perubahan mekanisme pernafasan selama bicara

7/24/2019 Perubahan Mekanisme Pernafasan Selama Bicara

1/27

Skip to main content Other Journals from APS

Login

Search for this keyword

Advanced Search

HOME

ARTICLES

INFO FOR

EDITORS

SUBSCRIBE

SUBMIT

Changes to respiratory mechanisms during

speech as a result of different cues toincrease loudness

Jessica E. Huer! "harath #handrasekaran! John J. $olstencroftJournal of Applied Physiology Pulished % June &''( Vol.)* no. +! &%,,-&%* DOI:%'.%%(&/0applphysiol.'%&1).&''

Article

2igures 3 4ata 5nfo

E-letters

P42

Abstract

6he purpose of the present study was to determine whether di7erent cues toincrease loudness in speech result in di7erent internal targets 8or goals9 forrespiratory movement and whether the neural control of the respiratory

system is sensitive to changes in the speaker:s internal loudness target. 6hisstudy e;amined respiratory mechanisms during speech in 1' young adults at

Search
http://jap.physiology.org/content/98/6/2177#main-contenthttp://www.the-aps.org/mm/Publications/Journalshttp://jap.physiology.org/user?destination=fronthttp://jap.physiology.org/searchhttp://jap.physiology.org/http://jap.physiology.org/content/currenthttp://jap.physiology.org/abouthttp://ajpcell.physiology.org/editors-messagehttp://www.the-aps.org/mm/Publications/Purchasinghttp://jappl.msubmit.net/cgi-bin/main.plexhttp://jap.physiology.org/content/98/6/2177http://jap.physiology.org/content/98/6/2177.figures-onlyhttp://jap.physiology.org/content/98/6/2177.article-infohttp://jap.physiology.org/content/98/6/2177.e-lettershttp://jap.physiology.org/content/98/6/2177.full-text.pdf+htmlhttp://www.the-aps.org/mm/Publications/Journalshttp://jap.physiology.org/user?destination=fronthttp://jap.physiology.org/searchhttp://jap.physiology.org/http://jap.physiology.org/content/currenthttp://jap.physiology.org/abouthttp://ajpcell.physiology.org/editors-messagehttp://www.the-aps.org/mm/Publications/Purchasinghttp://jappl.msubmit.net/cgi-bin/main.plexhttp://jap.physiology.org/content/98/6/2177http://jap.physiology.org/content/98/6/2177.figures-onlyhttp://jap.physiology.org/content/98/6/2177.article-infohttp://jap.physiology.org/content/98/6/2177.e-lettershttp://jap.physiology.org/content/98/6/2177.full-text.pdf+htmlhttp://jap.physiology.org/content/98/6/2177#main-content


2/27

comfortale level and increased loudness levels. 5ncreased loudness waselicited using three methods< asking su0ects to target a speci=c soundpressure level! asking su0ects to speak twice as loud as comfortale! andasking su0ects to speak in noise. All three loud conditions resulted in similarincreases in sound pressure level . However! the respiratory mechanisms

used to support the increase in loudness di7ered signi=cantly depending onhow the louder speech was elicited. $hen asked to target at a particularsound pressure level! su0ects used a mechanism of increasing the lungvolume at which speech was initiated to take advantage of higher recoilpressures. $hen asked to speak twice as loud as comfortale! su0ectsincreased e;piratory muscle tension! for the most part! to increase thepressure for speech. However! in the most natural of the elicitation methods!speaking in noise! the su0ects used a comined respiratory approach! usingoth increased recoil pressures and increased e;piratory muscle tension. 5nnoise! an additional target! possily improving intelligiility of speech! wasre>ected in the slowing of speech rate and in larger volume e;cursions even

though the speakers were producing the same numer of syllales.

6HE E?A@5A65O O2 HO$the respiratory system supports speech production isa long-standing area of study. Previous studies have demonstrated that therespiratory system provides a steady driving pressure for speech productiony alancing recoil and muscular pressures in the system 83! 8! ! !3! !9.Loud speech reBuires a higher suglottic pressure. 6o achieve this! mostindividuals egin to speak at a higher lung volume 8LC9 to take advantage ofhigher recoil pressures 8"! 8! !9! although all speakers do not increase theirLC to the same degree 8!9. A second mechanism! more often used yspeakers closer to or elow end-e;piratory level 8EEL9! is to increasee;piratory muscle tension to achieve the reBuired higher driving pressures.However! not all studies have demonstrated such clear trends in respiratoryfunction for loud speech. 2or e;ample! $inkworth and 4avis 8#$9 reportedvariale respiratory patterns in =ve women speaking in noise. Speaking innoise is known to elicit the Lomard e7ect in which speakers naturally speaklouder under conditions of ackground noise 8!#! !89. 6he variaility in therespiratory patterns in the $inkworth and 4avis study was distinctly di7erentfrom previous studies that reported a pattern of initiating speech at higherLCs. 6he su0ects in the $inkworth and 4avis study increased their soundpressure level 8SPL9 y %'D%+ d"! which was at least as much as theincreases in SPL reported in previous studies of respiratory patterns.$inkworth and 4avis 8#$9 hypothesied that the di7erences etween their

=ndings and earlier =ndings were due to the method used to elicit louderspeech. 6he ulk of the data on respiratory kinematics and increasingloudness have used a small set of cues< 19 asking su0ects to speak twice orfour times as loud as comfortale 8"9 and 29 asking individuals to target aspeci=c SPL 8usually (D%' d" higher than their comfortale level9 on an SPLmeter 8!9. #ues to speak twice or four times as loud have resulted inincreased LC initiations 8LC5s9 8"9! and targeting cues have resulted inincreased volume initiations and larger volume e;cursions 8!9.

6he type of cue used to elicit an increase in loudness has not een studiedpreviously mainly ecause of the assertion that respiratory kinematics do notchange as a result of how an individual is cued to increase loudness. 6he

=ndings from $inkworth and 4avis 8#$9 suggest that this may not e true.6heir data demonstrate that multiple strategies may e employed y
http://jap.physiology.org/content/98/6/2177#ref-3http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-9http://jap.physiology.org/content/98/6/2177#ref-13http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-4http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#ref-12http://jap.physiology.org/content/98/6/2177#ref-18http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#ref-4http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-4http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#ref-3http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-9http://jap.physiology.org/content/98/6/2177#ref-13http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-4http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#ref-12http://jap.physiology.org/content/98/6/2177#ref-18http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#ref-4http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-4http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-20


3/27

speakers to speak louder when they are not speci=cally instructed to increasetheir loudness.

6here are many potential di7erences etween instructing a su0ect to speaklouder and placing a su0ect in an environment that automatically triggers anincrease in loudness. 5nstructing su0ects to speak louder may cause them to

focus on eing loud more so than they would in a natural situation! such aswhen louder speech is reBuired to avoid communicative reakdown 8in anoisy environment9. Speaking in noise may involve communication goals notelicited y an instruction to increase loudness. 6hese additional goals mayinclude increasing the intelligiility of the speech signal to improve itstransmission through noise 8!#9. 2inally! speakers may perceive the diFcultyof the goal of increasing loudness di7erently depending on the situation."eing asked to increase loudness may e perceived as more diFcult thanspeaking in noise! in which the increase in loudness is viewed as moreautomatic 8!89. Additionally! speaking at %' d" aove comfortale 8typically*(D)' d" SPL9 may sound harder to su0ects than speaking twice as loud as

their comfortale loudness level.Any of the aove di7erences could potentially change the speaker:s internaltarget. An internal target is the neural representation of the desiredmovement parameters and re>ects the goals of the individual:s movements.5n the present study! the goal is to e louder. 6he internal target results in ascaling up of the movement plans that go out to the muscles in the periphery.5t is possile that the way an individual is cued may a7ect internal targetsand! therefore! how the movement parameters are modi=ed. 2or e;ample! ina noisy environment! two targets may e;ist< to increase loudness and to slowspeech to improve transmission of the message. 5n this case! the internaltarget may e for louder and slower speech! and movement parameters

would e modi=ed to e oth scaled up and slowed down.

Studies of lim movement have demonstrated that changes to peripheralmovements provide information aout internal targets 8%9. 2or e;ample! inthe study y Gentilucci and colleagues 8%9! reaching and grasping movementwas a7ected y the lael on the o0ect! such as largeI or small!I regardlessof the actual sie of the o0ect. 5n this case! automatic word reading had agreater e7ect on the internal target than the actual perception of the o0ect!and the internal target was re>ected in the movement parameters forreaching and grasping. "ecause the internal target of a movement isre>ected in the kinematics associated with achieving that target! it is possilethat the movements of the respiratory system will e altered ased onchanges to the internal target associated with speci=c cues to increaseloudness.

6he purpose of the present study was to determine whether di7erent cues toincrease loudness will result in di7erent respiratory kinematic patterns.E;amination of mechanisms for increasing loudness under di7erent cues isimportant from a motor control perspective since a greater understanding ofthe respiratory system:s sensitivity to internal targets will enhance ourknowledge of its role in speech production. niBue kinematic patterns! asedon cue! would support the view that cues create di7erent internal targets andthat the neural control of the respiratory system is sensitive to changes ininternal targets for speech.
http://jap.physiology.org/content/98/6/2177#ref-12http://jap.physiology.org/content/98/6/2177#ref-18http://jap.physiology.org/content/98/6/2177#ref-7http://jap.physiology.org/content/98/6/2177#ref-7http://jap.physiology.org/content/98/6/2177#ref-12http://jap.physiology.org/content/98/6/2177#ref-18http://jap.physiology.org/content/98/6/2177#ref-7http://jap.physiology.org/content/98/6/2177#ref-7


4/27

6he aim of the present study is also important from a clinical perspective inspeech therapy. 6here are a numer of speech disorders that result inreduced loudness levels! such as Parkinson:s disease. 5ndividuals withsymptoms of low vocal loudness often undergo therapies that cue them! inmany ways! to increase their loudness. "ecause the respiratory system plays

a primary role in increasing loudness! an understanding of how cues a7ectrespiratory function in normal speakers will assist in choosing the est cuesfor treatment.

MATERIALS AND METHODS

Subjects

6hirty normal young adults! %( women and %( men! participated in the study.6he mean age of the women was && yr! mo! and the mean age of the men

was && yr! %' mo. Speakers were grouped y se; ecause previous studieshave demonstrated di7erences in respiratory kinematics during speechetween the se;es 8!$! !9.Su0ects indicated that they had 19 no history of voice or respiratoryprolems 8including asthma9! neurological disease! or head or necksurgeryK 29 never received formal speaking or singing trainingK 39 no recentcolds or infectionsK and 49 een nonsmoking for the past ( yr. 6hey had aody mass inde; etween %) and 1' as measured on the day of testing 8&9!normal speech! language! and voice! and spoke General orth Americandialect of English. 6hey had normal hearing as indicated y a hearingscreening at 1' d" hearing level for octave freBuencies etween &(' and*!''' H! ilaterally! completed in a Buiet room. Su0ects had normal vitalcapacity 8C#9! forced C#! and forced e;piratory volume in % s de=ned as*'M of e;pected values ased on age! gender! height! weight! andethnicity. Lung capacities were measured using a digital spirometer 8SPKCacu@ed 4iscovery Handheld SP9.

Procedures and Speech Stimuli

Procedures for data collection were approved y Purdue niversity:s#ommittee on the se of Human Nesearch Su0ects. Su0ects said twosentences< 19 "uy "oy a puppyI and 29 ou uy "oy a puppy now if he

wants one.I se of two sentences of speci=ed length ensured that utterancelength was controlled and was not a factor in the kinematic measurements.Su0ects were instructed to say one sentence per reath and to speak clearlyand audily in each condition. 6he e;perimenter was visile to the su0ectduring all conditions and was seated aout *' in. away. 6he conditions wereas follows.

Comfortable level.

Su0ects were instructed to read the sentence at your comfortale loudnessand pitchI 8#O@29.

COMF +10.
http://jap.physiology.org/content/98/6/2177#ref-10http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-5http://jap.physiology.org/content/98/6/2177#ref-10http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#ref-5


5/27

Su0ects were instructed as follows< 6he numer goes up as you get louder.$hen you read the sentence this time! 5 want you to keep that numeretween ?? and ??.I 6he SPL targets were inserted for the ?? in theinstruction. 6he SPL targets for this condition were set at %' d" 8& d"9aove the su0ect:s comfortale SPL. 6he SPL meter was set at fast response.

6he output from the SPL meter was enlarged and pro0ected onto a televisionscreen that faced the su0ect. 2eedack was provided continuously to thesu0ect.

2COMF.

Su0ects were instructed to read the sentence at what you feel is twice yourcomfortale loudness.I o feedack was provided during this condition. 5fthis condition did not follow the #O@2 condition! su0ects were instructed toread the sentence at your comfortale loudness and pitchI until their SPLlevel was similar to their original SPL level for the #O@2 condition efore

eing given the twice-as-loud cue. 6he decision of similarity etween the SPLlevels was made y the e;aminer using the SPL meter.

NOISE.

@ulti-talker noise 8A4i6E#! St. Louis! @O9 was turned on in the room. 6henoise was delivered at ,' d"A relative to the su0ect:s ears via free =eld.Su0ects were instructed to read the sentence.I o cue was given regardingloudness in this condition. 6he speakers used to deliver the noise were placed1) in. in front of the su0ects.

2or each condition! the shorter sentence was completed =rst! and eachsentence was said %( times consecutively. 6he #O@2 condition was alwayscompleted =rst. 6he order of the three loud conditions 8#O@2 Q %'! &R#O@2! and O5SE9 was counteralanced across su0ects.

Su0ects also produced ma;imum capacity tasks 89. 6hese tasks were usedto otain an estimate of the ma;imal capacity of the lungs! ri cage 8N#9! andadomen 8A9. @easurements during speech production were e;pressed as apercent of capacity so that comparisons could e made across individuals ofdi7ering sies. 5n all cases! su0ects were e;pected to produce threecomparale ma;imum capacity tasks. C# measured as a part of the inclusion

criteria was used! and LC measures were e;pressed as a percentage of thelargest C# produced. C# maneuvers were also completed at least three timeswith the Nespitrace ands in place. 6hese trials were in addition to those usedto otain C# for su0ect inclusion and were used to otain N# capacity 8N##9since the N# moves ma;imally during a C# maneuver. 6o determineadominal 8A9 capacity 8A#9! two maneuvers were completed! at leastthree times each! with the Nespitrace ands in place. 2irst! the su0ect wasinstructed to hold his/her reath at EEL and suck his/her stomach inma;imally. Second! the su0ect was instructed to hold his/her reath at EELand e;tend his/her stomach ma;imally. 6he comination of the ma;imum inand the ma;imum out were taken to e total A#. At least three steady

cycles of rest reathing 8N"9 were collected efore the start of each trial ofthe ma;imum capacity maneuvers.
http://jap.physiology.org/content/98/6/2177#ref-9http://jap.physiology.org/content/98/6/2177#ref-9


6/27

Equipment

6he acoustic signal was transduced via a condenser microphone that wasconnected to an SPL meter 8uest model %,''9. 6he microphone was placed+ in. from the su0ect:s mouth at a (T angle. 6he microphone signal was

recorded to digital audiotape and later digitied into a personal computerusing Praat 8!9. 6he signal was digitied at .% kH and resampled at %*kH. 6he resampling process applied a low-pass =lter at )!''' H forantialiasing.Nespiratory kinematic data were transduced via respiratory inductiveplethesmography using the Nespitrace system 8Amulatory @onitoring9. Anelastic and was placed around the N#! 0ust under the a;illa! to transducemovements of the N#. A second elastic and was placed around the adomenat the level of the umilicus! ensuring that it was elow the last ri! totransduce movements of the adomen. Nespiratory kinematic data weredigitied at &!''' H. 4ata from a second microphone were collected with

these data so that an acoustic record would e digitied in comination withthe respiratory kinematic data.

Measurements

6he =rst two trials of each sentence in each condition were discarded. 6hene;t %' consecutive sentences that were produced without error were chosenfor analysis. SPL was measured using Praat 8!9 for each sentence.

Respiratory kinematic measurements.

Nespiratory kinematic measurements were made using algorithms written in@atla 8@ath$orks9. "efore any measurements were made! the respiratorykinematic signals were low-pass =ltered at ' H to remove noise.

"ecause LC change re>ects the comined e7ect of changes in the N# and Avolumes 8!!9! the sum of the N# and A signals was computed and correctedfor the respective N# and A contriutions to LC change. 6wo nonspeechtasks were used to determine the N# and A contriutions to LC change.Su0ects were instructed to rela;! and data were collected for two (-speriods of N". 4ata was also collected for three (-s periods of speech-likeIreathing 8SL"9. 2or this task! su0ects were instructed to read the longersentence silently to themselves! one time per reath. At least three steadycycles of N" were collected efore the start of the each SL" data collectionperiod. LC data was collected during the N" and SL" tasks using a SP8Cacu@ed niversal Centilation @eter9. 6his SP has a very small dead space.

6hese data were digitied along with the respiratory kinematic data at &!'''H.

6he data from the SP! N#! and A signals during the N" and SL" tasks wereused to determine the correction factors for the N# and A. 6he @oore-Penrose pseudoinverse function was used in @atla to determine the leasterrored solution for the correction factors 8k%and k&9. 6he psuedoinversefunction solved for k%and k&in the formula SP U k%8N#9 Q k&8A9 for each set ofN#! A! and SP data points in the N" and SL" tasks. 6his estimation of the LCsignal was veri=ed y visually checking the S@ signal Vk%8N#9 Q k&8A9W
http://jap.physiology.org/content/98/6/2177#ref-1http://jap.physiology.org/content/98/6/2177#ref-1http://jap.physiology.org/content/98/6/2177#ref-11http://jap.physiology.org/content/98/6/2177#ref-1http://jap.physiology.org/content/98/6/2177#ref-1http://jap.physiology.org/content/98/6/2177#ref-11


7/27

against the original SP signal for a SL" trial. 6he estimated LC signal was thencomputed for each point during the sentence production tasks using theformula LC U k%8N#9 Q k&8A9.LC! N#! and A initiations and terminations were measured relative to EEL.EEL was measured from troughs of three steady N" efore the start of each

set of sentence repetitions. LC! N#! and A initiations and terminations weree;pressed as a percent of C#! N##! and A#! respectively. Speech initiationswere de=ned as the point where voicing egan! and speech terminationswere de=ned as the point where voicing ended! as indicated y themicrophone signal collected with the kinematic data 8see Fi'. !! lines

Aand B9. 6he audio signal was used to verify that the initiations andterminations were selected accurately and that no part of the speech signalwas cut o7. LC! N#! and A e;cursions were calculated as the volume atinitiation minus the volume at termination and e;pressed as a percent of C#!N##! and A#! respectively 8see Fi'. !! line Ato line B9.

Do(nlo)* +',re

O-en in ne( t)

Do(nlo)* -o(er-oint

Fi'. !.

@easurement points for the respiratory kinematic traces. 6race from male speaker during

the comfortale condition! longer sentence. Top< line Ais the point where lung 8LC9! ri

cage 8N#9! and adominal 8A9 volume initiations were measured. Line Bis the point

where LC! N#! and A volume terminations were measured. Bottom


8/27

waveform9 is the volume at the end of e;piration for the previous utterance. Point Dis

the top of inspiration for the current utterance.

Percent C# e;pended per syllale was measured y dividing the LC e;cursiony the numer of syllales for each utterance. Percent C# inspired was

measured from the LC signal y sutracting the volume at the end ofe;piration after the previous utterance from the volume at the top ofinspiration efore the current utterance 8see Fi'. !!point Dtopoint C9.

Timing Measurements

4uration was measured as the time etween speech initiation and speechtermination of each utterance. Syllales per second was measured ydividing the duration of the utterance y the numer of syllales produced.

6he phonation onset time was de=ned as the time from the end of inspirationto the start of speech for the sentence 8see Fi'. !!point Dto line A9.

Statistics

@eans were computed for each su0ect for each condition. 6he di7erences inthe means were assessed in two-factor repeated-measures AOCA. 6hewithin factor was loudness 8condition9! and the etween factor wasse;. T)le. !provides a summary of the statistical results. 6ukey:s honestlysigni=cant di7erence tests were completed for all factors and interactionsthat were signi=cant in the AOCA. 6he alpha level for the AOCAs and the

6ukey:s honestly signi=cant di7erence tests was set at PX '.'%.

Vie( -o-,-

Vie( inline

T)le !.

Statistical summary for condition, se, and condition !y se interactions

6o test for a learning e7ect across the %' trials used for analysis! a matched-pairs t-test was computed etween the mean of the =rst three trials and themean of the last three trials for each measurement. 6he alpha level was setat PX '.'%! as it was for the AOCAs. 6here were no signi=cant di7erencesetween the two means for any of the measurements! suggesting that therewas no signi=cant learning e7ect.

5ntermeasurer reliaility was completed on two male and two femalesu0ects! randomly chosen. 5ndependent t-tests were computed etween the=rst and second measurement for each variale. one of the alpha levelsneared signi=cance! ranging from PU '.%&* to PU '.**%! indicating goodintermeasurer reliaility.

RESULTS

2or SPL! there was a signi=cant condition e7ect ut no signi=cant se; or

interaction e7ects. 6he three loud conditions were produced at a signi=cantlyhigher SPL than the #O@2 condition. 6here were no signi=cant di7erences in
http://jap.physiology.org/content/98/6/2177#F1http://jap.physiology.org/content/98/6/2177#F1http://jap.physiology.org/content/98/6/2177#T1http://jap.physiology.org/highwire/markup/90616/expansion?width=1000&height=500&iframe=true&postprocessors=highwire_figures%2Chighwire_mathhttp://jap.physiology.org/content/98/6/2177#%23http://jap.physiology.org/content/98/6/2177#F1http://jap.physiology.org/content/98/6/2177#F1http://jap.physiology.org/content/98/6/2177#T1http://jap.physiology.org/highwire/markup/90616/expansion?width=1000&height=500&iframe=true&postprocessors=highwire_figures%2Chighwire_mathhttp://jap.physiology.org/content/98/6/2177#%23


9/27

SPL for the three loud conditions. 6he mean SPLs for the conditions were ,)d" 8SE U '.(( d"9 for #O@2! *) d" 8SE U '.+% d"9 for #O@2 Q %'! ** d" 8SEU '.,1 d"9 for &R #O@2! and )' d" 8SE U '.,% d"9 for O5SE.

Timing Measurements

2or duration! there was a signi=cant condition e7ect ut no signi=cant se; orinteraction e7ects. 6he sentences produced in the O5SE condition wereproduced over a signi=cantly longer duration than those in the #O@2condition 8see Fi'. #9.

Do(nlo)* +',re

O-en in ne( t)

Do(nlo)* -o(er-ointFi'. #.

6iming measurements< mean di7erence from comfortale speaking level 8#O@29 for the

1 loud conditions V#O@2 Q %' d" 8#O@2 Q %'9! twice the #O@2 volume 8&R #O@29! and

speaking over ackground noise 8O5SE9W. Lines show standard errors. YSigni=cant

changes from #O@2 8PX '.'%9.

2or syllales per second! there was a signi=cant condition e7ect ut no se; orinteraction e7ects. Signi=cantly fewer syllales per second were produced inthe O5SE condition compared with the #O@2 condition 8see Fi'. #9.2or phonation onset time! there was a signi=cant condition e7ect ut no se;or interaction e7ects. 6he phonation onset time was signi=cantly shorter inthe #O@2 Q %' and &R #O@2 conditions compared with the #O@2 condition8see Fi'. #9.

Respiratory Kinematic Measurements

2or LC5! there were signi=cant condition and se; e7ects ut no interactione7ect. LC5 was signi=cantly higher in the #O@2 Q %' condition compared withthe #O@2 condition 8see Fi'. 39. LC5 was signi=cantly higher for the womencompared with the men.
http://jap.physiology.org/content/98/6/2177#F2http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F2.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F2.large.jpghttp://jap.physiology.org/highwire/powerpoint/90590http://jap.physiology.org/content/98/6/2177#F2http://jap.physiology.org/content/98/6/2177#F2http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F2http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F2.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F2.large.jpghttp://jap.physiology.org/highwire/powerpoint/90590http://jap.physiology.org/content/98/6/2177#F2http://jap.physiology.org/content/98/6/2177#F2http://jap.physiology.org/content/98/6/2177#F3


10/27

Do(nlo)* +',re

O-en in ne( t)

Do(nlo)* -o(er-oint

Fi'. 3.

5nitiation and termination measurements< mean di7erence from #O@2 for the 1 loud

conditions 8#O@2 Q %'! &R #O@2! and O5SE9. Lines show standard errors. LC5! lung

volume initiationK LC6! lung volume terminationK N#C5! ri cage volume initiationK N#C6!

ri cage volume terminationK A"C5! adominal volume initiationK A"C6! adominal volume

termination. YSigni=cant changes from #O@2 8PX '.'%9.

2or LC termination 8LC69! there were signi=cant condition and se; e7ects utno interaction e7ect. LC6 was signi=cantly higher for the #O@2 Q %'condition compared with the #O@2 condition 8seeFi'. 39. LC6 wassigni=cantly higher for the women than for the men.2or LC e;cursion 8LCE9! there was a signi=cant condition e7ect ut no se; orinteraction e7ects. LCE was signi=cantly larger in the O5SE condition than inthe #O@2 condition 8see Fi'. "9.

Do(nlo)* +',re

O-en in ne( t)
http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F3.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F3.large.jpghttp://jap.physiology.org/highwire/powerpoint/90593http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F4http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F4.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F4.large.jpghttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F3.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F3.large.jpghttp://jap.physiology.org/highwire/powerpoint/90593http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F4http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F4.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F4.large.jpg


11/27

Do(nlo)* -o(er-oint

Fi'. ".

E;cursion measurements< mean di7erence from #O@2 for the 1 loud conditions 8#O@2 Q

%'! &R #O@2! and O5SE9. Lines show standard errors. LCE! lung volume e;cursionsK

N#CE! ri cage volume e;cursionK A"CE! adominal volume e;cursion. YSigni=cantchanges from #O@2 8PX '.'%9.

2or percent of C# inspired! there was a signi=cant condition e7ect ut no se;or interaction e7ects. 6he percent of C# inspired was signi=cantly larger inthe O5SE condition than in the #O@2 and #O@2 Q %' conditions 8see Fi'.&9.

Do(nlo)* +',re

O-en in ne( t)

Do(nlo)* -o(er-oint

Fi'. &.

Percent vital capacity 8C#9 e;pended and percent C# inspired measurements< mean

di7erence from #O@2 for the 1 loud conditions 8#O@2 Q %'! &R #O@2! and O5SE9. Lines

show standard errors. YSigni=cant changes from #O@2 8PX '.'%9.

2or percent C# e;pended per syllale! there was a signi=cant condition e7ectut no se; or interaction e7ects. Percent C# e;pended per syllale wassigni=cantly higher in the O5SE condition compared with the #O@2 condition8see Fi'. &9.2or N# volume initiation 8N#C59! there was a signi=cant condition e7ect ut nose; or interaction e7ects. N#C5 was signi=cantly higher in the #O@2 Q %' andO5SE conditions compared with #O@2 condition 8see Fi'. 39.2or N# volume termination 8N#C69! there was a signi=cant condition e7ect utno se; or interaction e7ects. N#C6 was signi=cantly higher in the #O@2 Q %'condition compared with the #O@2 condition 8see Fi'. 39.2or N# volume e;cursion 8N#CE9! there was a signi=cant condition e7ect utno se; or interaction e7ects. N#CE was signi=cantly larger in all loudconditions compared with the #O@2 condition 8see Fi'. "9.2or A volume initiation 8AC59! there was a signi=cant condition e7ect ut no

se; or interaction e7ects. AC5 was signi=cantly lower in the &R #O@2condition compared with the #O@2 condition 8see Fi'. 39.
http://jap.physiology.org/highwire/powerpoint/90602http://jap.physiology.org/content/98/6/2177#F5http://jap.physiology.org/content/98/6/2177#F5http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F5.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F5.large.jpghttp://jap.physiology.org/highwire/powerpoint/90607http://jap.physiology.org/content/98/6/2177#F5http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F4http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/highwire/powerpoint/90602http://jap.physiology.org/content/98/6/2177#F5http://jap.physiology.org/content/98/6/2177#F5http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F5.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F5.large.jpghttp://jap.physiology.org/highwire/powerpoint/90607http://jap.physiology.org/content/98/6/2177#F5http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F4http://jap.physiology.org/content/98/6/2177#F3


12/27

2or A volume termination 8A"C69! there was a signi=cant condition e7ect utno se; or interaction e7ects. AC6 was signi=cantly lower in the &R #O@2condition! compared with the #O@2 condition 8see Fi'. 39.2or A volume e;cursion! there were no condition! se;! or interaction e7ects8see Fi'. "9.

DISCUSSION

6he purpose of the present study was to test whether the kinematic patternsof the chest wall will di7er ased on how speakers were cued to increaseloudness. All three loud conditions resulted in similar SPL increases< Z10 dB.However, the respiratory mechanisms used to increase loudness differed depending howthe increase in loudness was elicited. 6he di7erent kinematic patternssuggest that the cues resulted in di7erent internal targets and that the neural

control of the respiratory system for speech is a7ected y changes to thespeaker:s internal loudness target.

5n the #O@2 Q %' condition! su0ects primarily used a mechanism ofeginning to speak at a higher LC to utilie higher recoil pressures. LC and N#volume initiations and terminations were all signi=cantly higher in the #O@2Q %' condition than in the #O@2 condition. 6hese results are directly in linewith the =ndings from Stathopoulos and Sapiena 8!9 who used the samecueK however! they asked su0ects to target a loudness ( d" SPL higher thancomfortale. Phonation onset time was shorter than in the #O@2 condition!indicating that su0ects egan speaking closer to the top of inhalation in the

#O@2 Q %' condition. 6he mechanism of employing higher recoil pressureswas used to the greatest e;tent in the #O@2 Q %' condition compared withthe other two loud conditions! as demonstrated y the fact that the changefrom #O@2 for LC and N# volume initiations and terminations was greatest inthe #O@2 Q %' condition 8see Fi'. 39. 2urthermore! the #O@2 Q %' conditionwas the only loud condition in which LC5s and LC6s and N#C6s weresigni=cantly higher than in #O@2. se of higher LC and N# volume initiationsand terminations was the predominant pattern for most of the trials in the#O@2 Q %' condition 8seeFi'. /9.

Do(nlo)* +',re

O-en in ne( t)

Do(nlo)* -o(er-oint

Fi'. /.
http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F4http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F6.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F6.large.jpghttp://jap.physiology.org/highwire/powerpoint/90612http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F4http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F6.large.jpg?download=truehttp://d13geadg2uyg93.cloudfront.net/content/jap/98/6/2177/F6.large.jpghttp://jap.physiology.org/highwire/powerpoint/90612


13/27

Histogram of respiratory kinematic initiations and terminations showing the numer of

trials in which the #O@2 condition was higher or lower than the #O@2 Q %'! &R #O@2!

and O5SE conditions! respectively.

5n the O5SE condition! N#C5 and percent C# inspired were higher than in the#O@2 condition. 6hese results suggest that speech was initiated at a higherLC! even though the LC5 results were nonsigni=cant. Additionally! most trialsdemonstrated the pattern of increased LC and N# volume initiations andterminations during the O5SE condition 8see Fi'. /9. 6here was less changefrom #O@2 in LC5s and N#C5s in the O5SE condition than in the #O@2 Q %'condition! indicating that increased recoil pressure was used in the O5SEcondition ut not to the e;tent it was used in the #O@2 Q %' condition8see Fi'. 39.5n the &R #O@2 condition! LC and N# volume initiations and terminations didnot change relative to the #O@2 condition! indicating that the lungs and N#were not more e;panded when speech was initiated or terminated. 6herefore!

the primary mechanism for increasing loudness in the &R #O@2 conditioncould not have een the use of higher recoil pressures and higher LCs.However! as in the #O@2 Q %' condition! phonation onset time wassigni=cantly shorter in the &R #O@2 condition compared with the #O@2condition! indicating that the su0ects egan to speak closer to the top ofinhalation in the &R #O@2 condition than in the #O@2 condition.

6he larger reliance on the use of the higher recoil pressures in the #O@2 Q %'condition may have een a result of su0ect perception. Su0ects may haveperceived maintaining an SPL at nearly )' d" as diFcult and planned inadvance to achieve this goal. Su0ects may not have perceived the loudness

target in the &R #O@2 or O5SE conditions to e as high as the level in the#O@2 Q %' condition and! therefore! may not have planned to need as muchrespiratory driving pressure even though they produced similar SPLs in allconditions. 6hese data suggest that how the loudness target was perceivedy the speaker a7ected the mechanisms used to support the loud speech andthe neural control of the respiratory system.

Su0ects did not utilie increased recoil pressures in the &R #O@2 conditionand utilied them to a lesser degree in the O5SE condition than in the #O@2Q %' condition. However! they achieved the same overall increase in SPL.

6herefore! another mechanism must have een used in the &R #O@2 and

O5SE conditions. 5n the &R #O@2 condition! AC5 and AC6 were signi=cantlylower than in the #O@2 condition! indicating that the A was morecompressed during the &R #O@2 condition. "ecause the A muscles are oneof the ma0or e;piratory muscle groups 8#! 39! the tucked position of the Asuggests that the speakers generated higher e;piratory muscle forces usingtheir A muscles. 6he use of e;piratory muscle tension to generate higherpressure for louder speech was most prevalent in the &R #O@2 condition. 6he&R #O@2 condition was the only loud condition in which there weresigni=cant changes from #O@2 in the A measurements 8see Fi'. 39.2urthermore! the pattern of decreased AC5 and AC6 was demonstrated formost trials in the &R #O@2 condition! more than in the other two loud

conditions 8see Fi'. /9.
http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#ref-2http://jap.physiology.org/content/98/6/2177#ref-3http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#ref-2http://jap.physiology.org/content/98/6/2177#ref-3http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6


14/27

5n the O5SE condition! AC5 and AC6 were lower than in the #O@2 or #O@2Q %' conditions with AC5 almost as low as in the &R #O@2 condition!although the changes were nonsigni=cant 8see Fi'. 39. 6his suggests thatsu0ects did use the mechanism of increasing e;piratory muscle tension inthe O5SE condition ut not to the e;tent it was used in the &R #O@2

condition. 2urthermore! lower AC5 and AC6 were used on nearly as manytrials in the O5SE condition as in the &R #O@2 condition 8see Fi'. /9.6he iggest di7erence in engaging the adomen etween the &R #O@2 andO5SE conditions is present in the AC6 data. 6he change in AC6 from #O@2is much greater for &R #O@2 than O5SE! whereas the change in AC5 from#O@2 is more similar etween the two conditions 8seeFi'. 39. 6he largedecrease in AC6 in the &R #O@2 condition may suggest that su0ectsunderestimated the amount of pressure that would e reBuired to achieve theloudness target. 6hey may have realied the need for more driving pressureas they moved through the utterance and used the A to generate highere;piratory muscle pressures. 6his is sustantiated y the fact that the LC5 and

N#C5 were lowest for the &R #O@2 condition! indicating that su0ects had lessrecoil pressure availale in the &R #O@2 condition than in the other two loudconditions 8see Fi'. 39.5t is interesting to note that there were no clear trends in the use of the A forthe #O@2 Q %' condition. one of the A volume measurements 8initiations!terminations! and e;cursions9 changed signi=cantly from #O@2 to #O@2 Q%'! and there is no clear trend in the trial data for the A measurements8see Fi'. /9. 6his =nding is in line with previous studies that havedemonstrated no contriution from the A toward increasing loudness88! !9. However! given the =ndings for the &R #O@2 and O5SE conditions inthe present study! the conclusion that the A does not participate in

increasing loudness appears to e related to how the increase in loudness iselicited.5n addition to the di7erences in how the increase in loudness was supportedy the respiratory system for each condition! the O5SE condition stood outas di7erent from the #O@2 Q %' and &R #O@2 conditions in a numer ofways. 2irst! the O5SE condition was the only loud condition in whichutterance duration was signi=cantly longer! fewer syllales per second wereproduced! larger LCE were used! and a higher percent of C# was e;pendedper syllale compared with #O@2. "ecause the length of the utterances didnot change across any of the conditions! all of these =ndings can eaccounted for y a slower! more delierate speech rate. Speakers may have

perceived the need to use a slower speech rate in the O5SE condition toimprove the intelligiility of the speech signal in the noise. 6hese datasuggest that goals for speech production were more comple; in the O5SEcondition than in the other two loud conditions! i.e.! to improve intelligiilityin addition to increasing loudness. 6his target for improved intelligiility inthe O5SE condition was re>ected in the respiratory kinematics.

Second! in the O5SE condition! the su0ects comined the two proposedmechanisms for increasing loudness! higher recoil pressures! and moree;piratory muscle tension. 6he use of a comined strategy may relate to thenaturalness of the cue. 6he most natural cue was the O5SE condition sincethe su0ects in the study had! presumaly! spoken in noise previously in thecourse of their daily life. 6he &R #O@2 condition may seem naturalK however!speakers seldom think aout douling their loudness without additional
http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-19http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F3http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-19


15/27

environmental cues as to how much loudness increase is reBuired 8speaker-listener distance! increased room sie! etc.9. 6herefore! the &R #O@2 cue isnot as natural as the O5SE cue. 6he comined respiratory strategy wasdemonstrated to the greatest e;tent in the O5SE condition.

se of a comination of the two mechanisms may e advantageous forsu0ects since it may e less work for the system than using predominantlyone mechanism. 5n using predominantly higher recoil pressures! a greaterinspiratory e7ort must e e;pended to reathe to a higher LC and to controlthe high recoil pressures y checking the descent of the N# 83! 89. 5n usingpredominantly more e;piratory muscle tension! a greater e;piratory musclee7ort must e e;pended to produce higher driving pressures for speech. Acomination of these approaches would reduce oth the inspiratory ande;piratory muscle loads! spreading the work across a larger set of muscles.2urthermore! y not reathing to high LC or reathing farther elow EEL!speakers stay closer to the mid-LC range! which has een suggested to e

most eFcient for speech production.Last! in the O5SE condition! su0ects inspired a greater percent of C# eforeeach utterance than in the #O@2 and #O@2 Q %' conditions. 6his isparticularly interesting since the #O@2 Q %' condition was the only conditionin which the LC at which speech was initiated was higher. However! changesto N" have een reported as a result of no;ious stimulation. 6he presence ofno;ious visual stimulation during N" has een shown to increase oth tidalvolume and freBuency of reathing 8!"9. 6he presence of saw-tooth noiseduring N" has een shown to increase ventilation in a group of individualswith high an;iety 8!&9. 5ndividuals may have reathed more deeply duringthe O5SE condition! resulting in a greater percent of C# inspired efore an

utterance. 6here are two possile e;planations for the =nding of greater C#inspired! ut not greater LC5! in the O5SE condition. 19 5ndividuals may havelet more air out efore an utterance in the O5SE condition than in the #O@2Q %' condition. 6his possiility is supported y the fact that phonation onsettime was not signi=cantly shorter than #O@2 in the O5SE condition as it wasin the #O@2 Q %' condition. 29 Alternately! individuals may have e;piredmore after an utterance and efore inhaling for the ne;t utterance in theO5SE condition. 6his alternate hypothesis is supported y a study ofdecererate cats in which stimulation of the midrain periacBueductal gray innoise resulted in louder vocaliations and increased laryngeal adductor ande;ternal oliBue activation ut no change in diaphragm activation 8!%9.

6he data from the present study suggest that speaking in noise elicitsadditional goals and di7erent respiratory patterns than other conditions thatwere used to elicit an increase in SPL. However! the data do not support$inkworth and 4avis:s 8#$9 claims that speaking in noise does not elicit apatterned and consistent response from the respiratory system. 6here areclear patterns in the data! and the variaility present does not match thelevel reported y $inkworth and 4avis 8see Fi'. /9. 6he variaility reportedin their study may have een due to the reduction in all auditory feedackcaused y delivering the noise through headphones. 6his methodology wasaltered in the present study since the noise was delivered via free =eld.Although cereral control of the respiratory system for speech production isnot well understood! hypotheses can e made aout the areas of the rainlikely to e involved in the formation of internal targets and planning outputto the respiratory muscles ased on studies of the neural control of
http://jap.physiology.org/content/98/6/2177#ref-3http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-14http://jap.physiology.org/content/98/6/2177#ref-15http://jap.physiology.org/content/98/6/2177#ref-17http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#F6http://jap.physiology.org/content/98/6/2177#ref-3http://jap.physiology.org/content/98/6/2177#ref-8http://jap.physiology.org/content/98/6/2177#ref-14http://jap.physiology.org/content/98/6/2177#ref-15http://jap.physiology.org/content/98/6/2177#ref-17http://jap.physiology.org/content/98/6/2177#ref-20http://jap.physiology.org/content/98/6/2177#F6


16/27

respiration. #ortical and sucortical activation! in addition to rain stemactivation! would e e;pected since the respiratory system must econtrolled voluntarily to achieve the goals of the speech task 8!/9. 2ink et al.8/9 found volitional reathing to involve signi=cantly higher activation of thesupplementary motor area 8S@A9 than ventilator-controlled reathing. $hen

resistance was added to increase the work of inspiration! there was also asigni=cant increase in the activation of the primary motor area and thepremotor area 8/9. 6hese authors suggest that the increased activation withincreased resistance might have een a result of task-related changes insensory inputI 8Nef. /! p. %1'9. 6he O5SE condition in the present studyalso involved a task-dependent change to sensory input and! therefore!generation of the resulting respiratory patterns may involve some of thesame cortical areas suggested y 2ink and colleagues 8/9. 2urther listening tocontinuous noise increases cereral metaolism in the auditory areas andmay increase the metaolic demand! leading to deeper reathing in noise8!"9.

@c[ay and colleagues 8!/9 e;amined suprapontine activation during avoluntary hypernea reathing task. 6hey found increased activation of theS@A ilaterally and the right premotor area 8!/9. 6hese authors suggestedthat the increase in activity in the S@A may relate to the learned nature ofthe task 8!/9. 6his might e;plain the S@A activation in the 2ink et al. 8/9study as well since they asked su0ects to reathe more deeply than normal.@c[ay et al. 8!/9 suggest that the right premotor area activation may e dueto an increased attentional reBuirement with the voluntary hypernea task.

6he S@A and premotor areas may e involved in the #O@2 Q %' conditionsince it is not a natural task and is likely to reBuire a greater amount ofattention than a more natural cue like the O5SE condition.

Also! in the #O@2 Q %' and O5SE conditions! su0ects increased the volumeat which speech was initiated! demonstrating preplanning of respiratorymovements to achieve the internal target. 6his was not demonstrated in the&R #O@2 condition! where su0ects tended to continue to speak at lower LCsusing predominantly e;piratory muscle tension. 6he premotor area or S@A!which is known to e involved in motor planning! may have een moreactivated in the #O@2 Q %' and O5SE conditions more than in the &R #O@2condition.

6he point of this study was not to suggest that one cue to increase loudnessis etter than another from a clinical perspective. However! an understandingof how cues a7ect respiratory function will assist in choosing the est cuesfor treatment. 6he results indicate that cues to increase loudness elicitdi7erent respiratory patterns. "ecause it is not possile to test everysituation in which an individual may need to increase his/her loudness! itwould e ene=cial to treat individuals using multiple cues to ensure severalpatterns of respiratory function are supported y therapy. 2urthermore! it isimportant to realie that results from studies of respiratory kinematics cannote compared across studies without considering the cue used to elicit theincrease in loudness. Last! it is important to consider the eFciency of thepatterns elicited y these cues! from a work perspective! when planning atreatment. 6he kinematic patterns elicited in the O5SE condition appeared toe the most eFcient and reBuired the least muscular e7ort from the speaker.

6he kinematic pattern elicited y the #O@2 Q %' condition was also eFcientin that recoil pressures were used to a great e;tent! reducing the e;piratory
http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-14http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-14http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-16http://jap.physiology.org/content/98/6/2177#ref-6http://jap.physiology.org/content/98/6/2177#ref-16


17/27

muscle e7ort. 6he kinematic patterns elicited y the &R #O@2 conditionappeared to e the least eFcient of the cues used in the present study sincethe large ma0ority of respiratory driving pressure was generated y increasinge;piratory muscle tension. Speech in the &R #O@2 condition was produced atLCs where recoil pressures are lower than in the #O@2 Q %' and O5SE

conditions! resulting in a greater amount of pressure to e generated ymuscle tension.

5n summary! the data from the present study suggest that di7erent cues toincrease loudness result in di7erent internal targets and that the neuralcontrol of the respiratory system for speech is sensitive to changes in thespeaker:s internal target. 2or e;ample! in the most natural the elicitationmethod 8O5SE9! the su0ects used a comined respiratory approach< othincreased recoil pressures and increased e;piratory muscle tension.2urthermore! changes to speech rate that accompanied changes in SPL in theO5SE condition were re>ected in the respiratory kinematics. 6he control of

the respiratory system also seemed to re>ect speakers: perceptions ore;pectations. One e;ample of this was in the #O@2 Q %' condition where astrategy of using primarily higher recoil pressures was demonstrated possilyecause of su0ects: e;pectations of diFculty in reaching this loudnesstarget. Another e;ample of this was in the &R #O@2 condition where aprimarily A strategy was used! possily ecause su0ects mis0udged theamount of respiratory drive needed for this condition. 5n future studies! itwould e ene=cial to study the e7ects of these cues across a continuum. 2ore;ample! does a lower level of noise induce the same changes to respiratorypatterns and speech rate\

GRANTS

6his research was funded y ational 5nstitute on 4eafness and Other#ommunication 4isorders Grant %N'1 4#-'(,1%-'%.

Footnotes

6he costs of pulication of this article were defrayed in part y thepayment of page charges. 6he article must therefore e herey marked

ad"ertisementI in accordance with %* .S.#. Section %,1 solely to indicate

this fact.

#opyright ] &''( the American Physiological Society

REFERENCES

%. 0
http://jap.physiology.org/content/98/6/2177#xref-ref-1-1http://jap.physiology.org/content/98/6/2177#xref-ref-1-1


18/27

"oersma P)n* 1eenin2 D.Praat8.% ed.9 Amsterdam< 5nstitute of Phonetic

Sciences! &''1.

&. 0

#ampell [email protected] electromyographic study of the role of the adominal muscles in

reathing.# P$ysiol!!%< &&1D&11! %)(&.

1. 0

4raper @H L)*e4o'e* 5 )n* 16itteri*'e D.Nespiratory muscle in speech.#

Speec$ %ear &es#< %+D&,! %)().

Me*line

. 0

4romey #)n* R)7i' LO.5ntentional changes in sound pressure level and rate

perubahan mekanisme pernafasan selama bicara

Documents