
a,∗,
Patrícia
Maria
Mendes
Balata
b,
Laís
Guimarães
Vieira
c,
Thaís
Moura
a,
Hilton
Justino
da
Silva
a aUniversidadeFederaldePernambuco(UFPE),Recife,PE,Brazil b
HospitaldosServidoresdoEstadodePernambuco,Recife,PE,Brazil c
LGVAssessoriaeConsultoriaMédica,Recife,PE,Brazil
Received11January2016;accepted15July2017 Availableonline19August2017
KEYWORDS Speechtherapy; Voice; Dysphonia; Electromyography feedback Abstract
Introduction:Thereisevidencethatallthecomplexmachineryinvolvedinspeechactsalong withtheauditorysystem,andtheiradjustmentscanbealtered.
Objective:Topresenttheevidenceofbiofeedbackapplicationfortreatmentofvocaldisorders, emphasizingthemuscletensiondysphonia.
Methods:AsystematicreviewwasconductedinScielo,Lilacs,PubMedandWebofSciences databases,usingthecombination ofdescriptors,andadmittingasinclusioncriteria:articles published in journals with editorial committee, reporting cases or experimental or quasi-experimentalresearchontheuseofbiofeedbackinrealtimeasadditionalsourceoftreatment monitoringofmuscletensiondysphoniaorforvocaltraining.
Results:Thirty-threearticleswereidentifiedindatabases,andsevenwereincludedinthe qual-itativesynthesis.Thebeginningofelectromyographicbiofeedbackstudiesappliedtospeech therapywerepromisingandpointedtoanewmethodthatenabledgoodresultsinmuscle ten-siondysphonia.Nonetheless,thediscussionoftheresultslackedphysiologicalevidencethat couldserveastheirbasis.Thesearchforsuchexplanationshasbecomeachallengeforspeech therapists,anddeterminedtworesearchlines:onededicatedtotheimprovementofthe elec-tromyographicbiofeedbackmethodologyforvoicedisorders,toreduceconfoundingvariables, andtheotherdedicatedtotheresearchofneuralprocessesinvolvedinchangingthemuscle engramofnormalanddysphonicpatients.
夽 Pleasecitethisarticleas:AmorimGO,BalataPM,VieiraLG,MouraT,SilvaHJ.Biofeedbackindysphonia---progressandchallenges.Braz
JOtorhinolaryngol.2018;84:240---8.
∗Correspondingauthor.
E-mail:[email protected](G.O.Amorim).
PeerReviewundertheresponsibilityofAssociac¸ãoBrasileiradeOtorrinolaringologiaeCirurgiaCérvico-Facial. https://doi.org/10.1016/j.bjorl.2017.07.006
1808-8694/©2017Associac¸˜aoBrasileiradeOtorrinolaringologiaeCirurgiaC´ervico-Facial.PublishedbyElsevierEditoraLtda.Thisisanopen accessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).
Biofeedbackindysphonia---progressandchallenges 241 Conclusion: Thereisevidencethattheelectromyographicbiofeedbackpromoteschanges in theneuralnetworksresponsibleforspeech,andcanchangebehaviorforvocalemissionswith quality.
© 2017 Associac¸˜ao Brasileira de Otorrinolaringologia e Cirurgia C´ervico-Facial. Published by Elsevier Editora Ltda. This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/4.0/). PALAVRAS-CHAVE Fonoaudiologia; Voz; Disfonia; Feedbackde eletromiografia
Biofeedbackemdisfonia---progressoedesafios
Resumo
Introduc¸ão: Háevidênciasdequeocomplexoaparatoenvolvidonoatodafala, juntamente comosistemaauditivo,eseusajustespodemseralterados.
Objetivo: Apresentar evidências da aplicac¸ão de biofeedback (biorretroalimentac¸ão) para tratamentodedistúrbiosvocais,enfatizandoadisfoniadetensãomuscular.
Método: Realizou-seumarevisãosistemáticanasbasesdedadosdeScielo,Lilacs,PubMede WebofSciences,utilizandoacombinac¸ãodedescritoreseadmitindocomocritériosdeinclusão: artigospublicadosemrevistascomcomitêeditorial,relatosdecasosoupesquisasexperimentais ouquaseexperimentaissobreousodebiofeedbackemtemporealcomo fonteadicionalde monitorac¸ãodetratamentodedisfoniadetensãomuscularouparatreinamentovocal. Resultados: Trintaetrêsartigosforamidentificadosembasesdedados,eseteforamincluídos nasíntesequalitativa.Oiníciodosestudosdebiofeedbackeletromiográficosaplicadosàterapia fonoaudiológicaforampromissoreseindicaramumnovométodoquepermitiubonsresultados nadisfoniadetensãomuscular.Noentanto,adiscussãodosresultadoscareciadeevidências fisiológicasquepudessemservirdebase. Abusca portaisexplicac¸õestornou-se umdesafio para os fonoaudiólogose determinouduaslinhasde pesquisa: umadedicada àmelhoria da metodologiadebiofeedbackeletromiográficoparadistúrbiosdavoz,parareduzirasvariáveisde confusãoeoutradedicadaàpesquisadeprocessosneuraisenvolvidosnaalterac¸ãodoengrama musculardepacientesnormaisedisfônicos.
Conclusão:Háevidênciasdequeobiofeedbackeletromiográficopromovemudanc¸asnasredes neuraisresponsáveispelafalaepodemudarocomportamentoparaemissõesvocaiscom qual-idade.
© 2017 Associac¸˜ao Brasileira de Otorrinolaringologia e Cirurgia C´ervico-Facial. Publicado por Elsevier Editora Ltda. Este ´e um artigo Open Access sob uma licenc¸a CC BY (http:// creativecommons.org/licenses/by/4.0/).
Introduction
Oral communicationdependsonthe abilitytoproducean enormous variety of sounds that together characterize a language.Theproductionofthesesoundsresults inhighly specificconfigurationsofthevocal tract,whichfilters the soundproducedinthelarynxandmodulatesitthrough coor-dinatedmovementsofthelips,tongueandmandible.1These interactionsinvolvethemusclesandtheautonomicand cen-tral nervous system and are intriguing topic of research involvingnormalsubjectsandthosewithvoicedisorders.
There is evidence that all the complex machinery involved in speech acts along with the auditory system, which enables the individual to develop natural dynamic adjustmentsoveraspeech,inresponsetoperceptiblenoise pollution,tomakethevoicemoreharmoniousandadequate tothespeech.2Theseadjustmentsnotalwaysaresufficient tocompensatethevoicedisorder.
Dysphoniaconstitutesabout40% oftheamendmentsin voice clinics, and the other 60% of the amendments are represented by voice enhancement requests, concerning
todeviation from the perceptive properties of the voice that drawattention to the speaker, that means esthetics of voice.3 Dysphonia is a communication disorder, repre-sentedbyanydifficultyinvocalemissionthatpreventsthe voicefromfulfillingitsbasicroleoftransmittingverbaland emotionalmessageofan individual,3 dueto. imbalanceof themusclesresponsibleforvoiceproduction.4Thealtered vocalpatterns,asaconsequenceofdysphonia,resultfrom muscular adjustments, at glottic and supraglottic levels, developedby thesubject, in responseto thepresence or notoflaryngealinjury.5Indysphoniabymuscletension,itis assumedthattheincreasedtensionoftheextrinsicmuscles ofthelarynxelevatesthelarynxintheneckanddisturbsthe slopeofthelaryngealcartilages.6Consequently,the extrin-sicmusclesofthelarynxareaffectedandchangethetension ofthevocalfolds,determiningthevocaldisorder.7
Thecerebralcortexistheoriginofvolitionalspeech con-trol.Nerveimpulsesarepropagated toactivatethemotor nucleus of the brainstem (ambiguous nucleus, nucleus of the solitary tract and parabrachial core) and the spinal cord,modulatingtheaction of laryngeal muscles, aswell
asarticulators,thoracicandabdominalmuscles.The conse-quencesofthisactivationinvoicedisordersarenewmental representations of the phonatory behavior, which will requireatherapeuticprocessforrelearningthefunction.8
The various types of vocal quality are produced by changesinlaryngealkinesiology,whoseintrinsicand extrin-sic muscles can be in imbalance. This abnormality points out the need to speech diagnosis in order to establish a treatment plan.The non-instrumentaldiagnostic methods includevocal history,todetermineabuseor misuseofthe voice;investigationofstressandpsychologicalfactorsthat canalterthevocalproduction,aswellaspalpationofthe neck,determiningmuscletensionandelevationofthe lar-ynx,withthepossibilityofquantificationwithafourtofive pointsLikert scale. These methods, although widely used in clinicsand research, have low diagnostic power, being judge-dependents.7
Theinstrumentaldiagnosticmethodsinclude videolaryn-goscopy,radiography,needleor surface electromyography and, more recently, although still experimental, func-tionalmagneticresonanceimaging.7,9,10Radiographyallows to document and assess the increase in anteroposte-rior muscle contraction, and the presence of nodules, polyps and cysts, but has low diagnostic power for voice disorders.
Electromyography(EMG)enablesevaluationofthe elec-tricalactivityofthelaryngealmusclesandthereforeofthe tensionofmusclefibersandthe integrityof thelaryngeal nervoussystem.11Additionally,electromyographywith elec-trodesplacedontheskin(calledsurfaceEMG---sEMG)canbe usedinoperantlearningforinternalautonomicmodification ofabehaviorthroughconcurrentinformationtoatask.The sEMGallowstheindividualtoobservethephysiological pro-cess,inrealtime,contributingtobehavioralchange,i.e., toestablishcorrection,meetingobjectivesclearlydefined bythephysiologyofthealteredorganfunction.
Thepathophysiologyofmuscletensiondysphoniaisnot yetfully elucidated.Somehypothesis admitthat the ten-sionin extrinsicmuscles move the larynxup in theneck, whichaltersthe slopeof laryngealcartilages.Asa result, theintrinsicmuscles ofthe larynxareaffectedandtense thevocalfolds,promotingdisphonya.7
Inthepresenceofdysphonia,theautonomicandthe cen-tralnervous system record the sound change and trigger accommodation to reduce the disturbances. This adap-tivephysiologicalmechanismcanbemonitoredandviewed throughprogramsthatconverttheregistrationof proprio-ceptiveor tactile receptors in imagesin a process called biofeedback.12
Biofeedbackoccursduringvocalemission,enablingthe issuer to make adaptations for correction of vocal disor-ders.In dysphonia,thesecorrectionscanalsocallfor the speaker’sattentiontotheintensityofthedisorder,because thereis a natural tendency tounderestimateit. Thus, in thebiofeedback,internalautonomicevents,suchasmuscle tension,areelectronicallyamplifiedallowingtheindividual (bio)toreceivetheinformation(feedback),whichisusually unavailable.12
Several studies13---18 have demonstrated the possibility ofbiofeedback tohelpindividuals modifying physiological patterns,admittingthatanindividual,whoknowsthe func-tioningofhisbody,canpromoteoneselfchanges.
Weidentifiedthatdysphonicindividualsmayhave diffi-culty in changingthe divertedvocal standard, asa result of the engram of the muscle activity that automated it, whereas mental processes are subject to changes from inputsofvisual,auditoryandkinestheticnature,provided byfeedbackchannels.12,17,19Thesefactsrouseourinterest instudyingdysphonia,intheviewofneuroscience.
The objective of this systematic review is to present the evidence of biofeedbackapplication for treatment of vocaldisorders,emphasizingthemuscletensiondysphonia, aswellasforvocallearningofprofessionalsingers.
Methods
SystematicreviewwasconductedinthedatabasesofScielo, Lilacs, PubMed and Web of Sciences, using the combina-tion of descriptors <feedback>, <voice>, <biofeedback>, <dysphonia>,<tension>,<therapy>, <surface electromyog-raphy>,<singers>.Basedoninformationfromtheabstracts and titles, potentially useful articles were evaluated in thereview.Itwereappliedtheinclusioncriteria,namely: articles published in journals with editorial committee, reporting cases or experimental or quasi-experimental researchontheuseofbiofeedbackinrealtimeasadditional sourceoftreatmentmonitoringofmuscletensiondysphonia or for vocaltraining.Itwasnotset anyrestrictiononthe studydesign,language,biofeedbackmethodor methodof assessment,duetothescarcityofstudies.
Twojudges(G.A.andP.M.B.)independentlyreviewedthe titlesandabstractsofthelocatedarticlesanddetermined those with potential tointegrate the review, from which thefulltextswereobtainedforfurtherexaminationofthe judges. At this stage, each of them selected studies that metthedefinedgoals,andthedissentingjudgmentswere resolvedbyconsensus.
While recognizing that the meta-analysis is the gold standard for review, in this article it has not been possi-bletoadoptthismethodologygiventhescarcityofstudies andthesmallnumberofelementsineachstudy.
Results
Thirty-three articles were identified in databases, but 16 wereexcludedalreadyintheanalysisoftitlesandabstracts: four by using biofeedback without surface electromyog-raphy; three for being written in German; six systematic reviews;oneforshowinganalysisofthecortexinthevoice articulation;andtwofor havingsEMGasthemaintheme, withoutbiofeedbackapproach.
Seventeenarticleswerelisted bythe judgesfor analy-sis of the full text,from which resulted the exclusion of sixfullarticleswithinaccuratemethodologicaldescription, leavingelevenarticlesforevaluationofeligibility.Thefull reading of thetext resulted in the exclusionof four arti-cles, one by approaching functional dysphonia associated withother illness,differentfromdysphonia,andthreeby investigatingpopulationswithoutdysphoniausingfunctional magnetic resonanceandvocalbiofeedback.Sevenarticles wereincludedinthequalitativesynthesis(Fig.1)(Table1).
Biofeedbackindysphonia---progressandchallenges 243
Identified studies in databases (PubMed = 10, EBSCO=4; SCOPUS=20)
(n = 33)
Removed articles (n = 17)
Selected articles (n = 11)
Complete articles to eligibility evaluation (n = 11)
Included articles in the qualitative synthesis (n = 7)
Removed articles Use of biofeedback without sEMG (n=4)
Written in German (n=3) Systematic reviews (n=6) Cortex analysis of voice articulation (n=1)
sEMG as main theme (n=2)
Exclusion of complete articles (n=6) for inaccurate methodological description
Exclusion (n=4)
Approach of functional dysphonia associated to other injuries (n=1) Use of functional magnetic resonance for
other populations (n=3)
Identification
Screening
Elegibility
Inclusion
Figure1 Flowchartoftheselectionofarticlesforsystematicreview.
Discussion
We decided to divide the articles into two groups, one withbiofeedbackapplicationfor correctionofmuscle ten-siondysphoniaorvoicedisorderssecondarytothelaryngeal muscletension,12,20---22andanotherdedicatedtothe appli-cationof thismethodin thevocallearningofprofessional singers.2,10,16
The twogroupsofarticlesweremaintainedto contem-platethebiofeedbackemploymentopportunitiesinspeech therapy,giventhattechnologicaladvancesinimaging stud-iesenableddemonstrationofnewevidence,eitherforvocal training,ortotreatmentoftensionalvocaldisorders.
Biofeedbackforcorrectionofmuscletension dysphoniaorvoicedisorderssecondarytomuscle tension
The vocal disordershave been studied for a long timeto determinetheircausesaswellasforthedefinitionof effec-tive therapies. Allen, Bernstein and Chait20 conducted a studywithelectromyographicbiofeedbackonamalechild,8 yearsold,withhyperfunctionaldysphonianotresponsiveto vocalrehabilitationandrelaxationtechniques.Thepatient hadvocal nodulesandsuchanintensemuscletensionthat therewasindicationforsurgery.
The child was subjected to electromyographic evalu-ations of the larynx muscles tension, and of the voice quality, performed by a professional and the child’s par-ents.Thetreatmentconsistedofa30-minutesessiontwice aweek,withvisualbiofeedbacktraining,followedby5min ofhabituation.Ateachsession,itwereheld20consecutive exercises, first during vocal rest and then during speech, always trying to reduce muscle tension,by observing the
feedbackoflaryngealmuscles.Theinitialtensionreduction criterionwas0.5V,invocalrest,and5V,duringspeech, inatleast80%ofexercisesinthethreeconsecutivesessions. Thetherapycompletioncriterionwasbasedonthenormal baselinesforadults,asforchildrentherewerenonormative data.Followingthreeand6months,muscletensionatrest decreasedfrom7.5Vto3.2V;atspeech,from49.5Vto 10V,valuethatwasmaintainedafter6months.20
The differential of this study20 was the evaluation of voicequalitybythechild’sparents,featuringsocialvalidity ofthetreatment,i.e.,utilityandmeetingoftheparents’ expectationsfortheresults obtainedwithtreatment, giv-ing a value with social applicability.23 The social validity requires comparing two basic strategies. The first is the objectiveevaluationbasedonnormativedata.Thesecond issubjective, based onthe perception of an expert or of people accustomed to the event under evaluation.23 The child’sparents,inthisexperiment,pointedtothevalidityof thetreatmentbecauseofthegeneralizationoftheeffects onthevoice,intheresidenceandother non-clinical sett-ings,indicatingstabilityofthereductionofvocaltension. Anotherimportantfindingofthisresearchwastheevidence ofadrasticreductionofvocalnodules,dispensingthe sur-gicaltreatmentpreviouslyscheduled.
Thesecondstudyofthisgroup22 describesthecaseofa 26year-oldmalewithahistoryof tensiondysphoniaafter severelaryngitisandviralinfluenza,withsubsequent func-tionaldysphoniaofthevestibularvocalfold,6monthsago. Inotorhinolaryngologicconsultation,it wasidentifiedthat the patient made use of analgesic and anti-hypertensive medicationthatdidnotpromoteoraffectdysphonia.
Treatment was established with electromyographic biofeedback consisting of two weekly sessions, lasting 20---30min,witha10minhabituation,inaroomwithsound attenuation,1---20repeatedcountsandcasualconversation,
Amorim
GO
et
al.
Table1 Descriptionofthestudiesincludedinthisreview.
Authors Date Objective No.ofparticipants(age) Positionoftheelectrodesinthe biofeedbackwithsEMG
Results AllenKD, BernsteinB, ChaitDH5 1991 Reductioninlaryngeal muscletension
1(8years) Ipsilateralandvertically,inparallel alignmentonthethyroidmembranewith earth-electrodeplacedonthewrist
Socialvalidityofreducedtensionin laryngealmusclesandvocalnodules, withprevioussurgicalindication WatsonTS,
AllenSJ, AllenKD14
1993 Reductioninmuscle tensioninvestibularfold dysphonia
1(26years) Ipsilateralandvertically,inparallel alignmentalongthemajoraxisofthe thyroidmembranewithearth-electrode placedonthewrist
Disappearanceofsymptomsand persistenceofresultsinthe evaluation,after6months Ruscello11 1999 Reductionin
phonologicalerrorsof acquisitionand automation
4(9,10,14,29years) Biofeedbackwithnasalairflowcontrol Biofeedbackself-administeredunder supervision,havingasastandardthe correctpronunciationbyaninstructor Biofeedbackself-administeredunder supervision,withrecordingofthepatient’s pronunciation,inusualandcorrectspeech articulation
Socialvalidityofarticulatory relearningandreducedlaryngeal muscletension
WarnesE,Allen KD19
2005 Reductioninparadoxical vocalfoldmobilityand respiratorydistress
1(16years) Ipsilateralandvertically,inparallel alignmentalongthemajoraxisofthe thyroidmembrane,usingthethyroid cartilageasadditionalanatomicalmarker
Socialvalidityofreducedlaryngeal muscletension,reducedchestpain andcripplingrespiratorydistress, andpersistenceofresultsinthe evaluationafter6months YiuEM-L, VerdoliniK, ChowLPY9 2005 Learningofvocal productionwith laryngealrelaxation Definitionofthebest conductinbiofeedback (concurrentorterminal)
5(21---27years) Electrodespositionedsymmetricallytothe midlineatfourpositions:(a)1cmfromthe lipcommissure,(b)0.5cmfromthemidline ofthechin,(c)1cmfromthemandibular midline,(d)0.5cmfromthemidline,on thethyroidmembrane,and(e)an earth-electrodefirmlyfixedonthewrist
Absenceofbiofeedbackcontribution forvocallearningwithrelaxationof laryngealmuscles,butpresenceof orofacialrelaxation,attributedto thepossibledifficultyofnormal individualstoreducelaryngeal muscletension
KirkpatrickA, McLester JR15
2012 Learningofrelaxationof thelaryngealdepressor musclesforimproved voicequalityofsingers
22(>21years) Bilaterallyandbeforethethyroidcartilage, justabovethelaryngealprotuberance
Biofeedbackhelpsinactivatingthe depressormusclesofthelarynx, increasingtheamplitudeandtonal quality
NiziolekCA, GuentherFH2
2013 Effectofthevocal boundariesin
compensatoryresponses toauditorydisturbances inrealtime
36(19---33years) Visualbiofeedbackwithfunctional magneticresonanceimagingwhilehearing therecordingoftheownnormalphonation andwithdiscreteormarkedchangeofa vowel
Theacousticcorrectionofspeech errorsactivatesbrainareasandmay beenhancedbybiofeedback
Biofeedbackindysphonia---progressandchallenges 245 always in this order, separated by 3min habituation.
Throughout the procedure, the patient was instructed to reducethetensionofthelarynxmuscles,whetherby watch-ing the integrated electromyographic signal or by deep breathing,arelaxationtechniquethatshouldberepeated inthehouseholdtwiceaday.22
The patient was evaluated after three and 6 months, byperceptual, vocaland visualmeasurements. Therewas reductioninEMGaveragelevels:from59Vto7Vinthe non-vocalization;from51Vto9Vincounting;and,from 70V to12Vin conversationand thesereductions were maintainedafter6months.Inperceptualvoiceanalysisby theVocalScreeningProfileBuffaloIIWilson24(rangingfrom 1 --- normal--- to 5 ---severe deviation), thedeviation was rated 4,atthe beginningoftherapy,passing to1.5,after treatment.22
The authors22 made an interesting comment concern-ing the behavior differentiation in counting biofeedback compared to conversation. Contrary to the initial expec-tations, the intensity of reductionof muscle tensionwith electromyographybiofeedback regardingthe conversation wasdifferentfromthatobservedinthecounting,leadingto theconclusionthatthecountingandtheconversation con-formdifferentdynamics.Counting involvesresponses that do notdepend ona listener. The conversationrequires a partner,whosevocalbehaviormediatessubtleandexplicit socialcontingencies,requiringchangesinthediscriminative stimulusforvocalsettings.Fromthis,therewasindication ofstartingthetreatmentwithconversation,sinceitinvolves morecomplextasksandmoreeffort,thusenablingearlier changeandachievingresultsinashortestperiodoftime.
The thirdstudy12 was notdedicated tomuscle tension dysphonia.Itsinclusionderivedfromtherelevanceof the approachespresented bytheauthoronexplainingtheuse ofbiofeedbackforphonemicspeechdisturbances.Hetests andprovestheeffectivenessofthevisualelectromyographic biofeedbackemploymenttocorrectphonologicaldisorders in children, refractory to other treatments. The distur-bances consisted in the wrong articulation of a phoneme (acquisition)orthemisuseofaphonemealreadyacquired, inaspontaneouscontext(automation).Thesechangesmade the children reading, writing and literacy difficult, from whichderivedtheimportanceofspeechtherapy.
Inthecontextualizationtoexplainthefavorableresults ofelectromyographicbiofeedback,theauthor12arguesthat theprimarydifferencefromotherspeechtherapeutic tech-niquesistoreduce theauditorystimulusandincrease the visualpathwaythroughinstrumentalsignal. This visualiza-tionmakeseasierforthepatienttoidentifyanacquisition or an automation error. The therapist offers the patient thecorrectphono-articulatoryproductionthatenablesthe recording of themental model. Then hestimulates emis-sionsbythepatient,askinghimtomentallyattachthetask and try to correctly repeat the emission, by using visual feedback.
There were two information sources on feedback in these experiments. One consisted in detecting nasal air flow through pneumotachograph and presenting it on the computer screen as a graphic signal. The second source ofinformationwasthespectrogram oftheacousticsignal presentedinrealtimesothepatientcouldmakehis adjust-mentsinordertoachievethetargetspeech.Thevisionof
thealterationwasawarningoferrorproductionandenabled theindividualtomodifyhisperformance.
In the firstsession,12 a phonoaudiologist demonstrated theuseofequipmentandexemplifiedtherightandwrong phonation, recording them to serve as a model to be achievedbythepatient.Onceassuredthepatient’s under-standing about the therapy, the patient performed the exercises,insessionslasting60min,whichweremonitored bythespeechtherapistinvideoconferencing.
Toexplainthereasonswhyadolescentandadultpatients persisted with acquisition and automation errors, the author12 contextualizesthatthephonologicaldevelopment occursuntiltheage ofeight,whentheindividualreaches theupperlimitofspeechandlanguagedevelopment.From thiscompleteness, thechanges willonly be possible with therapy,becausetheindividualdoesnothavethenecessary informationtomakethecorrections.Otherstrategiesneed tobeprovidedsothattheindividualcanformanewengram, welldifferentiatedwithrespecttotheerroneousone,tobe recognizedandcorrectedwheneverarticulated.12,13
Ruscello12 concludesthepaperstatingthatthe biofeed-back gives good results not because it is restricted only tothespectrogram visualization,butbecausethemethod introducesinthecomplexofsystemsinvolvedin speecha newelement that has receptive and expressive features, facilitatingitsintegrationintothecorrectspeech articula-tion.
Thefourthstudy21aimedatpresentingacaseofafemale teenager,16yearsold,Caucasian,withhistoryof paradox-icalmovementofvocalfoldssignedbyvideolaryngoscopy, associatedwithchestpain,muscletensiondysphonia, diffi-cultyinbreathingandafeelingofsuffocation.Despitebeing treated with breathingexercises for 12 months, progress wasnotnoticed, andatreatment withelectromyographic biofeedbackwasindicated.
The treatment consisted of a weekly session, for 10 weeks.Itstarted withtwosessionsfor electromyographic evaluation of muscle tension, without biofeedback.From then on, each session started with 5min of vocal basal rest, followed by 10min of relaxation of muscle tension undervisualization of the biofeedback spectrogram, with theguidanceofachievingthecriterionofreductionof mus-clecontraction,butwithoutanyotherguidance.Thepatient shouldgetthemusclerelaxationbytrialanderror,observing thespectrogram.
Associatedwithbiofeedback,thepatientevaluatedher perceivedadaptationbyusingasixpointsLikertscale,and theintensityofchestpainbya10pointsvisualanalogscale. Additionally,thepatient’smotherevaluatedthepain inter-ferencewithactivitiesofherdaughter,alsoemployingasix pointsLikertscale, withtheobjectiveof determiningthe socialvalidityofthetreatment.Graduallynewtension lev-elswereestablished,lowerthanthepatient’sbaseline,and thetherapyceasedwhennormallevelshavebeenachieved. Theinitialleveltobereachedwasfixedto10V, there-fore 2V lower than the baseline levels of the patient. Newtargets weresetfollowing thecriteria:(a)whenever, inthreeconsecutivesessions,thepatientreachedthegoal or exceededit achievinggreater relaxation,the goal was reducedby2V;(b)ifintwosubsequentsessionsthepatient didnotreachthe settarget, thetarget wasincreasedby 1V.Intherangeof5minbetweenoneseriesandanother,
the patient could see the spectrogram to understand its operation.
After4weeks,thepatienthadnotneeded tomissany school activity, and did not report chest pain, while her motherreported no interference of vocal disorder in the dailyactivities ofthe girl.Despitereportinggoodresults, theauthors21consideredthatthechangesmayhaveresulted fromnonspecificeffectsderivedfromtheexerciseintensity, morethananybiofeedbackfeature.
The set of articles from that first block took place from 1991 to 2005 period, therefore predating the mag-neticresonanceimagingstudiesfortheapplicationofvoice biofeedback. This set was presented in this systematic reviewtopointoutthepredominantreasoningwithinthese twodecades,foundingthehypothesisthatthismethodhad thepotentialtomodifythevocalengram.However,the evi-denceswerebasedonsmallsamples,oncase studieswith differentmethodologies,thuswithoutthepossibilityof gen-eralization,aswellemphasizedWarnesandAllen.21
The mechanisms with which it was sought to explain the effects of electromyographic biofeedback for voice were based on behavioral reinforcement, increased self-confidence and motivational boost, but the results were conflicting.10,25,26Yetthelineofresearchhasnotbeen aban-doned,althoughthereisscarcityofpublications.
This hypothesis fragilitymotivated studies with profes-sional singers, since the absence of large vocal disorders couldfacilitatetheidentification ofnewevidences onthe voicebiofeedback.Thechoiceofthispopulationwasaform of standardizing some intervening variables. In addition, thesestudiescouldclarifygapsonthemechanisms respon-siblefortheresultsofresearchandguidemoreobjectively theuseofthemethodinvoicetherapy.
Biofeedbackinthevocallearningofprofessionals Fromapopulationwithvocal healthandgoodorexcellent emissionquality,studieshavesoughttoidentifythe possi-bilityoftheelectromyographicbiofeedbackfacilitatingthe biomechanicallearninginvoiceproduction.
Astudyaimedtoidentifyamethodforfacilitationinthe acquisitionofnewskillstovoiceproductionwithlaryngeal relaxation,aswellastodeterminewhethertheterminalor concurrentbiofeedbackoptimizesthislearning.10
Afterperformingapilotproject todeterminethetype andpositionofelectrodes,aswellastheformoffixation, 18womenandfourmenwithameanageof22.41yearsold (range19---27years),withnohistory ofhearing or speech problems andwithnoexperience of vocal therapy,which self-declaredhealthy,wereincludedinthestudy.After affix-ingapairofelectrodes0.5cmapartfromeachotherinthe midlineofthe thyroidmembrane, andasecond pair1cm awayfromthelipcommissure,bilaterally,eachparticipant satin front of twocomputer screens.In one of these, it weredesigned11blocksofstimulussentences,toberead aloudlookingtorelaxthelarynxbyobservingthesEMGof thethyroidsite,whichwasdesignedontheotherscreen.10 After the first two reading stimuli, participants were randomizedtoreceiveconcurrentor terminalbiofeedback in each reading. In the concurrent stimulus, participants watchedthe muscular electrical activityasthey readthe
sentences.Intheterminalbiofeedback,theyreadthe sen-tenceandthenwatchedthestaticgraphicofthesEMG.The whole procedurelastedapproximately 50---60min andwas repeatedafteroneweek,forfixingrealxoflarynx,although withoutbiofeedback.10
The study results did not corroborate other findings. Hence,biofeedbackdidnotappeartofacilitatethelearning ofvocaltask.Itwasobservedlearningofrelaxationinthe facialsite,butnotinthethyroidsite,contrarytotheinitial hypothesisofthisstudy.Thethirdandmostinteresting find-ingwastheequalityinterminalandconcurrentbiofeedback onlearning.
The importance of this experiment was to demon-stratethattheconcentrationofattentiononatask(which consistedinreducingmuscletensionofnormalindividualsto valueslowerthantheirbaselinelimit)degradesthemotor learning,whichpointedtothecomplexityofvocallearning. When comparing the results of this study10 withthese of Warmes and Allen,21 it is observed the weakness that markedtheevidenceonelectromyographicbiofeedbackfor voice,untilthefirstdecadeofthetwenty-firstcentury.For patientswithmuscletensiondysphonia,theresultsshowed thatmorecomplextasksintherapywithelectromyographic biofeedbackcouldprovidebetterconditionsforgood ther-apeuticresultsinlesstime,whichwasnotthecaseinthe experimentwithsingers.Forthem,theincreased complex-ityoftheexperimentcausedthenon-learningofthevoice taskformusclerelaxation.
Thesecondstudy16soughtapplicationofthe electromyo-graphic biofeedbackin thevocal enhancementofsingers, toassistthemtoissueaclear,strongandcomfortablevoice withahighrange,for keepingthelowpositionof the lar-ynx,whichprotectsagainstinjuriesinveryloudemissions.16 Three objectiveswereset out:toprovetheusefulness of sEMGasindicatoroftheactivityofdepressormusclesofthe larynx;totesttheusefulnessofthemethodtoteachsingers tolowerthelarynx,activatingthedepressormuscles,andto maintainthisstanceastheysingand,ultimately,to deter-mine whether the laryngeal posture improves the quality of thesingingtone or promotesachange whatsoeverin a scientificallymeasurablecomponentofthesoundspectrum. Twenty-twotrainedclassical singers werecomparedto eightuntrainedsingers,usingsEMGwithelectrodesplaced bilaterallyoverthethyroidcartilageinanattempttoisolate thesternohyoidandsternothyroidmuscles---primary depres-sorsandstabilizersofthelarynx.Itwasrequestedforeach singertoissuefourtimesthevowel/a/inapitchslightly louderthanhisvoice(baritone,tenor,mezzo-soprano and soprano),inordertogetoutofthevocalcomfortzone.
Inallcases,theelectromyographicbiofeedbackmadeit possible toindicate the activity of thedepressor muscles ofthelarynx,aswellasfacilitatedthelearningofclassical singerstokeeplowthelarynxduringsinging,toensurethe clearest,mostbeautifulandharmonicemissions.
Noticingthatsingerswithtrainingwereabletoimprove in up to12% their baseline performance witha few min-utes oftraining,withnodifferencebetween genders,the authors15encouragedtocarryoutfurtherresearchwiththe samedesign,toaccumulateevidence.
Attentionmustbedrawntothedetailoftheabsenceof genderdifferencesinvocalimprovementwith electromyo-graphic biofeedback,because,according totheauthors,16
Biofeedbackindysphonia---progressandchallenges 247 thedatumallowstohypothesizethatthechangespromoted
bythetherapyshouldbeintense,astheywerelargerthan the anatomicallarynx differences in both genders.These changesshould alsobebroad, astheyallow relaxationof various laryngeal muscles according to the needs of the patient.
Analyzingtheresultsofresearchwithdysphonicpatients andsingers,itbecameevidenttheneedtoconsiderthe neu-ralnetworkinvolvedintheissueandthevoicesettingina speech,beitemissionsofisolatedphonemes,speakingor singing.
Thecomprovationofthesehypothesesstartedfromthe seconddecadeofthetwenty-firstcentury,withthestudies onfunctionalmagneticresonanceimagingandbiofeedback withbirds,identifyingthathearingchangespromotedmotor correctionsinthesinging.27Thisfindingraisedthepossibility that,inhumans,biofeedbackcouldpromotechangeinvocal behaviorbyrestructuringtheneuralnetwork.
At this time,it was already admitted that the speech motor system depends on the auditory feedback, which is responsible for controlling the vocal articulations from moment to moment. Thus, if during the emission of the voiceadisturbanceoccurs,theauditorysystemimmediately counter-actsandenablescorrectiveadjustmentsoffsetting theimbalance.Theexplanationforthismechanismisthat each sound corresponds to a perceptual auditory area.28 Thesebrainregionsintheauditorycortexareusedtoupdate andrefinethemotorcommandsthatguidetheacoustic sig-nalalongtheseregions.
Similarlytothisphysiologicalsystem,theprograms avail-ableintheInternetnetworkwithauditoryorvisualfeedback enablecorrectionofphonatoryacousticmistakes,butthese correctionsarestillrudimentary.Inthenervoussystem,it is assumed that the auditory perceptionsare notcreated equally, that is, tothe sound of the same phoneme, the hearingrecordcanbedistinguished,sothatthesoundsthat exceedacertainvolumethresholdarebetteridentifiedthan others.
This discretionary ability of the nature of the produc-tionofadiscourseisahypothesiswhereitisassumedthat smallvocal deviationsarenot perceivedby theindividual sender,phenomenoncalledperceptualmagneteffect.Not withstanding,alldeviationsthatreachacertainthreshold areself-recognizedandsubmittedtothevocaladjustment mechanism.The hypothesiswasplausiblebutraised ques-tionwhethertheauditoryerror dependedonlyonthelow levelofhearingdistanceofalimitorintentionaltargetor wastriggeredbyatypeofhigh-levelmodification.
In 2013, it was held a research2 employing electro-magnetic biofeedback and functional magnetic resonance imaging(fMRI)toexaminetheneuralnetworksand deter-minethechangesinresponsetosuddendisturbancesofthe auditory feedbackloop, triggeredby unexpected acoustic modifications,admittingthatachangeoffeedbackcloseto aborderregionmightevokeamoreintenseresponsethan achange lyingsafelywithinan acceptablevariabilityof a certainsoundinaspeech.
Eighteen dexterous subjects, aged between 19 and 33 yearsold(meanof23.5years),withgenderdistributionof 1:1,tookpartinabehavioralexperiment.AllhadEnglishas afirstlanguage;didnotreporthearingalterationsorspeech disordersandhadnocontraindicationtofMRI.
Theexperimentconsistedoftwophases,onebehavioral andthe other imagery. Inthe behaveoralphase, allwere subjected to a pretest to define the parameters of pro-ductionandperceptionofspeechofdifferentvowelstobe employedintheexperiment.2
Intheimagecapturephase,brainactivitywasmeasured by fMRI, during the biofeedback task, within a triggering eventofthecoordinatingmechanismofstimulus.The anal-ysisoffMRImapsdemonstrated,bymonitoringtheauditory feedback,thatthespeechmotorsystemcanquicklycorrect smallarticulatoryerrors.Whenconsideringthevariabilityof eachsubject,onecanidentifythatsimilarchangescan sys-tematicallyevokeresponsesof differentmagnitude,when theyoccur indifferent partsoftheauditory space.Ifthe changes occur in an area close to the limit for the cat-egoryin which a sharpchange wasnoticeable, it triggers corticalactivityandseverebehavioralcompensation.Ifthe changewasfarfromthemodifiedcategorylimit,the behav-ioralchangeneverexceeded8%oftheunchangedacoustic behavior.2
This experiment2 hasshown for the firsttime thatthe auditorycontrolwithbiofeedbackissensitive tolinguistic categoriesandcan triggerdifferent behaviorsof the neu-ral network, facilitating the automatic correctionsof the speechmotor system, especiallyin the superiortemporal andinferiorfrontalregions.
The technological improvement of electromyographs, surface electrodes and computer programs for graphic conversionofelectricalcurrentinspectrogramopenednew horizonsofresearchinthisarea,abouthalfwaythroughthe firstdecadeofthetwenty-firstcentury.Notwithstanding, thisdidnotmeangreater homogeneityof thetherapeutic resultsintheapplicationofelectromyographicbiofeedback forvoice.Successreportswerealternatedwiththeabsence ofvocalmodifications.
The researchers2 sought explanations for the dispar-ityof results from two pathways. A group wasdedicated to the improvement of the electromyographic biofeed-back methodology for voice disorders, seeking to reduce confoundingvariables, whose activities could explain the inabilitytoformevidence.
Asecondgroupwasdedicatedtotheresearchofneural processesinvolved in changingthemuscle engramof nor-malanddysphonicpatients,resultinginprovingthevalidity ofthistherapeutic method,agoalthat hasrecentlybeen achievedthroughstudy by functional magnetic resonance imagingwithsingers,whosefindingscanbegeneralizedto patientswithvoicedisorders.
Thedirectioningoffutureresearchshouldbefocusedon findingastandardizationofbiofeedbackmethodsforvoice, whichisstilladifficult challenge.There isevidenceasto the validity of the method to guide vocal emissions with quality,buttherearenocomparativeparametersthatallow generalizations.
Conclusions
and
further
research
There is evidence that the electromyographic biofeed-backpromotes changesintheneuralnetworksresponsible forspeech,especiallysuperior temporalandinferior pari-etalcortex, allowingfor conditionstopromotebehavioral
changeinnormalpatientsorwithvoicedisorder,butthisis stillahypothesis,becausethediscussionoftheexperiments lackedphysiologicalevidencethatcouldserveastheirbasis.
Funding
MCTI/CNPQ/Universal14/2014-FaixaA.
Conflicts
of
interest
Theauthorsdeclarenoconflictsofinterest.
References
1.Bouchard KE,Mesgarani N, Johnson K, Chang EF. Functional organizationofhumansensorimotorcortexforspeech articu-lation.Nature.2013;495:327---32.
2.NiziolekCA,GuentherFH.Vowelcategoryboundariesenhance cortical and behavioral responses to speech feedback alter-ations.JNeurosci.2013;17:12090---8.
3.CieloCA,FingerLS,Roman-NiehuesG,DeuschleVP,SiqueiraMA. Organic-functionaldysphoniaandcomplainsoffallergicand/or digestivedisturbance.RevCEFAC.2009;11:431---9.
4.Behlau M, Madazio G, Moreti F, Oliveira G, Santos LMA, Paulinelli BR, et al. Eficiência evalores de corte de proto-colos deautoavaliac¸ão doimpacto deproblemas devoz. In: 21◦CongressoBrasileirodeFonoaudiologia,2◦CongressoIbero AmericanodeFonoaudiologia.2013.p.9.
5.Pontes PAL, Vieira VP, Gonc¸alves MIR, Pontes AAL. Car-acterísticas das vozes roucas, ásperas e normais: análise acústica espectrográfica comparativa. Rev Bras Otorrino-laringol.2002;68:182---8.
6.AllenKD,BernsteinB,ChaitDH.EMGbiofeedbacktreatmentof pediatrichyperfunctionaldysphonia.JBehavTherExp Psychi-atry.1991;22:97---101.
7.KhoddamiS,AnsariN,Izadi F,MoghadamST.Theassessment methodsoflaryngealmuscleactivityinmuscletension dyspho-nia.SciWorldJ.2013;2013:1---7.
8.BehlauM,AzevedoR,MadázioG. Anatomiadalaringee fisi-ologiadaproduc¸ãovocal.In:BehlauM,editor.VozOlivrodo especialista.1sted.SãoPaulo:Revinter;2001.p.1---52. 9.MerlettiR,BotterA,TroianoA,MerloE,MinettoMA.Technology
andinstrumentationfordetectionandconditioningofthe sur-faceelectromyographicsignal:stateoftheart.ClinBiomech. 2009;24:122---34.
10.Yiu EM-L, Verdolini K, Chow LPY. Eletromyographic studyof motorlearningforavoiceproductiontask.JSpeechLangHear Res.2005;48:1254---68.
11.Allen KD. EMG biofeedback treatment of dysphonias and relatedvoicedisorders.JSpeechLangPatholApplBehavAnal. 2007;2:149---57.
12.RuscelloDM.Visualfeedbackintreatmentofresidual phono-logicaldisorders.JCommunDisord.1995;28:279---302. 13.Peper E, Harvey R, Akabayashi N. Biofeedback an evidence
based approach in clinicalpractice. JpnJ Biofeedback Res. 2009;36:3---10.
14.George R, ChungTD, Vedam SS, Ramakrishnan V, Mohan R, WeissE,etal.Audio-visualbiofeedbackfor respiratory-gated radiotherapy: impact of audio instruction and audio-visual biofeedback on respiratory-gated radiotherapy. Int J Radiat OncolBiolPhys.2006;65:924---33.
15.HenschenTL,BurtonNG.TreatmentofspasticdysphoniabyEMG biofeedback.BiofeedbackSelfRegul.1978;3:91---6.
16.KirkpatrickA,McLesterJR.Teachinglowerlaryngealposition withEMGbiofeedback.JSing.2012;68:253---60.
17.KirkpatrickA, McLesterJR. Trainingvocaliststo achieveand maintainalowerlaryngealwhilesingingusingsEMG biofeed-back. BiofeedbackFederation. United Kingdom:Biofeedback Federation;2012.p.253---60.
18.LindaCP.Comparisonofconcurrentandterminal electromyo-graphic biofeedback on the training of laryngeal muscle relaxation.UniversityofHongKong;2002.
19.Schneider-Stickler B, Knell C, Aichstill B, Jocher W. Biofeedback on voice use in call center agents in order to prevent occupational voice disorders. J Voice. 2012; 26:51---62.
20.CieloCA,ChristmannMK,RibeiroVV,HoffmannCF,PadilhaJF, SteidlEMS, et al.Musculoskeletal stress syndrome, extrinsic laryngealmusclesandbodyposture:theoreticalconsiderations. RevCEFAC.2014;16:1639---49.
21.WarnesE,AllenKD.Biofeedbacktreatmentofparadoxicalvocal foldmotionandrespiratorydistressinanadolescentgirl.JAppl BehavAnal.2005;38:529---32.
22.WatsonTS,AllenSJ,AllenKD.Ventricularfolddysphonia: appli-cationofbiofeedbacktechnologytoararevoicedisorder.Behav Ther.1993;24:439---46.
23.KennedyCH.Trendsinthemeasurementofsocialvalidity.Behav Anal.1992;15:147---56.
24.Mu˜nozJ,MendozaE,FresnedaMD,CarballoG,RamirezI. Per-ceptual analysis in different voice samples: agreement and reliability.PerceptMotSkills.2002;94:1187---95.
25.MarynY,DeBodtM,VanCauwenbergeP.Effects of biofeed-back in phonatory disorders and phonatory performance: a systematicliteraturereview.ApplPsychophysiolBiofeedback. 2006;31:65---83.
26.ThomasLB,StempleJC.Voicetherapy:does sciencesupport theart.CommunDisordRev.2007:49---77.
27.Sober SJ, Brainard MS. Vocal learning is constrained by the statisticsofsensorimotor experience.ProcNatlAcadSciUS A.2012;109:21099---103.
28.Golfinopoulos E, Tourville JA, Guenther FH. The integra-tion of large-scale neural network modeling and functional brainimaginginspeechmotorcontrol.Neuroimage.2010;52: 862---74.
