bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Envelope-tACSmodulatesintelligibilityofspeechinnoise
AnnaWilsch1,ToralfNeuling2,&ChristophS.Herrmann1,3*
1
ExperimentalPsychologyLab,DepartmentofPsychology,ClusterofExcellence“Hearing4all”,
EuropeanMedicalSchool,CarlvonOssietzkyUniversity,Oldenburg,Germany
2
3
DepartmentofPsychology,UniversityofSalzburg,Salzburg,Austria
ResearchCenterNeurosensoryScience,CarlvonOssietzkyUniversity,Oldenburg,Germany
*
Correspondence:
Prof.Dr.ChristophHerrmann
ExperimentalPsychologyLab
CarlvonOssietzkyUniversitätOldenburg
AmmerländerHeerstr.114-118
26131Oldenburg
Tel.: +494417984936
Fax: +494417983865
Email: [email protected]
Abstract
Corticalentrainmentoftheauditorycortextothebroad-bandtemporalenvelopeofaspeechsignalis
crucialforspeechcomprehension.Thisentrainmentresultsinphasesofhighandlowneuralexcitability
whichstructureanddecodetheincomingspeechsignal.Entrainmenttospeechisstrongestinthetheta
frequencyrange(4–8Hz),theaveragefrequencyofthespeechenvelope.Ifaspeechsignalisdegraded,
for example masked byirrelevant information such as noise, entrainment to the speech envelopeis
weakerandspeechintelligibilitydeclines.
Besides perceptually evoked cortical entrainment, transcranial alternating current stimulation (tACS)
can entrain neural oscillations by applying an electric signal to the brain. Accordingly, tACS-induced
entrainmentinauditorycortexhasbeenshowntoimproveauditoryperception.Theaimofthecurrent
studywastoexternallymodulatespeechintelligibilitybymeansoftACSsuchthattheelectriccurrent
correspondstotheenvelopeofthepresentedspeechstream.
ParticipantsperformedtheOldenburgsentencetestwithsentencespresentedinnoiseincombination
withtACS.Critically,thetimelagbetweensentencepresentationandtACSwasmanipulatedfrom0to
250msin50-mssteps(auditorystimuliweresimultaneoustoorprecededtACS).
First, we were able to show that envelope-tACS modulated sentence comprehension such that on
averagesentencecomprehensionatthetimelagofthebestperformancewassignificantlybetterthan
sentencecomprehensionoftheworstperformance.Second,sentencecomprehensionacrosstimelags
wasmodulatedsinusoidally.
Insum,envelopetACSmodulatesintelligibilityofspeechinnoisepresumablybyenhancing(timelag
within-phasestimulation)anddisrupting(timelagwithout-of-phasestimulation)corticalentrainment
tothespeechenvelopeinauditorycortex.
1
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Introduction
Thelifespanoftheworldpopulationiscontinuouslyincreasing.Thisincreaseinageentailsastrong
increase in hearing loss. In 2030, the prevalence of clinically relevant hearing loss is expected to be
doubled.Consequently,hearinglosswillbeoneofthesevenmostprevalentchronicdiseases(WHO
2004;WHO2013).
Despite huge advances in the development of hearing aid technology, challenging listening
situationssuchastheso-calledcocktailpartyproblem(Cherry,1953)arestillnotentirelycompensated
for.Thepresentstudywillsuggesttranscranialalternatingcurrentstimulation(tACS)asapossibilityto
improvespeechcomprehensioninadverselisteningsituations.
The ability to comprehend speech is tightly linked to the neural oscillatory response in auditory
cortex.Thatisbecauseauditorycortextracksthespectro-temporalstructureofspeechalsoknownas
corticalentrainment(e.g.,Abramsetal.,2008;Lalor&Foxe,2010;Luo&Poeppel,2007).Ingeneral,
neuraloscillationsareknowntoentraintothetemporalstructureofexternalstimuli.Acrossdifferent
sensory modalities, it has been shown that a rhythmically presented stimulus leads to a phase and
amplitudealignmentofneuraloscillationswithinthefrequencyofthepresentedrhythm.Entrainment
of oscillations results in distinct phases of high and low excitability on the population level, thereby
generating time windows of a higher firing likelihood (Engel et al., 2001; Lakatos et al., 2005). This
synchronization of neural oscillations and external stimuli facilitates the processing of relevant
information(Lakatosetal.,2008;Schroederetal.,2010).
Entrainment to the broad band temporal envelope of speech has been shown to be crucial for
speechcomprehension(e.g.,Doellingetal.,2014).Thestrongestresponsecanbefoundinthetheta
range around 4–8 Hz which corresponds to the average frequency rate of a speech signal
(Chandrasekaranetal.,2009;Ghitza&Greenberg,2009).Thefunctionalroleofcorticalentrainmentto
theenvelopehasbeenwidelydiscussed(forareview,seeDing&Simon,2014;).Whereassomeargue
thatentrainmentorphaselockingtotheenvelopeonlytrackstheacousticpropertiesoftheenvelope
(Howard & Poeppel, 2010; Millman et al., 2015; Steinschneider et al., 2013) others argue that
entrainment is a mechanism of syllabic parsing (Giraud & Poeppel, 2012) or sensory selection, (i.e.,
segragating the speech stream from background noise; Schroeder & Lakatos, 2009). Overall, neural
oscillations in the 4–8-Hz range are supposed to be a means to structure and decode the incoming
acousticstream(forareview,seePeelle&Davis,2012).
Theextenttowhichauditorycortexisabletotrackaspeechstreamdependsontheamountof
spectral detail. That is, the more degraded the speech stream the worse the neural tracking of the
envelope(Peelle&Davis,2012).Alongtheselines,Ahissaretal.(2001)couldshowthattheabilityto
2
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
comprehendspeechdependsonhowwellauditorycortextracksthespeechsignal(seealsoPeelleet
al.,2013).AstudybyDingandSimon(2013)manipulatedthesignal-to-noiseratioofcontinuousspeech
andstationarynoise.Theywereabletoshowthatentrainmenttotheenvelopeinthe4to8-Hzrange
decreaseswithincreasingnoiselevel.Thesefindingsindicatethatentrainmenttothespeechenvelope
iscorrelatedwithnoiselevelandspeechintelligibility.
Ifasuccessfulentrainmentofoscillationsintheauditorycortexisessentialforspeechintelligibility,
it should be possible to increase intelligibility of an acoustically degraded speech signal by inducing
corticalentrainmentandtodecreaseintelligibilitybydisruptingcorticalentrainment.Ifthiswouldbe
established,thespeechcomprehensionofhearingaiduserswhoexperiencehearingimpairmentscould
beimprovedbyexternallyinducingcorticalentrainmenttothespeechenvelope.Onetechniquethatis
capable to modulate ongoing brain oscillations is transcranial alternating current stimulation (tACS).
tACSinducescorticalentrainmentinafrequencyspecificmanner(forreviews,seeAntal&Paulus,2013
and Herrmann et al., 2013; Helfrich et al., 2014; Zaehle et al., 2010). By modulating the resting
membranepotentialofneuronsviatACS,oscillatoryactivityisguidedbytheexternaloscillationsonthe
populationlevel(Fröhlich&McCormick,2010).Atthesametime,tACSisabletodisruptentrainment
bymeansofstimulationthatisanti-phasictosynchronizedcorticalactivity(Helfrich,Knepper,etal.,
2014;Strüberetal.,2014).
Similartoentrainmenttoexternalstimuli,tACSinducedentrainmenthasbeenshowntomodulate
auditoryperception(forareview,seeHeimrathetal.,2016).Forexample,tACSinthealphafrequency
range(10Hz)ledtocorticalentrainmentwhichinturnmodulateddetectionofpuretonesembedded
innoiseinasinusoidalmanner.Thatis,pure-tonedetectiondependedonthephaseofthestimulation
signal(Neulingetal.,2012).FortACSinthedeltafrequencyrange(4Hz),phasedependentdetectionof
near-thresholdauditoryclicktrainscouldbeshownaswell(Rieckeetal.,2015).Bothstudieswereable
to provide insight that entrainment induced by tACS modulates the perception of near-threshold
auditory stimuli. Two studies that investigated the impact of tACS on speech stimuli showed that
phonemedetectionwithsyllablesismodulatedby40-HztACSinyoungeraswellasinhearingimpaired
adults(Rufener,Oechslin,etal.,2016;Rufener,Zaehle,etal.,2016).
TheaimofthecurrentstudywastotestwhethertACSmodulatesspeechintelligibility.Participants
performedthewell-establishedOldenburgsentencetest(OLSa;Wagneretal.,2001)withsentences
presentedinnoiseincombinationwithtACS.Critically,thetACSsignalcorrespondedtothebroad-band
envelope of the presented sentences (i.e., envelope tACS), in order to simulate the LFPs during
presentationofclearspeech.Here,wemanipulatedthetimelagbetweensentencepresentationand
tACSfrom0to250msin50-mssteps(auditorystimuliweresimultaneoustoorprecededtACS).This
wasdonefortworeasons:First,sofaritisunclearwhattimelagisappropriatesuchthattACSonset
3
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
andauditorystimulationonsetarealignedinauditorycortex.Second,inordertoobserveamodulation
of sentence comprehension by tACS, different time lags allowed for aligned/in-phase and possibly
beneficialentrainmentaswellasout-of-phaseordisruptingentrainmentpossiblyimpedingsentence
comprehension. Overall, we expected to show that envelope tACS in auditory cortex modulates
sentencecomprehensionandthatsentencecomprehensionfluctuatesacrosstimelags.
Methods
Participants
Nineteen healthy young adults (11 female, mean age = 23.5 years) without prior or current
neurologicalorpsychiatricdisordersparticipatedinthestudy.Theexperimentalprotocolwasapproved
ofbytheethicscommitteeoftheUniversityofOldenburg,andhasbeenperformedinaccordancewith
the ethical standards laid down in the 1964 Declaration of Helsinki. All participants gave written
informedconsentbeforethebeginningoftheexperiment.
Experimentalprocedure
Eachparticipantcompletedtheexperimentintwosessionsontwodifferentdaysinordertoassure
tACStobebelow20minutesperday.Theparticipantswereseateduprightinareclinerinadimlylit
roombeforethestimulationelectrodesandheadphoneswereattached.Afterwards,thetACSthreshold
wasdetermined.Tothisend,westartedthestimulation(3Hz,5sec)withanintensityof400μAand
asked the participants to indicate whether they perceived a skin sensation or phosphene. The
stimulationintensitywasincreasedbystepsof100μAuntiltheparticipantindicatedskinsensationor
phospheneperceptionoranintensityof1500μAwasreached.Incasetheparticipantalreadyreported
anadverseeffectat400μA,theintensitywasreducedtoastartlevelof100μA.Stimulationintensity
throughouttheexperimentcorrespondedthentothehighestintensitythatwasnotperceivedbythe
participant. Then, the participant’s absolute threshold of hearing was determined for each ear (i.e.,
subjective hearing threshold). After that the participant was familiarized with the speech test
(Oldenburgersentencetest,seebelow).Subsequently,theparticipantcompleted9listsofthetestthat
wererandomlyassignedto9tACSconditions(Figure1A),whichtookaround45minutes.Thedesign
wasdouble-blind,i.e.,neithertheparticipant,northeexperimenterknewtheorderoftheconditions
before the end of the experiment. After each list, the participants had to indicate whether they
perceivedtheelectricalstimulation.Followingthespeechtest,theparticipantswereaskedtocomplete
4
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
aquestionnaireonpossibleadverseeffectsoftACS(Neulingetal.,2013).SeeFigure1Aforaschematic
displayoftheexperimentalprocedure.
Oldenburgsentencetest
TheOldenburgsentencetest(OLSa;Wagneretal.,2001)isanadaptivetestthatcanbeusedto
obtainaspeechcomprehensionthreshold(SCT)duringnoise.Forthetest,30sentencesthatconsisted
offivewordsarepresentedandtheparticipantisaskedtorepeatthesentencesorally.Thetestmaterial
consistsof40testlistsofwhich22(ontwodays)wererandomlychosenforeachparticipant.Foreach
testsession,thetwotestlistswereusedtofamiliarizetheparticipantwiththetestmaterialaccording
tothehandbook.Subsequently,ninelistswerepresented,separatedbyself-pacedbreaks.
Thepresentedsentencesweresampledat44100HzanddeliveredviaMATLABtoaD/Aconverter
(NIDAQ,NationalInstruments,TX,USA),attenuatedseparatelyforeachear(Tucker-DavisTechnologies,
Alachua,FL,USA;modelPA5),andpresentedviaheadphones(HDA200,Sennheiser,Germany).The
noise of the OLSa was presented at 65 dB above individual hearing threshold, the intensity of the
sentenceswasadjustedaccordingtothetestmanual.Thepresentednoiseonsethadthesamespectral
characteristicsasthepresentedsentence,inordertoobtainanoptimalmaskingeffectofthesentence.
Thenoisewaspresentedfrom0.5spriortosentenceonsetuntil0.5saftersentenceoffset.Foreach
OLSalist,aspeechcomprehensionthreshold(SCT)wascomputed,whichisthedifferenceofthesound
pressureofthespeechsignalandthenoiseofthelast20sentences.Forexample,aSCTof-7dBimplies
that 50 % of the sentences presented with an intensity lower than 7 dB lower than the noise were
repeatedcorrectly.
5
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Figure1.Experimentalprocedure,tACSconditions,andelectrodeconfiguration.A.Schematicdisplayoftheexperimentaltime
line.B.Exampleofatestsentenceandthecorrespondingexperimentalconditions.Petrolcoloredtracesdepictthe6different
timelags,graytracesthedirectcurrentconditions.TimelagsrefertothepointintimeoftACSstimulationafterpresentation
oftheauditorysentence(i.e.,0–250ms). C.BlueandredboxesillustratepositionsofthetACSelectrodesonthehead.Blue
andredtracesdepictanexampleenvelopeofasentence(black:audiosignal)thatwasusedfortACSstimulation.
Transcranialalternatingcurrentstimulation
tACSwasappliedusingabattery-drivenNeuroConnDCStimulatorPlus(NeuroConnGmbH,Ilmenau,
Germany).StimulationelectrodeswereplacedinabipolarmontageonCz(7×5cm)andbilaterallyover
theprimaryauditorycorticesonT3andT4(4.18×4.18cm)accordingtotheinternational10-20EEG
system(Figure1B).Theimpedancewaskeptbelow10kΩbyapplyingaconductivepaste(Ten20,D.O.
Weaver,Aurora,CO,USA).
Thestimulationsignalcorrespondedtotheenvelopeoftheconcurrentspeechsignal(i.e.,envelope
tACS).Ascontrolcondition,anodalandcathodaldirectcurrent(DC+,DC–)matchedtothedurationof
the speech signal were stimulated as well as no electrical stimulation (sham) (see Figure 1B). The
envelope-tACSwasgeneratedbyextractingtheenvelopeofeachOLSasentence.Here,theabsolute
valuesoftheHilberttransformoftheaudiosignalwerecomputedandthenfilteredwithasecondorder
Butterworthfilter(10Hz,low-pass).TomaximizetACSefficacy,peaksandtroughsexceeding25%of
thehighestabsolutepeakwerescaledto100%.Thisway,thepeak-to-peakintensitywasnormalized.
Kubanek and colleagues (2013) reported a lag of around 100 ms between speech envelope and
auditory cortex oscillations that are entrained by the signal. In order to find the time lag at which
6
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
auditorystimulationandtACSwerealignedinauditorycortex(highestbehavioraleffect)andalsoto
testhowtACSmodulatesspeechcomprehensionacrossdifferenttimelags,envelopetACSwasassigned
to6differentdelayconditions(seeFigure1A).EnvelopetACSwasinitiatedbetween0and250ms(50
msstep-size)aftertheonsetofthespeechsignal.ThetACSsignal(sampledat44100Hz)wasdelivered
viaMATLABtotheD/AconverterandfedintotheexternalsignalinputoftheDCStimulator.
DataAnalysis
Foreachparticipant,wecomputedtheSCTforeachoftheninetACSconditions(seeFigure2A).We
contrastedbestandworstperformance(onlythesixenvelope-tACSconditions)ofeachparticipantwith
a t-test, regardless of the time lag. Then, in order to understand the nature of the modulation of
sentencecomprehensionbytACS,eachparticipant’sperformancedatawerepeak-alignedtothebest
performanceandz-transformed.Wedeterminedthispeaktobeatbinthree.Thatisthepositionofthe
timelagof100ms,correspondingtothemediantimelagofthebestperformancesofallparticipants
(seebelow).DuetothecircularfluctuationoftheSCTs(seeFigure2A)wefittedasinewavetothedata,
comparabletotheapproachofNeulingetal.(2012;seealsoNaueetal.,2011).Asacontrol,wealso
performedalinearfitandaquadraticfitonthealignedperformancedataofeachparticipant.Forthe
sinewavefit,thefollowingequationwasfitted:𝑦 = 𝑎 ∗ sin(𝑓 ∗ 2𝜋 ∗ 𝑥 + 𝑐),wherexcorrespondsto
thetACSbins[0;250ms].Theparametersa,f,andcwereestimatedbythefittingprocedure,wherea
wastheamplitudeofthesinewave,fthefrequency,andcthephaseshift.Parameterawasbound
betweenzeroandthree(notethatdatawerez-transformedandnotlikelytoexceedaz-valueofthree),
f between two and eight (we expected that a modulation frequency would be in the theta range
comparable to the syllable rate of the sentences), and c between zero and 12.57 (4 * π). The initial
parametersinsertedinthefunctionwererandomlydrawnfromwithintherangerestrictionsofeach
parameter, respectively. For the linear fit (𝑦 = 𝑎𝑥), the coefficient a was computed with a random
startingvalue.Thequadraticfit(𝑦 = 𝑎𝑥 0 + 𝑏𝑥)wascomputedtoestimatethelinearcoefficientband
thequadraticcoefficienta.Notethatalloftheequationsdonotcontainaparameterfortheintercept
becausethefitteddatawasz-transformedandhencethemeanwaszero.Themodelfitswerecomputed
with the lsqcurvefit function with Matlab (version 8.0, Optimization Toolbox) that allowed for 1000
iterationsinordertofindthebestmodel.CoefficientsofdeterminationR2wasreportedforeachfitas
anindicatorforthegoodnessofthemodelfit.
Finally,theBayesianinformationcriterion(BIC;Schwarz,1978)wascalculatedforeachfunction,in
ordertocomparethegoodnessoffitofthesinusoidal,quadratic,andlinearfits.TheBICcorrectsforan
unequalnumberofparameters.BICscoreswerecomparedwithanrmANOVAandpost-hoct-tests,in
7
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
ordertodeterminethefunctionthatrepresentsthemodulationinsentencecomprehensionbest.TtestswereFDR-corrected.Allanalyseswereexclusivelyperformedontheenvelope-tACSconditionsand
not on tDCS as well as sham (see Figure 1) in order to assess the possible modulation of speech
comprehensionsolelybasedontheeffectofvaryingtimelags.
Results
The aim of the present study was to find out whether envelope tACS modulates sentence
comprehensioninnoiseandtospecifythenatureofthemodulation.Participantslistenedtosentences
with different signal-to-noise ratios. Their ability to comprehend the presented sentences was
quantified by the sentence comprehension threshold (SCT), the signal-to-noise ratio at which
participantscomprehended50%ofasentence.tACSwasinducedsimultaneouslyorwithatimelagof
50to250ms.Figure2AdepictstheSCTsforeachtimelag(i.e.,delaybetweenauditoryonsetandtACS
onset),aswellasanodalandcathodaltDCSandsham.TotestwhethertACSatdifferenttimelagsledto
amodulationofsentencecomprehensionatall,wesortedtheSCTsbybestandworstperformanceof
eachparticipantandcomparedbothgroupswithat-test(Figure2B).Thet-testshowedthattheSCTs
ofthebestperformancesweresignificantlylowerthantheSCTsoftheworstperformances(t(18)=8.8,
p<0.0001,Figure2B).ThiseffectindicatesthattACSmodulatedsentencecomprehension.Theoverall
improvementinsentencecomprehensionwas0.86dBonaverage(sd=0.43dB).Themediantimelag
thatledtothebestperformancewas100msandfortheworstwas50ms.Critically,timelagsimproving
orimpedingperformancewerenotconsistentacrossparticipants.
Figure2.ModulationofsentencecomprehensionthresholdbytACSatdifferenttimelags.A.Bargraphsrepresentsentence
comprehensionthresholds(SCTs;signal-to-noiseratioindB)averagedacrossallparticipantsateachtACSstimulationtimelag,
aswell asfor thecontrolconditionsDC+, DC–, and sham.Note, that lowervaluesindicatebetterperformance.Errorbars
displaystandarderrorofthemean. B.Bargraphs representbestandworstSCTaveragedacrossparticipants,regardlessof
timelags.Errorbarsdisplaystandarderrorofthemean.TheasteriskindicatesasignificantdifferencebetweentheSCTofthe
bestperformanceandtheworstperformance.
8
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Therefore,forfurtheranalysesallparticipants’SCTsofallsixtimelagswerepeak-alignedsuchthat
thebestperformanceconditionwasatthethirdposition,originallycorrespondingto100ms(Figure3A,
leftpanel).Next,asinefit,aquadraticfit,andalinearfitwereperformedonthealigneddata,inorder
tounderstandthenatureofthetACSinducedmodulationofspeechcomprehension.
Figure3.Sinusoidal,quadratic,andlinearfits. A.Sinusoidalfit.Leftpanel:Theredcurvedisplaysgrandaveragesinusoidalfit.
Petrolcoloredcurvesaresingle-subjectz-transformedsentencecomprehensionthresholds(SCTs)inSNR.Bestperformanceis
alignedto0.1stimelag.Rightpanel.Redcurvesdisplayeachparticipantssinusoidalfit,petrolcoloredcurvesdisplayz-transformed
SCTs.B.Quadraticfit.AnalogoustoA.onlythattheredcurvedisplaysquadraticthefit.C.Linearfit.AnalogoustoA.andB.only
thattheredlinedisplaysthelinearfit.
The grand average and single-subject sinusoidal fits are displayed in Figure 3A. The average
goodness-of-fitisR2=0.73(sd=0.13).Parameterswereestimatedresultinginthisequation,averaged
acrossparticipants:𝑦 = 1.15 ∗ sin(6.1 ∗ 2𝜋 ∗ 𝑥 + 6.04).Thequadraticfithadanaveragegoodnessof-fitofR2=0.13(sd=0.11)withcoefficientsof𝑦 = 23.03𝑥 − 4.63𝑥(seeFigure3Bforgrandaverage
andsingle-subjectfits).Theaveragegoodness-of-fitofthelinearfitwasR2 =0.04(sd=0.04)withthe
followingcoefficient:𝑦 = 0.08𝑥.ThermANOVAacrosstheBICscoresofthethreefitsshowedthatthe
goodness-of-fitdifferedsignificantly(F(2,36)=41.09,p<0.0001).Post-hoct-testsrevealedthattheBIC
score of the sinusoidal fit was significantly lower than the scores of the quadratic and linear fit
(quadratic:t(18)=–7.18;p<0.0001;linear:t(18)=–5.95,p=0.0002).Thatmeansthatthegoodnessof-fitofthesinusoidalfitwassignificantlybetterthanthefitofthequadraticandlinearfunctions.Thus,
thesinewaverepresentsthemodulationofspeechintelligibilitybyenvelope-tACSbest.
Discussion
Theaimofthepresentstudywastoshowthatenvelope-tACSinauditorycortexmodulatessentence
comprehension.Weemployedaspeechintelligibilitytaskincombinationwithenvelope-tACSinduced
atdifferenttimelagstofindthedelaywiththehighestbehavioralbenefits.
The present study provides first evidence that sentences comprehension can be modulated by
envelope-tACSandthatthismodulationfluctuatessinusoidallyacrossdifferenttimelags.Thisfinding
supportspreviousclaimsontheimportantroleofcorticalentrainmentinauditorycortexforspeech
9
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
comprehension(forareview,seeZoefel&VanRullen,2015).Thesinusoidalfluctuationisinlinewith
theresultsofNeulingetal.,2012.Theywereabletoshowthatpure-tonedetectiondependsonthe
phaseofthetACSsignal.Thatis,theoptimaltACSphaseallowedincreasedexcitabilityofauditorycortex
and consequently improved auditory perception. Similarly, time lags in the present study varied to
assessatwhichtimelagtheenvelope-tACSsignalandthespeechenvelopearealignedandinphasein
auditorycortex.Thatway,tACSisassumedtoinduceaphaseresetatthepointintimewhenthespeech
signalreachesauditorycortexattheoptimalphase.Consequently,withadifferenttimelagtACSand
thespeechsignalwerenotaligned.Thatmostlikelydisruptedentrainmentofauditorycortextothe
speechenvelopeandtherebydisruptedspeechcomprehension.
However,thesinusoidalmodulationofspeechcomprehensionimpliesthatafteronecycleofthe
oscillationorsinecurve,tACSagainappearstobemorebeneficialforsentencecomprehension.Thisis
mostlikelyduetotheperiodicityofentrainedneuraloscillationsandthespeechenvelope.Thecircular
regularities of the envelope (speech as well as tACS) most probably induced an oscillatory neural
response in auditory cortex due to cortical entrainment of the envelope. Consequently, speech
comprehension, depending on entrainment to the envelope oscillates in this circular manner. The
average frequency of the fitted sine waves was 6.1 Hz which is within the range of the amplitude
modulationofspeechatfrequenciesthatarecrucialforintelligibility(4to16Hz;Drullmanetal.,1994;
Shannonetal.,1995).
Themedianofthetimelagswiththebestperformancewasat100msafterspeechonset.Thistime
lagcorrespondstothelatencyoftheauditoryN100,anegativeevokedpotentialelicitedbyauditory
stimulationafterapproximately100ms(forareview,seeNäätänen&Picton,1987).TheN100hasbeen
arguedtobeasignatureofauditorypatternrecognitionandintegration(Näätänen&Winkler,1999).
Although its characteristics such as amplitude, latency, and origin can be manipulated by different
featuresofastimulus,theN100remainsoneoftherobustinitialresponsestospeech(andnon-speech
sounds)inauditorycortex.Moreover,studiescomputingcrosscorrelationsbetweentheenvelopeofa
sentenceandtherespectiveEEGresponseorcomputingspectro-temporalresponsefunctionsfound
thepeakoftheeffectsatalatencyof100ms(Ding&Simon,2012;Hortonetal.,2013).DingandSimon
explainthatthislatencyindicatesthattheEEGresponsetoaspeechsignalisdrivenbytheamplitude
modulationsofthestimulus100msago.
Aiken and Picton (2008) report that their transient response models, another measure for the
relationshipbetweenspeechsignalandEEGresponse,peaksatalaterlatencyapproximatelyaround
180 ms. These divergent findings on the latency of the EEG response are in line with our findings.
Although,ourdatashowthatanenvelope-tACSlagof100mshasastrongsentencecomprehension
benefit,theindividuallatencyofbestcomprehensionperformancewasfoundatanyofthesixtimelags
10
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
rangingfrom0msto250ms.Itappearsasifthelatencyfromstimuluspresentationtoentrainmentin
auditorycortexvariesstronglyacrossindividuals.Thus,toestimatetheindividualtimelagofauditory
stimulationandtACS,furtherexperimentswillbeneededtomeasureindividualentrainmentlatencies
withEEGpriortoenvelope-tACSapplication.
Theimportanceofstudyingtranscranialelectricstimulation(tES)liesinitspossibilitiesofclinical
applications.DifferentstudieshavedemonstratedthediverseclinicalbenefitoftACS(forasummaryof
therapeuticalapplicationsoftES,seeVosskuhletal.,2015).Prehn-Kristensenetal.(2014),forexample,
showedthat0.75HztESimprovesslowwavesleeptherebyenhancingtheconsolidationofdeclarative
memory in children with attention-deficit/hyperactivity disorder. Furthermore, different studies
investigatedtheeffectofrepetitivetransorbitalalternatingcurrentstimulation(rtACS)onneurological
deficits.Here,thestimulatingelectrodesareplacedatorneartheeyeball.rtACShasbeenshownto
significantlyrestorevisioninopticneuropathy(Sabeletal.,2011)aswellasinstrokepatientssuffering
fromopticnervedamage(Galletal.,2016).Fedorovetal.(2010)alsoinducedrtACSonischemicstroke
patients and showed that non-invasive rtACS modulates brain plasticity and enhances neurological
recovery.tACShasalsobeensuccessfullyappliedinthetreatmentofParkinson’sdisease.Brittainetal.
(2013) were able to significantly reduce tremor by inducing indivudally adjusted tACS in- and out-of
phaseofthefrequencyofthetremorrhythm.Withadifferentapproach,Krauseetal.(2014)improved
motorperformanceinpatientssufferingfromParkinson’sdiseasebyinducing20-HztACStothemotor
cortex.
Alongtheselines,theresultsofthepresentstudypavethewayforanotherclinicalapplicationof
tACS.Thereporteddatasupportthehypothesisthatenvelope-tACSisbeneficialfortheintelligibilityof
sentenceswithabadsignal-to-noiseratio(SNR).PeoplesufferingfromhearinglossexperiencebadSNRs
constantly. For example, hearing aids are fitted to improve the SNR by filtering, amplifying, and
compressingacousticsignals.Digitalhearingaids,incontrasttoanaloghearingaids,evenrespondto
ongoinganalysisofthesignalandthebackgroundonamorecomplexlevel(forareview,seePichoraFuller&Singh,2006).Inadditiontotheconventionalmodeofoperationofhearingaids,envelope-tACS
couldbeappliedtocounteracthearingimpairment.Envelope-tACSwouldthenimprovetheneuralSNR
in auditory cortex by enhancing cortical entrainment to the relevant speech envelope and thereby
improving speech comprehension. However, cortical responses to speech in older adults, the main
audienceofhearingaidusers,needtobeinvestigatedinordertosuccessfullyimplementenvelopetACS
inhearingdevices.
11
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Conclusion
Altogether,inthisstudywehavedemonstratedthattACSindeedmodulatesintelligibilityofspeechin
noisepresumablybyenhancinganddisruptingcorticalentrainmenttothespeechenvelopeinauditory
cortex.WewereabletoshowthatthereisatimelagatwhichtACSappearstobemostbeneficialalbeit
theoptimaltimelagvariesbetweenparticipants.Moreover,wesuggestthatenvelopetACShasthe
potentialtocounteracthearingimpairmentwhenimplementedinhearingaids.
Authorcontributions
TN,CSH:designedthestudy;TN:acquiredthedata;TN,AW:analyzedthedata;AW,TN,CSH:wrotethe
article.
ConflictofInterestStatement
CSHhasappliedforapatentforthemethodofenvelope-tACSandhasreceivedhonorariaaseditorfrom
ElsevierPublishers,Amsterdam.
Acknowledgments
This research was funded by German Research Foundation (Deutsche Forschungsgemeinschaft, DFG
ClusterofExcellence1077“Hearing4all”).WethankJonasObleserforfruitfuldiscussions.
12
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
References
Abrams,D.A.,Nicol,T.,Zecker,S.,&Kraus,N.(2008).Right-HemisphereAuditoryCortexIsDominant
forCodingSyllablePatternsinSpeech.JournalofNeuroscience,28(15),3958–3965.
Ahissar,E.,Nagarajan,S.,Ahissar,M.,Protopapas,A.,Mahncke,H.,&Merzenich,M.M.(2001).
Speechcomprehensioniscorrelatedwithtemporalresponsepatternsrecordedfromauditory
cortex.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica,98(23),
13367–72.
Aiken,S.J.,&Picton,T.W.(2008).Humancorticalresponsestothespeechenvelope.EarandHearing,
29(2),139–157.
Brittain,J.-S.,Probert-Smith,P.,Aziz,T.Z.,&Brown,P.(2013).Tremorsuppressionbyrhythmic
transcranialcurrentstimulation.CurrentBiology,23(5),436–440.
Chandrasekaran,C.,Trubanova,A.,Stillittano,S.,Caplier,A.,&Ghazanfar,A.A.(2009).TheNatural
StatisticsofAudiovisualSpeech.PLoSComputationalBiology,5(7),e1000436.
Cherry,E.(1953).Someexperimentsontherecognitionofspeech,withoneandwithtwoears.The
JournaloftheAcousticalSocietyofAmerica,25(5),975–979.
Ding,N.,&Simon,J.Z.(2012).Neuralcodingofcontinuousspeechinauditorycortexduringmonaural
anddichoticlistening.JournalofNeurophysiology,107(1),78–89.
Ding,N.,&Simon,J.Z.(2013).AdaptiveTemporalEncodingLeadstoaBackground-InsensitiveCortical
RepresentationofSpeech.JournalofNeuroscience,33(13),5728–5735.
Ding,N.,&Simon,J.Z.(2014).Corticalentrainmenttocontinuousspeech:functionalrolesand
interpretations.FrontiersinHumanNeuroscience,8(May),311.
Doelling,K.B.,Arnal,L.H.,Ghitza,O.,&Poeppel,D.(2014).Acousticlandmarksdrivedelta-theta
oscillationstoenablespeechcomprehensionbyfacilitatingperceptualparsing.NeuroImage,85
Pt2,761–8.
Drullman,R.,Festen,J.M.,&Reiner,P.(1994).Effectoftemporalenvelopesmearingonspeech
reception.TheJournaloftheAcousticalSocietyofAmerica,95(2),1053–1064.
Engel,A.K.,Fries,P.,&Singer,W.(2001).Dynamicpredictions:oscillationsandsynchronyintop-down
processing.NatureReviews.Neuroscience,2(10),704–16.
Fedorov,A.,Chibisova,Y.,Szymaszek,A.,Alexandrov,M.,Gall,C.,&Sabel,B.A.(2010).Non-invasive
alternatingcurrentstimulationinducesrecoveryfromstroke.RestorativeNeurologyand
Neuroscience,28(6),825–833.
Fröhlich,F.,&McCormick,D.A.(2010).Endogenouselectricfieldsmayguideneocorticalnetwork
activity.Neuron,67(1),129–43.
Gall,C.,Schmidt,S.,Schittkowski,M.P.,Antal,A.,Ambrus,G.G.,Paulus,W.,…Sabel,B.A.(2016).
Alternatingcurrentstimulationforvisionrestorationafteropticnervedamage:Arandomized
clinicaltrial.PLoSONE,11(6),1–19.
Ghitza,O.,&Greenberg,S.(2009).Onthepossibleroleofbrainrhythmsinspeechperception:
intelligibilityoftime-compressedspeechwithperiodicandaperiodicinsertionsofsilence.
Phonetica,66(1–2),113–26.
Giraud,A.-L.,&Poeppel,D.(2012).Corticaloscillationsandspeechprocessing:emerging
computationalprinciplesandoperations.NatureNeuroscience,15(4),511–7.
Heimrath,K.,Fiene,M.,Rufener,K.S.,&Zaehle,T.(2016).ModulatingHumanAuditoryProcessingby
TranscranialElectricalStimulation.FrontiersinCellularNeuroscience,10(53),1–18.
Helfrich,R.F.,Knepper,H.,Nolte,G.,Strüber,D.,Rach,S.,Herrmann,C.S.,…Engel,A.K.(2014).
SelectiveModulationofInterhemisphericFunctionalConnectivitybyHD-tACSShapesPerception.
PLoSBiology,12(12).
Helfrich,R.F.,Schneider,T.R.,Rach,S.,Trautmann-Lengsfeld,S.A.,Engel,A.K.,&Herrmann,C.S.
(2014).Entrainmentofbrainoscillationsbytranscranialalternatingcurrentstimulation.Current
Biology,24(3),333–339.
Herrmann,C.S.,Rach,S.,Neuling,T.,&Strüber,D.(2013).Transcranialalternatingcurrent
stimulation:areviewoftheunderlyingmechanismsandmodulationofcognitiveprocesses.
FrontiersinHumanNeuroscience,7,1–13.
13
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Horton,C.,D’Zmura,M.,&Srinivasan,R.(2013).Suppressionofcompetingspeechthrough
entrainmentofcorticaloscillations.JournalofNeurophysiology,109(12),3082–3093.
Howard,M.F.,&Poeppel,D.(2010).Discriminationofspeechstimulibasedonneuronalresponse
phasepatternsdependsonacousticsbutnotcomprehension.JournalofNeurophysiology,
104(5),2500–11.
Krause,V.,Wach,C.,Südmeyer,M.,Ferrea,S.,Schnitzler,A.,&Pollok,B.(2014).Cortico-muscular
couplingandmotorperformancearemodulatedby20Hztranscranialalternatingcurrent
stimulation(tACS)inParkinson’sdisease.FrontiersinHumanNeuroscience,7(January),928.
Kubanek,J.,Brunner,P.,Gunduz,A.,Poeppel,D.,&Schalk,G.(2013).TheTrackingofSpeechEnvelope
intheHumanCortex.PLoSONE,8(1),e53398.
Lakatos,P.,Karmos,G.,Mehta,A.D.,Ulbert,I.,&Schroeder,C.E.(2008).Entrainmentofneuronal
oscillationsasamechanismofattentionalselection.Science,320(5872),110–3.
Lakatos,P.,Shah,A.S.,Knuth,K.H.,Ulbert,I.,Karmos,G.,&Schroeder,C.E.(2005).Anoscillatory
hierarchycontrollingneuronalexcitabilityandstimulusprocessingintheauditorycortex.Journal
ofNeurophysiology,94(3),1904–11.
Lalor,E.C.,&Foxe,J.J.(2010).Neuralresponsestouninterruptednaturalspeechcanbeextracted
withprecisetemporalresolution.EuropeanJournalofNeuroscience,31(1),189–193.
Luo,H.,&Poeppel,D.(2007).Phasepatternsofneuronalresponsesreliablydiscriminatespeechin
humanauditorycortex.Neuron,54(6),1001–1010.
Millman,R.E.,Johnson,S.R.,&Prendergast,G.(2015).TheRoleofPhase-lockingtotheTemporal
EnvelopeofSpeechinAuditoryPerceptionandSpeechIntelligibility.JournalofCognitive
Neuroscience,27(3),533–545.
Näätänen,R.,&Picton,T.W.(1987).TheN1WaveoftheHumanElectricandMagneticResponseto
Sound:AReviewandanAnalysisoftheComponentStructure.Psychophysiology,24(4),375–425.
Näätänen,R.,&Winkler,I.(1999).Theconceptofauditorystimulusrepresentationincognitive
neuroscience.PsychologicalBulletin,125(6),826–859.
Naue,N.,Rach,S.,Strüber,D.,Huster,R.J.,Zaehle,T.,Korner,U.,&Herrmann,C.S.(2011).Auditory
Event-RelatedResponseinVisualCortexModulatesSubsequentVisualResponsesinHumans.
JournalofNeuroscience,31(21),7729–7736.
Neuling,T.,Rach,S.,&Herrmann,C.S.(2013).Orchestratingneuronalnetworks:sustainedaftereffectsoftranscranialalternatingcurrentstimulationdependuponbrainstates.Frontiersin
HumanNeuroscience,7,161.
Neuling,T.,Rach,S.,Wagner,S.,Wolters,C.H.,&Herrmann,C.S.(2012).Goodvibrations:oscillatory
phaseshapesperception.NeuroImage,63(2),771–8.
Peelle,J.E.,&Davis,M.H.(2012).Neuraloscillationscarryspeechrhythmthroughtocomprehension.
FrontiersinPsychology,3,1–17.
Peelle,J.E.,Gross,J.,&Davis,M.H.(2013).Phase-LockedResponsestoSpeechinHumanAuditory
CortexareEnhancedDuringComprehension.CerebralCortex,23(6),1378–1387.
Pichora-Fuller,M.K.,&Singh,G.(2006).EffectsofAgeonAuditoryandCognitiveProcessing:
ImplicationsforHearingAidFittingandAudiologicRehabilitation.TrendsinAmplification,10(1),
29–59.
Prehn-Kristensen,A.,Munz,M.,Göder,R.,Wilhelm,I.,Korr,K.,Vahl,W.,…Baving,L.(2014).
Transcranialoscillatorydirectcurrentstimulationduringsleepimprovesdeclarativememory
consolidationinchildrenwithattention-deficit/hyperactivitydisordertoalevelcomparableto
healthycontrols.BrainStimulation,7(6),793–799.
Riecke,L.,Formisano,E.,Herrmann,C.S.,&Sack,A.T.(2015).4-HzTranscranialAlternatingCurrent
StimulationPhaseModulatesHearing.BrainStimulation,8(4),777–783.
Rufener,K.S.,Oechslin,M.S.,Zaehle,T.,&Meyer,M.(2016).TranscranialAlternatingCurrent
Stimulation(tACS)differentiallymodulatesspeechperceptioninyoungandolderadults.Brain
Stimulation,9(4),560–565.
Rufener,K.S.,Zaehle,T.,Oechslin,M.S.,&Meyer,M.(2016).40Hz-Transcranialalternatingcurrent
stimulation(tACS)selectivelymodulatesspeechperception.InternationalJournalof
14
bioRxiv preprint first posted online Jan. 3, 2017; doi: http://dx.doi.org/10.1101/097576. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
Psychophysiology,101(January),18–24.
Sabel,B.A.,Fedorov,A.B.,Naue,N.,Borrmann,A.,Herrmann,C.,&Gall,C.(2011).Non-invasive
alternatingcurrentstimulationimprovesvisioninopticneuropathy.RestorativeNeurologyand
Neuroscience,29(6),493–505.
Schroeder,C.E.,&Lakatos,P.(2009).Low-frequencyneuronaloscillationsasinstrumentsofsensory
selection.TrendsinNeurosciences,32(1),9–18.
Schroeder,C.E.,Wilson,D.A.,Radman,T.,Scharfman,H.,&Lakatos,P.(2010).DynamicsofActive
Sensingandperceptualselection.CurrentOpinioninNeurobiology,20(2),172–6.
Schwarz,G.(1978).Estimatingthedimensionofamodel.TheAnnalsofStatistics,6(2),461–464.
Shannon,R.V,Zeng,F.G.,Kamath,V.,Wygonski,J.,&Ekelid,M.(1995).Speechrecognitionwith
primarilytemporalcues.Science,270(5234),303–304.
Steinschneider,M.,Nourski,K.V,&Fishman,Y.I.(2013).Representationofspeechinhumanauditory
cortex:Isitspecial?HearingResearch,305(1),57–73.
Strüber,D.,Rach,S.,Trautmann-Lengsfeld,S.A.,Engel,A.K.,&Herrmann,C.S.(2014).Antiphasic40
HzOscillatoryCurrentStimulationAffectsBistableMotionPerception.BrainTopography,27(1),
158–171.
Vosskuhl,J.,Strüber,D.,&Herrmann,C.S.(2015).TranskranielleWechselstromstimulation.Der
Nervenarzt,1516–1522.
Wagner,K.,Kühnel,V.,&Kollmeier,B.(2001).EntwicklungundEvaluationeinesSatztestsfürdie
deutscheSprache,Teil1:DesigndesOldenburgerSatztests.ZeischriftFürAudiologie,1,4–15.
Wöstmann,M.,Fiedler,L.,&Obleser,J.(2016).Trackingthesignal,crackingthecode:speechand
speechcomprehensioninnon-invasivehumanelectrophysiology.Languange,Cognitionand
Neuroscience,1–15.
Zaehle,T.,Rach,S.,&Herrmann,C.S.(2010).TranscranialAlternatingCurrentStimulationEnhances
IndividualAlphaActivityinHumanEEG.PLoSONE,5(11),e13766.
Zoefel,B.,&VanRullen,R.(2015).TheRoleofHigh-LevelProcessesforOscillatoryPhaseEntrainment
toSpeechSound.FrontiersinHumanNeuroscience,9,1–12.
15