An Experimental Study Of The Effects Of Fluency Feedback And Stuttering Feedback On The Subsequent Frequency Of Stuttering Utterance Duration, And Latency Of Response

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content AN EXPERIM ENTAL, STUDY O F TH E E F F E C T S OF
FLU EN CY FEED B A C K AND STUTTERING FEED B A C K
ON TH E SUBSEQUENT FREQUENCY OF STU TTERIN G ,
U TTER A N C E DURATION, AND LATENCY O F RESPONSE
by
F lo r a M ichelle W ark Je n se n
A D isse rta tio n P re s e n te d to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In P a r tia l F u lfillm e n t of the
R e q u irem en ts for the D eg ree
DOCTOR O F PHILOSOPHY
(C om m unicative D iso rd e rs)
S ep tem b er 1972
INFORMATION TO USERS
This dissertation was produced from a microfilm copy of the original docum ent.
While th e most advanced technological means to photograph and reproduce this
docum ent have been used, the quality is heavily dependent upon the quality of
the original submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from th e docum ent
photographed is "Missing Page(s)". If it was possible to obtain the
missing page(s) or section, they are spliced into the film along with
adjacent pages. This may have necessitated cutting thru an image and
duplicating adjacent pages to insure you com plete continuity.
2. When an image on the film is obliterated w ith a large round black
mark, it is an indication that the photographer suspected that th e
copy may have moved during exposure and thus cause a blurred
image. You will find a good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
p h o to g rap h ed the photographer followed a definite method in
"sectioning" th e material. It is custom ary to begin photoing at the
upper left hand corner of a large sheet and to continue photoing from
left to right in equal sections w ith a small overlap. If necessary,
sectioning is continued again - beginning below the first row and
continuing on until complete.
4. The majority of users indicate th at the textual content is of greatest
value, however, a somewhat higher quality reproduction could be
made from "photographs" if essential to th e understanding of the
dissertation. Silver prints of "photographs" may be ordered at
additional charge by writing the O rder Department, giving the catalog
number, title, author and specific pages you wish reproduced.
University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
A Xerox Education Company
J
\' 73-9314
I
I JENSEN, Flora Michelle Wark, 1936-
f AN EXPERIMENTAL STUDY OF THE EFFECTS OF
f FLUENCY FEEDBACK AND STUTTERING FEEDBACK
I ON TOE SUBSEQUENT FREQUENCY OF STUTTERING,
i UTTERANCE DURATION, AND LATENCY OF RESPONSE.
I University of Southern California, Ph.D., 1972
Speech Pathology
r
P -
I University Microfilms. A X E R O X Company. Ann Arbor, Michigan
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED.
UNIVERSITY O F SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANOELES, CALIFORNIA 8 0 0 0 7
This dissertation, written by
..............F_lora__Michen.e. W ark.Je.ngen..............
under the direction of hex.... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Gradu
ate School, in partial fulfillment of require
ments of the degree of
D O C T O R OF P H I L O S O P H Y
CfLujLxu-to***
\ J Dean
D ate.......
ATION COMMITTEE
PLEASE NOTE:
Some pages may have
in d is tin c t p rin t.
Filmed as received.
U niversity Microfilms, A Xerox Education Company
D edicated to:
P a u l, m y husband, who so unselfishly sup p o rted m y
e ffo rts to a cc o m p lish the Ph. D.
Stevie and E rik , who m ay not even have noticed but
n o n eth eless helped in th e ir special ways.
M om and Dad, who have alw ays supported me and
whom I know a re v ery proud.
ACKNOWLEDGMENTS
The author e x p re s s e s deep gratitude to h e r C h a irm a n and
I
I T e a c h e r, D r. R u s s e ll H aney, fo r his su p p o rt and guidance throughout
!
jthe planning and execution of this investigation. Special ap p rec iatio n
is extended for his w illingness to d ire c t this r e s e a r c h under the diffi-
j
cult c irc u m s ta n c e s of its being conducted on the o th e r side of the
i continent.
The author is fu rth e r indebted to the m e m b e rs of h e r d i s s e r t a
tion com m ittee fo r th e ir in te r e s t, co n stru ctiv e c ritic is m , and sp ecia l
contributions to the ideas e x p re s s e d herein.
j
G rateful acknow ledgm ent is given to the faculty, staff, and
i
!
\ students of the C om m unication S ciences L a b o ra to ry , U n iv ersity of
I
; F lo rid a , who so g e n e ro u sly sh a re d th e ir ta le n ts , e ffo rts, and equip
m ent, and to D r. P au l M o o re, C h a irm a n of the D ep artm en t of Speech,
and D r. H a rry H ollien, D ire c to r of the C om m unication S ciences L a b
o ra to ry , who g racio u sly o ffered th e ir support.
F in ally , the author e x p re s s e s h er v ery d e e p e st ap p rec iatio n to
S h e r husband P au l, not only for h is constant en co u ra g em en t, patience,
and su p p o rt, but equally, fo r his guidance throughout this re s e a r c h .
iii
T A B L E O F CO N TEN TS
jLIST OF T A B L E S ......................................................................................
i
Il i s t o f i l l u s t r a t i o n s ..................................................................
I
'C hapter
I. IN T R O D U C T IO N ...............................................................
S tatem en t of the P ro b le m
Q uestions and H ypotheses
Im p o rtan ce of the Study
II. REVIEW O F THE LITER A TU R E ...........................
F e e d b a c k Models
N o rm al Speech and DAF
S tu tterin g and DAF, M asking and Shadowing
i S tu tterin g as a P e rc e p tu a l D efect
I L atency of R esponse
| U tte ra n c e D uration
i
! HI. M ETHODS AND PR O C E D U R E S .................................
O verv iew of D esign
Subjects
M a te ria ls and A pparatus
P ro c e d u re s
IV. RESU LTS AND DISCUSSION........................................
F re q u e n c y of S tuttering
U tte ra n c e D uration
L atency of R esponse
C hapter Page
V. SUMMARY, CONCLUSIONS AND IM PLICATIONS. . . 52
S um m ary
j C onclusions
Im plications
APPENDIXES
| A. Stim ulus W o r d s .......................................................................... 62
i B. Tally of S tu tte rin g D uring P ra c tic e P e r i o d ............... 70
C. In stru ctio n s to the J u d g e s ................................................... 72
D. Judged F re q u e n c y of S tuttering and F lu en cy
for T h r e e T re a tm e n ts of A uditory F e e d b a c k . . . 74
E. D ifference S co res in U tte ra n ce D uration fo r
T hree T re a tm e n ts of A uditory F e e d b a c k ............. 78
F. D ifference S c o re s in L atency of R esponse for
j T hree T re a tm e n ts of A uditory F e e d b a c k ................. 85
R E FE R E N C E S ...................................................................................................... 92
v
LIST O F T A B L E S
T able P ag e
1. D istrib u tio n of Stim ulus W ords Among P r a c tic e ,
P re tre a tm e n t, T re a tm e n t, and P o s ttre a tm e n t
P e rio d s W ithin E ach T re a tm e n t S e s s io n .............................. 22
2. N um ber of F eed b ac k P re s e n ta tio n s fo r E ach Subject
in the Two E x p e rim e n ta l T re a tm e n t P e r i o d s .................... 31
3. S u m m ary of the A n aly sis of V arian ce of M ean
D ifferences in S tu tterin g F req u en cy for T h re e
T re a tm e n ts: No F eed b ac k , F luency F eedback,
and S tuttering F e e d b a c k .................................................................. 40
4. S u m m ary of the A nalysis of V ariance of M ean
D ifferences in U tte ra n ce D uration fo r T h ree
T re a tm e n ts: No F eed b ac k , F luency F eed b ack ,
and S tuttering F e e d b a c k .................................................................. 44
5. S u m m ary of A n aly sis of V arian ce of M ean D ifferences
in L atency of R esp o n se fo r T h re e T re a tm e n ts : No
F eed b ack , F lu en c y F e e d b a c k , and S tuttering
F e e d b a c k ................................................................................................ 48
LIST O F ILLU STR A TIO N S
F igure Page
1. A T h e o re tic a l M odel of the Speech M echanism as a
G losed-C ycle C o n tro l S y stem (F a irb a n k s )........................... 7
2. A T h e ra p e u tic a lly -O rie n te d F eed b ack Model of the
Total C om m unication S y ste m ( M y s a k ) ................................. 10
3. D isplay of E x p erim en tal R oom , C o n tro l R oom ,
and A ccom panying E quipm ent .................................................. 24
4. Illu stra tio n s of E v en ts R elating to Identification and
M e a su re m e n t of L atency of R esponse and U tte ra n c e
D u r a tio n ................................................................................................... 36
C H A PTE R I
INTRODUCTION
In re c e n t y e a rs , r e s e a r c h dealing with the behavior c alled
stu tte rin g has been c e n te re d around two th e o re tic a l m o d els; the o p eran t
i
jtheory of speech co n tro l (Skinner, 1957) and the feedback th eo ry of
I
'Speech co n tro l (F a irb a n k s, 1954). While the bulk of the r e s e a r c h has
lem erged out of the o p e ra n t m odel, th e se re s u lts have given us lim ited
u n d erstan d in g of the n atu re of stu tte rin g because of the lim ited ex p lan -
I
a to ry pow er of the o p e ra n t conditioning th eo ry . The feedback th e o ry ,
jon the o th e r hand, h as the potential of explaining stu tte rin g w ithin a
m o d el of the speaking p ro c e s s its e lf. This m odel has inspired a rath e r
lim ited am ount of r e s e a r c h , how ever.
S ta te m e n t of the P ro b le m
A ccording to the o p e ra n t th e o ry , stu tte rin g is co n sid ere d a
t
c la ss of b eh av io rs d esig n ated as o p e ra n ts and as such is subject to
c o n tro l by m anipulation of en v iro n m en tally desig n ated c o n se q u e n c e s.
jStudies that have been fo rm u la te d fro m this th e o re tic a l point of v ie w
I
have d e m o n strated th a t in sta n c e s of stu tte rin g can be co n tro lled by
1
2
irew ard o r punishm ent p a ra d ig m s. F luency d isru p tio n s of s tu tte r e r s
'have been co n sisten tly d e c re a s e d by a n u m b er of different resp o n se
contingent s tim u li (M artin and Siegel, 1966a; 1966b; and Siegel, 1970).
S tu tterin g also has been co n tro lled by an escap e p arad ig m . When
elim in atio n of delayed au d ito ry feedback (which w as co n sid ered noxious
was m ade contingent upon s tu tte rin g a new p a tte rn of speech em e rg e d
(G oldiam ond, 1965). The o p e ra n t m odel has view ed stu tte rin g in te rm s
of an o p en -cy cle sy ste m in w hich p e rfo rm a n c e is controlled by events
e x te rn a l to the sp e a k e r.
The feedback m odel, on the o th e r hand, has viewed speaking as
a c lo sed -c y cle se rv o m e c h a n is m sy ste m . The s y ste m is controlled by
its own feedback ra th e r than by feedback fro m e x te rn a l s o u rc e s . O ut
put is co n tro lled by feedback fro m the s e n so r unit (pro p rio cep tiv e,
!
k in e sth e tic , and auditory) w ith e m p h asis on aud ito ry feedback. F e e d
b a c k of the actu al output is co m p ared w ith the intended output. Any
d iffere n ce betw een intended output and actual output produces an e r r o r
sig n al. The output is then m odified by the s p e a k e r so as to red u ce the
e r r o r signal to zero .
If we m ak e the a ssu m p tio n that the intention of a sp eak er who
s tu tte rs is to sp e a k fluently, th e n stu tte rin g is s e e n as som e am ount of
" e r r o r signal" which h a s not b e e n m odified. C o n v e rse ly , fluency
•would r e p re s e n t an ap p ro x im atio n of z e ro e r r o r signal, or in o th e r
J
w o rd s, th e re would be a c lo se r m a tc h betw een intended output of
fluency and actual output.
Q uestions and H ypotheses
T he purpose of th is study, then, is to investigate the effects of
fluency feedback and stu tterin g feedback on the subsequent freq u en cy of
i
s tu tte rin g , u tteran ce du ratio n , and latency of re sp o n se . F eedback
I
Itheory and the proposed purpose of this study p ro m p t two q u estio n s.
j
T hey a r e as follow s:
j
1. If the sp e a k e r is given inform ation about his e r r o r sig n al
p erfo rm a n ce by way of rece iv in g auditory feedback of his
stu tte re d productions only (stu tterin g feedback), does he
show evidence of usin g the in fo rm atio n to ad ju st his produc
tion of subsequent sp e e c h and thereby achieve le s s e r r o r
| s ig n a l?
I
i
2. If the sam e sp e a k e r is given in fo rm atio n about his fluent
productions (fluency feedback), does this type of in fo rm a
tion influence his production of subsequent sp eech ?
T h ese questions a r e investigated by the use of the following
|
h y p o th e ses in w hich the independent v a ria b le s a re two types of auditory
feedback: (1) fluency feedback and (2) s tu tte rin g feedback. The d e p e n
d e n t v a ria b le s a r e th ree m e a s u r e s of sp e e c h p e rfo rm a n c e : (1) f r e
quency of stu tte rin g , (2) u tte ra n c e d u ratio n , and (3) latency of re s p o n s e .
| T he hypotheses being te s te d a re :
1. The m e an d e c re a s e in freq u en cy of stu tte rin g w ill differ
significantly (p < .05) am ong th re e tre a tm e n ts of aud ito ry
feedback: fluency feedback (u tteran ce is fed back im m e d i
ately if it is judged to have b een spoken fluently), s tu tte rin g
feedback (utterance i s fed b a c k im m ed iately if it is judged
T
to have been s tu tte re d ), and a control tre a tm e n t of no f e e d
back (neither judged fluent n o r judged stu ttered u tte ra n c e s
a re fed back).
a. The m ean d e c re a s e in the frequency of stu tte rin g w ill
be significantly g r e a te r (p < .01) for fluency feed b ack
than for no feedback.
b. The m ean d e c r e a s e in the frequency of stu tte rin g w ill
be significantly g r e a te r (p < .01) for stu tte rin g f e e d
back than fo r no feed b ack .
c. The m ean d e c r e a s e in the frequency of stu tte rin g w ill
be significantly g r e a te r (p < .01) for fluency feedback
than for s tu tte rin g feedback.
2. The m e a n d e c r e a s e in u tte ra n c e duration (am ount of tim e
spent producing an u tte ra n c e ) will d iffe r significantly
(p < . 05) am ong th re e tre a tm e n ts of auditory feedback:
fluency feedback, stu tte rin g feedback, and no feedback.
a. T he m e an d e c r e a s e in u tte ra n c e duration w ill be s i g
nificantly g r e a te r (p < .0 1 ) for fluency feedback than
fo r no feedback.
b. T he m ean d e c r e a s e in u tte ra n c e duration w ill be s i g
nificantly g r e a te r (p < .01) for stu tterin g feedback than
fo r no feed b ack .
c. T he m ean d e c r e a s e in u tte ra n c e duration w ill be s i g
nificantly g r e a te r (p < .0 1 ) for fluency feedback than
for stu tte rin g feedback.
3. The m e a n d e c re a s e in latency of resp o n se (am ount of tim e
fro m th e p re se n ta tio n of a stim u lu s w ord to the beginning
of the spoken re sp o n se ) will d iffer significantly (p < . 05)
am ong th re e tre a tm e n ts of auditory feedback: fluency f e e d
back, s tu tte rin g feedback, and no feedback.
a. The m ean d e c r e a s e in laten cy of resp o n se will be s i g
nificantly g r e a te r (p < .0 1 ) following fluency feedback
than following no feed b ack .
--------------------------------------------------------------------------------------------------------------------5_
b. The m ean d e c re a s e in latency of resp o n se will be s ig
nificantly g r e a te r (p < .01) follow ing stu tterin g f e e d
back than following no feedback.
I
| c. The m ean d e c re a s e in latency of resp o n se will be sig -
! nificantly g r e a te r (p < .01) following fluency feedback
j than following s tu tte rin g feedback.
!
Im portance of the Study
T his study has both th e o re tic a l and clin ica l im p o rtan ce. H o p e
fully, this line of investigation which v iew s stu tte rin g as p a r t of the
n o rm a l speaking p ro c e s s with re fe re n c e to a c lo se d -c y c le s e rv o m e c h
an ism m od el, will give us g r e a te r u n d erstan d in g of the b asic n atu re
of stu tte rin g . Is the m e c h a n ism one th a t is se lf-re g u la tin g , and if it is
then when given in fo rm atio n about its own b eh av io r, as w ith auditory
i
feedback, can it u tilize the in fo rm atio n p ro fita b ly ?
| T his study m a y also have re le v a n c e for clin ician s who provide
I
auditory feedback to clients by having th e m liste n to th e ir own sp eech
p e rfo rm a n c e by m e a n s of the tape r e c o r d e r . It would be helpful fo r
the clinician to know if having the c lie n t liste n to re c o rd in g s of his own
ispeech actu ally pro d u ces changes in h is su b seq u en t speech p e r f o r m
ance. O r, indeed, is it the feedback of one type of sp eech p e r f o r m
a n c e , i . e . , eith er the s tu tte re d u tte ra n c e s o r the fluently spoken u t t e r -
!
ances, w hich actually has an effect upon h is su b seq u en t p e rfo rm a n c e ?
C H A PTE R II
REVIEW OF THE LITER A TU R E
F eed b ack M odels
I ----------------------------
_>4, F a irb a n k s proposed a c lo s e d -c y c le se rv o m e c h a n ism
t
Jmodel to explain speaking. This w as the f i r s t m a jo r attem p t by any
au th o r to apply co n tro l th eo ry and c y b e rn e tic s (W iener, 1948, 1950) to
ithe p r o c e s s e s of speaking. A ccording to Lane and T ra n e l (1972), "The
(defining c h a r a c te r is tic of a se rv o m e c h a n ism is th a t a sam ple of the
i
joutput of the m e c h a n ism is fed back to som e point w here it co n tro ls the
I
ensuing output" (B row n and C am p b ell, 1958). F a irb a n k s p roposed that
o utput was co n tro lled by feedback of the s e n s o r unit (p ro p rio cep tiv e,
k in e sth e tic , and auditory) w ith e m p h a sis on the au d ito ry feedback ch an
nel (see F ig u re 1). Any d ifferen ce p ro d u ces an e r r o r sig n al. The o u t
put is then m odified by the sp e a k e r so as to red u ce the e r r o r sig n al to
|z e r o .
C h ase (1965a, 1965b) p ro p o sed an in fo rm atio n flow m odel for
the o rg an iz atio n and co n tro l of m o to r activ ity , in g e n e ra l, based on
s e n s o ry feedback. Of rele v an ce to sp eech p e rfo rm a n c e , C hase stated
th a t,
6
EFFECTOR UNIT
E F F E C T IV E DR IVING SIG N A L
MIXER
CONTROLLER
UNIT
OUTPUT
STORAGE MOTOR GENERATOR MODULATOR
ERROR
SIG N A L
IN P U T
SIG N A L
I_________
I_________
INPUT IOMPARA1
SENSOR 3 SENSOR 2 SENSOR 1
F E E D B A C K S IG N A L S
SENSOR UNIT
Figure 1. A th e o re tic a l model of the speech mechanism as a closed-cycle control system
(Fa i rbanks)
CHANNEL 2
8
A significant im p airm en t of sen so ry feedback inform ation
flow, w hether due to im p a irm e n t of tra n s d u c e r function, t r a n s
m iss io n , or sequential p ro c e ssin g at h ig h e r statio n s of the
n erv o u s s y ste m will re s u lt in a sig n ifican t im p a irm e n t of the
o rg an izatio n of m o to r activity. (1965a, p. 249)
H ardy (1965), guided by this m odel, developed one of his own fo r m o re
sp ecific app licatio n to language d is o rd e rs of ch ild ren . He postulated
c e n tra l n erv o u s sy ste m control of s e n so ry in fo rm atio n as a m ean s of
re la tin g aco u stic stim u li of the re c e p to r s y ste m to psychom otor
(speech) re s p o n s e s of the m otor output sy ste m . T h ese sy stem s a re
m e d ia te d via s to ra g e functions of the b ra in in re la tio n to the auditory
m ode and c e rta in asp ects of e x p erien ce. A fu rth e r a sp e c t of the m e d i
ation betw een se n s o ry input and m o to r output s y ste m s a r e the te m p o ral
and fo rm in teg ratio n functions of the b ra in . "T h ese probably s e rv e
re la tiv e to sto ra g e and re c a ll, in e r r o r detection and control, p rio r to
I
jan a p p ro p ria te m o to r output" (H ardy, 1965, p. 13).
S h e a re r (1959) viewed the hum an voice as a s e lf-re g u la tin g
sy ste m and noted that one of the m a jo r difficulties in voice re tra in in g
is the tendency of the old speaking habits to be self-m a in ta in in g which
re s u lts in the sp e a k e r again using his "habitual vo ice" once he leaves
th e clinic. He reco m m en d ed opening the c lo se d -c y c le sy ste m by v a r i
ous phonation e x e rc is e s in an e ffo rt to elim in ate the old habit p r io r to
beginning "se a rc h in g " behavior of d ifferen t voice c h a ra c te ris tic s which
w ill r e s u lt in a new vocal style. He th e o riz e d th at once the new style
9
had been in tro d u ced it could be estab lish ed by the c lo se d -c y c le s e r v o
m e c h a n ism .
M ysak (1959) p re se n te d an ela b o ra tio n of the F a irb a n k s ' fe e d
back m odel in an attem p t to explain th erap eu tic in terv en tio n of a r ti c u
la to ry d is o r d e r s . His elab o ratio n c o n sists p rim a rily of re c e p to r and
in te g ra to r units (sch em atized in F ig u re 2) w hich provide fo r "prim ing
'p ro c e d u re s ," o r sp eech s y ste m re a d ju stm e n t. " P rim in g " p re p a re s the
|
jspeaker fo r the adoption of an e x te rn a l c o n tro l signal, i . e . , the sound
iproduct of the clin ician . T his involves inform ing the s p e a k e r's sy ste m
th at it has developed a specific m is a rtic u la tio n and thus through e x t e r
n a l c o n tro l a "th erap eu tic e r r o r sig n al" has been e s ta b lish e d . M ysak
sta te s th at ". . . eventually this open cycle co n tro l (stim ulating and
guiding) w ill develop into a clo sed cycle co n tro l (in tern al fo rm atio n and
m o n ito rin g , or the 'in te rn a liz a tio n ' of the clin ician , as it w e re )"
(p. 149).
In a subsequent a rtic le , M ysak (I960) applied his m o d e l to the
p ro b le m of stu tte rin g . In th is in sta n c e , the open -cy cle p a r t of the total
sy ste m s e rv e s to introduce a "fluency e r r o r s ig n a l." An illu stra tio n of
I
this in Jo h n so n 's th e o ry (1956) would be the c r itic a l evaluation by
p a re n ts of the c h ild 's d isfluent sp eech . A ccording to M ysak, "if the
i
undeveloped s y s te m is exposed freq u en tly to this open cycle fluency
e r r o r signal, in te rn a liz a tio n m ay o c c u r, and the s y ste m m a y , c o n s e
quently, develop a closed cy cle fluency e r r o r sig n a l" (p. 191). When
RECEPTOR
UNIT
INTEGRATOR
UNIT
CONTROLLER
UNIT
EFFECTOR
UNIT
INPUT
JRECE TO R 1
L - ! _____1
RECE TO R 1
•
RECE TOR
[ r e c e
1
1 ____
t o r" J
- J
STO RAGE
cnvtDuno
UUvtKNUn
INTEGRATOR
2
•RROR
SIGNALX
EFFECTIVE DRIVING SIGNAL
— GENERATOR MODULATOR
1 '
1 i
OUTPUT
T
I
I
- V
T T T | |— i r - a 1 1 — 1 1 — u j
I I I JSENSOR* [ s e n s o r ! . SE N SO R jljSE N SO R kH
i i 1 I i M o r I i f^l i 1
I I L* J L* J ui J I — 1
I I L
I L
1 ___
T
j
FEEDBACK SIG N A L S
SENSOR UNIT
Figure 2. A th era p eu tica lly -o rie n te d feedback model of the total communication syste'i
(Mysak)
CHANNEL 2
CHANNEL I
11
the fluency e r r o r signal has been m ade to e x is t on an o p en -cy c le b asis,
the s y ste m re a d ju stm e n t is achieved by m anipulation of the e x te rn a l
s o u rc e s of the e r r o r signal. On the o th e r hand, if the open -cy cle
fluency e r r o r sig n al has becom e in te rn a liz e d , i . e . , h as becom e p art
of the c lo se d -c y c le sy ste m , then d ire c t re a d ju stm e n t of the s p e a k e r's
jsystem is re q u ire d . Two m ethods a r e then re c o m m e n d e d : (1) w eak
ening of the o p en -cy c le fluency e r r o r signal and (2) am plifying the
j
jclosed-cycle fluency zero e r r o r signal. The fo rm e r is to be achieved
[by d ire c t m anipulation of en v iro n m en ta l con seq u en ces and the la tte r by
stren g th en in g ze ro e r r o r sig n al productions, fo r ex am p le, by em p h a-
jsizing the s p e a k e r 's fluent p roductions.
N o rm al Speech and DAF
T he feedback sy ste m of speech c o n tro l contains m any co m p o
nents in a co m p licated a rra n g e m e n t and the s y s te m can read ily becom e
d is o rd e re d . D elayed A uditory F eed b ack (DAF) is one such m ethod.
L ee (1951) re p o rte d that when reading under D A F, n o rm a l s p e a k e rs
e x p e rie n c e d a breakdow n in sp eech production w hich on the su rfa c e
sounded like s tu tte rin g . This phenom enon w as r e f e r r e d to by Lee as
["artificial s tu tte r ."
S e v e ra l sp e e c h p a ra m e te rs h av e been found to be affected by
D A F . N eelley (1961) and F airb an k s and G uttm an (1958) found in c re a s e s
in o m iss io n s , sub stitu tio n s and additions of phonem es under DAF as
12"
co m p ared to n o rm a l auditory feedback. C hanges in v o cal pitch w ere
found by F a irb a n k s (1955) and S o d erb erg (1959). Black (1950) found
th a t s p e a k e r s ' v o ices becam e m o re in te n se . He also found prosody
changes with p ro g re s s iv e in c re a s e s in syllable prolongations a s delay
in c re a s e d fro m 0 to .18 seco n d s.
N eelley (1961), in an e ffo rt to d isc o v e r if delayed auditory fe e d
b a c k m ight be an im p o rta n t fa c to r causing the re p e titio n s and p ro lo n g a
tio n s in the speech of people identified as s tu tte r e r s , p e rfo rm e d a stud)
jcom paring s tu tte r e r s speaking u n d er n o rm a l feedback and n o rm a l
I
i
jsp eak e rs speaking under delayed auditory feedback. He found th a t
iwhile th e re w ere so m e p a ra lle ls , with re g a rd to the type of a rtic u la tio n
e r r o r s that w ere m a d e , the groups w ere actually quite d iffe re n t.
S tu tterin g and D A F, M asking, and Shadowing
When G oldiam ond (1965) had s tu tte r e r s re a d u n d e r a condition
■in w hich s tu tte rin g would actu ally elim in ate DAF (which he viewed as
an a v e rsiv e stim u lu s), he found th at ra th e r than in c re a s in g the f r e -
iquency of th e ir stu tte rin g , they changed th e ir style of sp eech . The new
sp e e c h p a tte rn w as c h a ra c te riz e d by prolonged vowel production as if
;the sp e a k e r w as try in g to slow down his ra te of sp eech in o rd e r to
r e - e s ta b lis h the c o r r e c t re la tio n sh ip betw een sp eech output and fe e d
b ack . The new sp e e c h p attern w as s tu tte r - f r e e ; it did n o t include the
rep e titio n s and s tru g g le that w e re c h a r a c te r is tic of the stu tte rin g .
13
G oldiam ond hypothesized th a t under DAF a new p a tte rn of speaking w as
su b stitu ted fo r the p a tte rn which had contained stu tterin g . As long as
stu tte rin g w as not d ifferen tially re in fo rc e d it would not becom e p a rt of
the new p a tte rn .
C u rle e and P e rk in s (1969) adapted G oldiam ond's technique fo r
the tre a tm e n t of stu tte rin g during c o n v e rsa tio n r a th e r than during o ra l
read in g . T h e ir subjects red u ced the freq u en cy of stu tterin g to zero
u n d er DAF w ith in stru c tio n s to sp eak slow ly, and then, through a
s e r ie s of "shaping" p ro c e d u re s they red u ced the am ount of delay until
the s tu tte r e r s w ere speaking under n o rm a l feedback without delay and
s till without stu tte rin g .
O ther p ro c e d u re s w hich have a lte re d n o rm a l auditory feedback
'have also produced changes in the sp eech output of s tu tte r e r s . When
I
I
j speaking w ith 50 dB m a sk in g , stu tte rin g is elim in ated and n o rm a l
i
^fluent sp eech b ecom es p red o m in an t (C h e rry and S a y e rs, 1956; M a ra is t
and Hutton, 1957; and Shane, 1955). B loodstein (1950) found that when
speaking w ith a w h isp er (a p ro c e d u re w hich e lim in a te s the vocal c h a r
a c t e r i s t i c s of auditory feedback) s tu tte r e r s seld o m s tu tte r. C h e rry
and S ay ers (1956) also found that s tu tte r e r s w ere able to sp eak n o r
m a lly , without stu tte rin g , when shadow read in g o r reading s im u lta n
eously with an o th er.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - r r
Stuttering as a P e rc e p tu a l D efect
The evidence cited above in d icates c le a rly th at output can
re a d ily be a lte re d by m anipulating s e n s o ry (auditory) feedback. T his
evidence su g g ests that stu tterin g m ay actu ally be, in p art, a p e rc e p t
ual p ro b lem . S o d erb erg (1968) o ffers th re e p o ssib le explanations fo r
th e se fin d in g s:
| 1. The phase d iffere n tial betw een a ir-c o n d u ctio n and bone-
: tissu e feedback is g re a te s t n e a r the fundam ental frequency
l of the s p e a k e r's voice. Under DAF, s tu tte r e r s tend to
' in c re a s e th e ir vocal pitch and thus achieve a c o rre c tio n of
| the feedback d iscrep a n cy .
2. The b o n e -tissu e feedback is out of phase with the air
conduction feedback and the DAF r e s to r e s a c o r r e c t r e l a
tionship betw een the two c h a n n e ls .
3. Under DAF, s tu tte r e r s prolong th e ir syllable production
and th e re b y reduce th e ir ra te of sp eech . This reduced
ra te m ay be m o re a p p ro p ria te fo r the s p e a k e r 's auditory
m o n ito rin g sy ste m w hich m ay have in h e re n t lim its w hich
a re d iffe re n t fro m n o n - s tu tte r e r s .
It is a p p ro p ria te at this point to m a k e a su m m a ry s ta te m e n t
con cern in g DAF, stu tterin g and m a sk in g te c h n iq u e s. F i r s t , DAF
in tro d u c e s a d isto rtio n into a s y s te m w hich for so m e re a s o n is alre ad y
functioning im p ro p e rly . Secondly, under DAF sp eech , the sp eak er is
— 15
expected to m ake evaluations and adjustm ents of his ow n speaking
m e ch an ism while in the p ro c e ss of speaking under a d is to rte d situ atio n .
And, finally, while DAF does re s u lt in s tu tte r - f r e e speech, th e speech
behavior is not n o rm a l, even after shaping p ro c e d u re s . That is , th e re
Is s till so m e evident am ount of syllable prolongation which is n o t ty p i
c a l of n o rm a l sp eech . The sp eak er has actually le a rn e d a new , s tu tt e r -
fr e e style of speaking. On the other hand, w hile speaking u n d er m a sk -
i
ling and shadow ing, the speech pattern w hich e m e rg e s is the " n o rm a l"
b r "fluent" speaking style of the sp e a k e r. It is not a new style of
speaking w hich e m e rg e s , but ra th e r, it is the fre q u e n c y of the sp e a k
e r 's own "fluent" style which in c re a s e s , while the frequency o f his
I
i"stu tterin g " d e c r e a s e s .
s
L atency of R esponse
Van R ip e r (1954, p. 381) has d e s c rib e d sy m p to m s of p o stp o n e
m e n t in w hich the sp e a k e r w ill often postpone saying a word e ith e r
because he fe a r s he cannot say it, or he feels th a t if he waits fo r a
I
b rie f m o m en t, he w ill be able to say it w ithout any difficulty. Love and
J e ffre s s (1971) found th a t s tu tte r e r s tend to have m o re freq u en t b rief
p au ses than do n o rm a ls during reading. Although stu tterin g is typi
cally identified by the re p etitio n s and pro lo n g atio n s, it is quite ap p aren :
th at the s ile n t in te rv a l o r period of sile n ce p r io r to speaking c a n be an
im p o rta n t p a r t of the stu tte rin g b ehavior.
------------------------------------------------------------------------------------------------------------------------re*
U tte ra n ce D uration
Je n se n (1966) em ployed both latency of resp o n se and u tte ra n c e
duration as m e a s u r e s of h esitatio n in the sp eech of young ch ild ren .
Since speech o c c u rs in tim e and s tu tte rin g r e s u lts in a lte ra tio n s of the
n o rm a l sp eech p a tte rn in tim e , m e a s u re m e n t of u tte ra n c e duratio n m ay
be c o n s id e re d a se n sitiv e in d icato r of sp eech change.
C H A PT E R III
I
METHOD AND PROCEDURES
j
O verview of D esign
This e x p erim en tal study w as designed to investigate the effects
|
|of two types of auditory feedback (fluency feedback and stu tterin g feed -
!
back) upon th re e m e a s u re s of speech behavior (frequency of stu tterin g ,
I
i
[utterance d u ratio n , and latency of re sp o n s e ). T h ese m e a su re s w ere
[derived in re la tio n to a se rv o m e c h a n ism m odel of the speech p ro c e s s .
The p re s e n ta tio n of the two types of feedback constituted the indepen
dent v a ria b le s and the th re e speech m e a s u re s w e re the dependent v a r i
ab les.
The study em ployed a group of nine adult s tu tte r e r s who se rv e d
[as th e ir own co n tro ls for two e x p e rim e n ta l tre a tm e n ts . The two e x p e r i
m en tal tre a tm e n ts w ere stu tterin g feedback and fluency feedback. A
th ird tre a tm e n t w as the control tre a tm e n t for w hich th e re w as no a u d i
to ry feedback p resen ta tio n . The effects of the tre a tm e n ts w ere c a lc u
la te d by co m p arin g p re tre a tm e n t w ith p o sttre a tm e n t re s p o n s e s . S tut-
|
te rin g feedback involved the re p la y of only those re s p o n se s which w e re
judged by the e x p e rim e n te r to be s tu tte re d . F lu en c y feedback
17
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - r e
constituted the rep la y of only those re s p o n s e s w hich the e x p e rim e n te r
judged to be flu en t.
| F re q u e n c y of stu tte rin g w as calcu lated fro m judgm ents m ade by
th ree lis te n e r s . L atency of resp o n se and u tte ra n c e duratio n m e a s u re s
w ere obtained fro m graphic level r e c o r d e r a n a ly se s, illu s tra te d la te r
>in this ch a p te r. L atency of resp o n se w as the tim e in terv al betw een the
p re se n ta tio n of the stim u lu s word and the beginning of the re s p o n s e .
i
[Utterance d u ratio n was the am ount of tim e th at p assed w hile the subject
Kvas producing h is total u tte ra n c e . This d u ratio n included any r e p e t i
tions and prolongations th a t w ere p a rt of the s tu tte re d resp o n se but
jwhich w e re included in h is total re sp o n se .
|
Subjects
The su b jects w ere nine ad u lts, seven m a le s and two fem a les
betw een 19 and 36 y e a rs of age. It was d e te rm in e d that the sm a ll n of
j
|9 would satisfy the re q u ire m e n ts of the a n a ly sis of v aria n ce assu m in g
that one would be able to get sufficient d a ta fro m eac h su b ject. Since
'the n a tu re of the study w as such th a t a la rg e am o u n t of d ata m ight be
obtained, this a lte rn a tiv e was taken ra th e r than in c re a s in g the num ber
of su b je c ts. Inspection of the following sectio n w ill re v e a l that a sub-
I
sta n tia l am ount of data w as to be obtained fro m e a c h subject; i . e . ,
180 re s p o n s e s . Thus, although the n w as s m a ll, the am ount of d ata
obtained fro m e a c h su b ject was g re a t enough so th a t the re q u ire m e n t
w
o f n o rm a l d istrib u tio n was m e t.
All the su b jects identified th e m s e lv e s as s tu tte r e r s by re s p o n d
ing to new spaper a d v e rtis e m e n ts w hich so licited subjects for the study.
In addition, all had p assed a sc re e n in g te s t by stu tterin g on at le a s t
th re e stim u li out of th irty (as judged by the e x p e rim e n te r) during an
in itia l p ra c tic e p erio d . An additional six individuals who identified
th e m se lv e s as s tu tte r e r s applied for the study but did not m e e t the
i
C rite rio n of s tu tte rin g on th re e out of th irty stim u lu s ite m s . Seven of
Jthe nine su b jects w e re c le a rly beyond the sc re e n in g c rite rio n for
^acceptance as a su b je c t in the study. T hese seven a r e v e rifie d by
judgm ents of stu tte rin g obtained d u rin g the in itia l p ra c tic e p erio d s (see
Appendix B). The o th e r two b arely m e t the sc re e n in g c rite rio n because
th e ir judged fre q u e n c y of stu tte rin g w as so low . Since th e ir inclusion
m ay be questioned, an exam ination of the p r e tre a tm e n t judgm ents of
stu tte rin g , p re se n te d in Appendix D, in d icate s that in each c a se , they
>were judged to have stu tte re d at le a s t th re e tim e s during one o r m o re
J
jof the th re e p re tre a tm e n t p erio d s.
M a te ria ls and A p p aratu s
S tim u li
F o u r h u n d red and ten w ords w e re se le c te d as s tim u li. T hese
w o rd s w ere sub seq u en tly divided into four lis ts w hich w ere used fo r
p ra c tic e , p re tre a tm e n t, tre a tm e n t and p o s ttre a tm e n t p erio d s within
------------------------- ----------------------------------------------------------------------------------- —------ 27 J-
each e x p e rim e n ta l sessio n .
It has been shown by Brown (1945) and S o d erb erg (1966) th a t
stu tte rin g and fluency often v a ry as a function of th re e phonetic v a r i
ables (word length, initial sound, and freq u en cy of o c c u rre n c e of the
w ord). T h e re fo re , it was decided th a t th e se phonetic fa c to rs should be
C ontrolled. To co n tro l for w ord length, one half of the w ords w ere one
b r two sy lla b les in length. A ccording to Z ipf (1935), s h o rte r w ords
I
jalso tend to be the m o st freq u en tly o c c u rrin g w o rd s. T h ere fo re , it
jbecame ap p are n t th a t these two v a ria b le s , length and frequency of
I
o c c u rre n c e , could be controlled to g e th e r. The s h o rt stim ulus w ords
|(one to two sy llab les) w ere taken fro m T h o rn d ik e -L o rg e 's (1944) lis t
I
jof the 500 m o st freq u en tly o c c u rrin g w o rd s. The long w ords (four to
|
six sy lla b le s) w e re taken fro m a li s t of w o rd s w hich occur le s s f r e
quently than the s h o rt w ords; they w e re se le c te d fro m the T h o rn d ik e-
L orge (1944) lis t of w ords o c c u rrin g a t le a s t once p er 1,000,000 w ords
In o r d e r to co n tro l the effect th at in itia l sound h as on fluency and s tu t
te rin g , all the w ords w ere selec ted fro m the two T h o rn d ik e-L o rg e wore
lists by use of a T able of R andom N u m b ers (E d w ard s, 1965). This
p ro c e d u re a s s u re d th at th e re would be no s y ste m a tic effect by re p e ti-
!
jtion of in itia l sounds.
i
The 410 stim u lu s w ords (205 o n e- o r tw o -sy llab le w ords and
205 f o u r - to six -sy lla b le w ords) w ere then a ssig n e d on a ro tatio n al
b a s is to the four li s t s . One list, containing 80 w o rd s, w as used fo r
----------— --------------------------------------------------------------zr
p ra c tic e . The o th e r th re e lists containing 110 w o rd s each w e re used
for the two e x p e rim e n ta l tre a tm e n ts and the co n tro l tre a tm e n t. F ro m
each li s t of 110 w o rd s, 30 (15 long and 15 sh o rt) w e re selected by use
of the table of ran d o m n u m b e rs to function as the p re tre a tm e n t te s t
w o rd s. T hese sam e w ords also functioned as the p o s ttre a tm e n t te s t
w o rd s. The re m a in in g 80 w ords functioned as tre a tm e n t stim u lu s
(words. The w o rd s within each p re tre a tm e n t and each tre a tm e n t list
I
I
(were then o rd e re d by use of a table of ra n d o m n u m b e rs (see T able 1).
|Thus, four w ord lis ts , e sse n tia lly equivalent w ith re g a rd to the pho-
inetic fa c to rs of length, frequency of o c c u rre n c e and in itial sound, w ere
developed (see Appendix A). As a re s u lt, w h atev e r e x p erim en tal
effects m ight be obtained would not be confounded by the phonetic c h a r
a c te r is tic s of the w ords being spoken.
A 2" x 2" slid e w as then p re p a re d for eac h w ord and a duplicate
slide m ad e for eac h p re tre a tm e n t and p o s ttre a tm e n t w ord. The slides
w e re m ad e by p rin tin g eac h stim u lu s w ord onto fro s te d a c e ta te . An
(electric IBM MTSC C o m p o ser, w as used for th is p ro ced u re to a s s u re
|
h ig h e st co n tro l of c la rity and ease of re a d a b ility . The ace tate w as then
m o u n ted into c a rd b o a rd slide h o ld e rs. T he slid e s w ere placed into
c a ro u s e l-ty p e p ro je c to r tr a y s .
E quipm ent
The study w as conducted in a tw o -ro o m e x p e rim e n ta l suite of
the C o m m u n icatio n S ciences L ab o ra to ry a t the U n iv ersity of F lo rid a .
n
T A B L E 1
DISTRIBUTION OF STIMULUS WORDS AMONG PRA CTICE,
PR E T R E A T M E N T , TR EA TM EN T, AND PO STTREA TM EN T
PERIODS WITHIN EACH TREATM ENT SESSION
1
_ a
P r a c tic e list 40 long 40 sh o rt
i
P r e tr e a tm e n t lis t'5 15 long 15 s h o rt
Session I T re a tm e n t lis t 40 long 40 sh o rt
J
b
P o s ttre a tm e n t lis t 15 long 15 s h o rt
1
i
P re tr e a tm e n t lis t 15 long 15 s h o rt
1
S ession II
c
T re a tm e n t lis t 40 long 40 sh o rt
1
i
P o s ttre a tm e n t lis t 15 long 15 sh o rt
P re tr e a tm e n t li s t 15 long 15 sh o rt
S ession III T re a tm e n t list 40 long 40 sh o rt
P o s ttre a tm e n t lis t 15 long 15 s h o rt
The long and s h o rt stim u li within eac h lis t a re a rra n g e d in
ran d o m ized o r d e r by use of a table of ran d o m n u m b e rs.
P o s ttre a tm e n t and p re tre a tm e n t s tim u li a r e the sam e fo r
ea c h se s sio n .
c
T re a tm e n t stim u li w e re p re se n te d u n til 40 in sta n c e s of fe e d
b ack had o c c u rre d o r until 80 stim u li had b een p re se n te d . The num ber
bf stim u li p re se n te d w ere v a ria b le fo r each of the two ex p e rim e n ta l
s e s s i o n s .
------------------------------------------------------------------------------------------------------------------ZT
The two ro o m s w ere s e p a ra te d by a one-w ay window. C om m unication
betw een the two ro o m s was m aintained through an in te rc o m s y ste m .
The su b ject w as in the e x p erim en tal room w hich w as equipped w ith two
E le c tro -V o ic e 664 m icro p h o n es, two ta b les, a c h a ir, and A coustic
R eflex-3 lo u d sp eak e r and a black fe lt s c re e n .
I
j The e x p e rim e n te r w as in the co n tro l ro o m w hich contained two
Kodak C a ro u se l-8 0 0 slide p ro je c to rs , an A m pex P R -10 two channel
i
tape r e c o r d e r , and a 1 kHz pulse g e n e ra to r. The four pieces of equip-
iment in the co n tro l ro o m w ere controlled by a single re m o te co n tro l
box. The m e c h a n ic a l o p eratio n s of the e x p e rim e n t w ere con cen trated
Into a single piece of equipm ent so th at the e x p e rim e n t could be con-
|
ducted by a single e x p e rim e n te r w ithout the aid of an a s s is ta n t. A ddi
tionally, th is p ro c e d u re allow ed for in c re a s e d p re c isio n and re liab ility .
Also in the co n tro l ro o m , th e re w as a tw o-channel Sony (TC-540)
ste re o type tape r e c o r d e r and an e le c tro n ic c o u n ter. The tw o -ro o m
suite and accom panying equipm ent a re p o rtra y e d in F ig u re 3.
j P e rm a n e n t re c o rd in g s of all the s e s sio n s w e re m ade on the tw o-
channel Sony tape r e c o r d e r . The su b je c t's sp eech w as re c o rd e d on
ch an n el 1. A 1 kHz tone b u rs t w as re c o rd e d on channel 2. This tone
(burst w as re c o rd e d sim u ltan eo u sly with the p re se n ta tio n of the stim ulus
i
w ord to the subject. Thus, latency of re sp o n se could be d eterm in e d by
m e a su rin g the in te rv a l betw een the p re se n ta tio n of the stim ulus and the
su b je c t's initiation of his re sp o n se . Subsequently, a m a s te r tape was
SCREEN
EXPERIMENTAL ROOM
MICROPHONES
SUBJECT 1 WAY MIRROR
TONE
g f f i
c h .l
SONY
c h .2
REMOTE CONTROL
BOX
PR -10
« — COUNTER o u t
1kHz
ON
AM P
PULSE
CONTROL ROOM EXPERIMENTER
Figure 3. Display of experimental room, control room, and accompany!ng equipment
m o
25
developed fro m th e se p e rm a n e n t re c o rd in g s. It w as fro m the m a s te r
tape that p e rc e p tu a l judgm ents of frequency of stu tte rin g and the p h y s
ical m e a s u re m e n ts of u tte ra n c e du ratio n and latency of resp o n se w ere
m a d e .
F eedback w as p re se n te d by re c o rd in g and playing back the s u b
je c t's u tte ra n c e s on a s ix -se c o n d or eig h t-seco n d tape loop on the
lPR-10 tape r e c o r d e r . D uring a pilot study, two p ro b lem s w ere e n -
|
jcountered in the u se of a continuous cycle tape loop which re c o rd e d and
jpresented feedback of the s p e a k e r's u tte ra n c e s : (1) if the s p e a k e r's
i
i
u tte ra n c e took longer than the length of the loop, then the in itial portion
of the loop w as e r a s e d and the additional u tte ra n c e w as re c o rd e d as the
loop continued to cy cle; consequently, the feedback w as d is to rte d since
it contained m ixed u tte ra n c e s ; (2) in those in sta n c e s when the su b ject
was to re c e iv e feedback, th e re was a six -se c o n d in te rv a l betw een
u tteran ce p ro d u ctio n s, w h e re a s th e re w as no tim e in te rv a l in those
in sta n ces w hen th e re w as no feedback. It w as felt th at this tim e i n t e r
val v aria tio n could affect the obtained r e s u lts .
To solve the f i r s t p ro b le m it w as decided to co n tro l the tape
ioop with a tone c o n tro l re la y sy ste m . A 1 kHz tone was placed on the
jchannel 1 tr a c k of the tape. W henever the tone p a sse d the playback
head of the A m pex P R -1 0 , the tape loop cam e to an autom atic stop.
T hus, the in itia l p o rtio n of the tape w as not e r a s e d . The su b ject n ev er
h e a rd the tone b e c a u se his sp eech , and what he h e a rd played back, w as
----------------------------------------------------------------------------------------------------------------------- Z F
on channel 2, w hile the tone w as on channel 1. The second p ro b lem
was solved by a d ecisio n to re c y c le the tape loop a fte r each u tte ra n c e
w hether o r not the su b ject w as to h e a r what he had re c o rd e d . If the
su b ject w as to re c e iv e feedback as the tape re c y c le d , then IS p re s s e d
a sw itch w hich tu rn ed on the lo u d sp eak e r in th e e x p e rim e n ta l room .
The lo u d sp eak e r w as not tu rn ed on following those u tte ra n c e s for which
jthere w as to be no feedback.
j An e le c tro n ic co u n ter, which w as activated each tim e the loud-
I
ispeaker w as tu rn ed on, kept ta lly of the n u m b er of tim e s th a t the su b -
i
pect had re c e iv e d feedback d u rin g each tre a tm e n t p erio d . Follow ing
1
the pilot study it w as decided th at each tre a tm e n t p erio d should re n d e r
40 in s ta n c e s of feedback p re se n ta tio n in o rd e r to m a x im iz e the effect
of the independent v a ria b le . In o rd e r to m in im iz e su b ject fatigue, no
tre a tm e n t period contained m o re than 80 stim u lu s p re s e n ta tio n s.
I
i
j Word s tim u li w e re p re se n te d v ia the two slide p ro je c to rs . The
slides w ere p ro je c te d through the window onto a black felt s c re e n in
front of the su b ject. One p ro je c to r p re se n te d the p r e tre a tm e n t and
p o s ttre a tm e n t w o rd s, the o th e r p re se n te d the p ra c tic e and tre a tm e n t
stim u li. Since each c a ro u s e l tra y held only 80 slid e s, it w as n e c e ss a ry
!to use two p ro je c to rs to effect a sm ooth tra n s itio n fro m one tre a tm e n t
period to the next.
27
P ro c e d u re s
E ach su b ject p articip a ted in the study on th re e se p a ra te days
within the span of a w eek. The subjects ex p erien ced one of the t r e a t
m ents (no feedback, fluency feedback, o r stu tte rin g feedback) on each
day. E ach tre a tm e n t se ssio n w as divided into four p e rio d s : p ra c tic e ,
p re tre a tm e n t, tre a tm e n t, and p o sttre a tm e n t. The su b je c ts w ere
I
I
assig n ed to seq u en ces of tre a tm e n ts by L atin Square design for two
jreasons: (1) so th a t possible o rd e r effects could be analyzed during
the s ta tis tic a l a n a ly sis of the data, and (2) to co n tro l for possible o rd e r
effects. U nfortunately, two of the subjects (each fro m a d iffere n t Latin
Square) had to r e r u n one of th e ir s e s sio n s. As a consequence, the
r e s u lts of only th re e su b jects would have been av ailab le to analyze
o rd e r e ffects. B ecause only th re e out of the p o ssib le six o r d e r s would
have been re p re s e n te d and because those th ree would eac h be r e p r e -
I
sen ted by one su b ject, it w as decided th at o rd e r effects could not be
!
an aly zed fro m the d ata. F u r th e r , it w as felt th a t the m o s t im p o rtan t
in fo rm a tio n would have been obtained fro m analyzing an in teractio n
betw een o r d e r and type of aud ito ry feedback and it had becom e im p o s
sib le to obtain such an in te ra c tio n with only th re e s u b je c ts. T hus, the
O riginal plan to analyze o rd e r effects through s ta tis tic a l an aly sis w as
e lim in ated . H ow ever, even so, by assigning su b jects to tre a tm e n ts by
L atin Square d esign, the g en era l p u rp o s e --c o n tro l fo r o rd e r e ffe c ts - -
w as sa tisfie d .
! E ach of the su b jects e n te re d the e x p e rim e n ta l room and w as !
Seated at the table facing the s c re e n . He w as told that w ords w ere j
going to be p re se n te d on the s c re e n in fro n t of him and th at his ta sk was!
to say each w ord im m e d ia te ly . He was also in stru c te d th at o c c a s io n
ally he would h e a r his own voice through th e loudspeaker and that he
w as sim ply to liste n to the rep la y w henever it o c c u rre d . He was i n
stru c te d a lte rn a tiv e ly th a t if th e re w as no replay, he was to sit and i
w ait for the p re se n ta tio n of the next w ord. He w as told th a t the s e s sio n
would take about an hour and th at the f ir s t p a rt would be p ra c tic e so
th a t he m ight get u sed to the p ro ced u re and be able to ask any questions^
w hich m ay a r is e .
P ra c tic e j
t I
i
The p ra c tic e p erio d la ste d fo r the p re se n ta tio n of 30 stim u lu s j
j
slid es and four p re se n ta tio n s of feedback. The p ra c tic e period la ste d
fo r about 10 m in u tes and allow ed the su b je c t to becom e fa m ilia r w ith
his ta sk and w ith the e x p erien ce of h e a rin g the feedback of his own
u tte ra n c e s o ver the lo u d sp eak e r. In addition, it allow ed adaptation to
I
the situation to take p la ce. It has been d e m o n stra te d that upon repeated
! i
read in g s of the sam e p a ssa g e , the freq u en cy of stu tte rin g d e c r e a s e s .
The o rig in al high rate of stu tte rin g has b een shown to be m o s t lik ely
due to the novelty of the speaking situ atio n . Cohen (1955) has shown
th at adaptation to the situ atio n tak es place within 10 m in u te s. It w as
[necessary to hav e adaptation to the situation co m p leted p rio r to b eg in - [
i :
jning the p re tre a tm e n t p erio d , in o rd e r to provide the p o ssib ility of
j
concluding w h eth er any changes w ere due to the adaptation effect or to j
' 1
ithe tre a tm e n t e ffe c ts. A tally of the freq u en cy of s tu tte rin g during th e j
screen in g p erio d is liste d in Appendix B. It m a y be noted fro m th ese
tabulations th at the freq u en cy of stu tterin g w as alw ays stable p rio r to
ibeginning the p r e tr e a tm e n t p erio d s. F u r th e r , it m ay be noted that fo r ,
each subject, the frequency of s tu tte rin g was fa irly stab le f r o m day to
I
day.
The f i r s t p ra c tic e p erio d se rv e d as s c re e n in g fo r su b ject s e le c t
;tion. During th e sam e p erio d , IE calculated the a v e ra g e length of the
s p e a k e r's s tu tte re d r e s p o n s e s . At the end of the sc re e n in g , and a s -
I
jsuming the su b je c t w as se le c te d for th e study, a d ecisio n w as m ade as
I
[to w hether the su b ject would be assig n ed a six -se c o n d o r eig h t-seco n d |
! I
[tape loop. The longer loops (eight seconds) w ere u s e d for su b jects !
i
[whose av erag e s tu tte re d re s p o n s e w as g re a te r than s ix seco n d s. The
; i
long tap e loop, if selec ted , w a s slipped onto the m ach in e and an addi-
I
jtional 20 p ra c tic e w o rd s w ere p re s e n te d . (Two su b je c ts, B. C . and
;
j V . S ., used the longer loops; the o th e r subjects used the six -seco n d
i
loops). The p ra c tic e p erio d w as extended b ecau se of the addition of the
two seconds to the tape loop an d the po ssib le need fo r fu rth e r a d a p ta
tion. Once a ta p e length had b een s e le c te d fo r a su b ject, th a t sam e
loop length w as used th ro u g h o u t the study.
IP re tre a tm e n t
The second period fo r ea c h se ssio n was the p re tre a tm e n t period
It co n sisted of the p resen ta tio n of 30 p re tre a tm e n t stim u lu s w o rd s.
;The subject spoke each w ord as soon as it w as p ro jec ted on the s c re e n
in fro n t of h im . T h ere w ere no in stan ces of feedback d u rin g the p r e -
tre a tm e n t p erio d . The sp eak er sa t in silence during the s ix - o r e ig h t-
second p erio d re q u ire d to re c y c le the tape loop. The p re tre a tm e n t
period w as included for the p u rp o se of providing a s e r ie s of u tte ra n c e s
fro m w hich c o m p a riso n s could be m ade after the su b ject ex p erienced
the tre a tm e n t p erio d .
T re a tm e n t
The tre a tm e n t period la ste d for 40 feedback p re se n ta tio n s or
juntil th e re had been 80 stim u lu s p rese n ta tio n s. It w as intended that
jthe tre a tm e n t period c o n s is t of 40 feedback p re se n ta tio n s w hether the
(treatm ent w as fluency feedback o r stu tterin g feedback. In e ith e r of
(those in sta n c e s how ever, if 40 feedback p re se n ta tio n s had not been
[accom plished in the p re se n ta tio n of 80 w ords, th at tre a tm e n t w as t e r
m in a te d . As indicated p rev io u sly , this p ro ced u re w as adopted in o r d e r
[to co n tro l the nu m b er of feedback p resen ta tio n s and yet to p re v e n t s u b
je c t fatigue. T able Z show s the num ber of feedback p re se n ta tio n s th a t
each su b ject re c e iv e d d u rin g ea c h tre a tm e n t p erio d .
TA BLE 2
NUMBER OF FEED B A C K PRESENTATIONS FOR EACH
SU B JECT IN T H E TWO EX PERIM EN TA L
TR EA TM EN T PERIODS
Subject
F luency
Feedback®
S tuttering
F eedbacka
LA 40 40
BC 38 35
LD 40 9
RN 40 0
RR 40 29
VS 40 20
DS 40 23
JW 40 10
RBW 40 26
' cl
F ig u re s under F lu en cy F eedback and S tuttering F eed b ack
!should be in te rp re te d to m ean th a t the n u m b er of feedback p r e s e n ta
tio n s accom plished in e a c h e x p e rim e n ta l period for each tre a tm e n t a re
'displayed h o rizo n tally in re la tio n to the s u b je c t’s in itia ls.
l l r
I :
: i
j It was d u rin g the tre a tm e n t p erio d that the su b jects w ere given
i I
jone of the th re e feedback tre a tm e n ts w hich co n siste d of the two experi-1
] i
m e n ta l independent v a ria b le s , fluency feedback and stu tte rin g feedback j
and a co n tro l tre a tm e n t of no feedback. In feren ces co n cern in g the !
effects of th ese v ario u s tre a tm e n ts w e re m ade by c o m p a rin g the r e s
ponses of the su b jects during the p erio d s p rio r to and a fte r receiving
the feedback tre a tm e n t. i
P o s ttre a tm e n t
The p o sttre a tm e n t period w as an ex ac t re p lic a of the p r e tr e a t
m e n t period; the sa m e w o rd s w e re p re se n te d . T h ere w e re no feedback!
p re se n ta tio n s. The p o s ttre a tm e n t p erio d w as included in o r d e r to
co m p are a su b je c t's re s p o n s e s a fte r re c e iv in g one of the feedback j
|
(treatm ents with his re s p o n se s b efo re rece iv in g it. |
; i
i j
| t
R esp o n ses j
The p e rm a n e n t re c o rd in g s of a ll su b jects during the p r e tr e a t
m e n t and p o s ttre a tm e n t p erio d s w e re analyzed fo r one p e rc e p tu a l
(m easure (frequency of stu tte rin g ) and two p h y sical m e a s u r e s (utterance
I
d u ra tio n and latency of re s p o n s e ). F i r s t , how ever, th e se ta p es w ere
j
i
edited so th at the tim e during the re c y c lin g of the tape loop w as e lim i-
jnated. It m ay be re c a lle d th at the su b je c t s a t in sile n ce d u rin g the
recy c lin g of the tap e loop b e c a u se it w as im p o rta n t p ro c e d u ra lly that
the sa m e period of tim e o c c u rre d betw een p re s e n ta tio n s of stim u li
; ---------------------------------------------------------------------------------- 33_j
: i
w hether the su b ject re c e iv e d feedback or not. H ow ever, th e se ad d i-
■ 1
jtional eight seconds of sile n c e betw een u tte ra n c e s would m ak e the |
i I
listen in g ta sk unduly long and boring for the judges and would r e q u ire
' |
^n ex ce ssiv e am ount of co stly read o u t paper for the graphic level ■
i
a n a ly sis. It w as fo r th e se re a so n s th a t the tapes w ere edited. How
e v e r, the 1 kHz tone signaling the p resen ta tio n of the stim uli w as
retain ed .
F req u en cy of S tu tterin g
T his p e rc e p tu a l m e a s u re m e n t w as m ade by having th re e sp eech i
! |
pathology grad u ate s tu d e n ts, e x p erien ced with stu tte rin g , judge the
jm aster ta p e s. E ach of the judges was c u rre n tly a clinician in train in g
at the U niversity of F lo r id a 's S tu tterin g C linic. P r io r to the listen in g j
' j
se ssio n s, the six tap es for each su b ject (three p re tre a tm e n t and th re e j
p o sttre atm e n t) w ere a rra n g e d random ly by use of a table of ran d o m |
inumbers (E d w ard s, 1965, p. 333). This p ro c e d u re w as followed so
th a t the ju d g e s 1 p e rc e p tio n s would not be influenced by th e ir knowledge
jthat a group of u tte ra n c e s w ere m ade eith e r p re tre a tm e n t o r p o s ttre a t-
!
)nent. The ju d g m en ts took place d u rin g two o n e -a n d -o n e -h a lf-h o u r
jsessions (see A ppendix C fo r In stru ctio n s to the Judges). A g re e m e n t
between two of the th re e judges defined an u tte ra n c e as e ith e r fluent or
stu tte re d .
Interjudge re lia b ility w as estab lish ed by having E] liste n to the
(tapes and fo rm judgm ents of s tu tte rin g under conditions id en tical to
i ,
I
[those em ployed by the th re e ju d g es. R e s p o n s e -b y -re sp o n se a n a ly sis j
I :
: j
of the judged in sta n ces of stu tte rin g w as m ad e. P e rc e n ta g e a g re e m e n t j
pf 85 p e rc e n t betw een El and two of the th re e judges fo r in sta n c e s of
stu tte rin g only was e s ta b lis h e d by the fo rm u la :
_________A g re e m e n t________
A g re e m e n t + D is a g re e m e n t
B ecause this p ro c e d u re is dependent upon re s p o n s e -b y -re s p o n s e a n a l
y sis ra th e r than total fre q u e n c ie s only, C u rle e and P e rk in s (1970)
[found this m ethod to be m o re rig o ro u s (conservative) as an e s tim a te of
judge re lia b ility as c o m p a re d to the m ethod of ranked o r d e r correlation:,
I !
i :
jwhich is typically cited in the lite r a tu r e .
i ;
i B ecause it has been shown by W illiam s, W ark and M inifie (1963)
j I
that th e re is e s se n tia lly no d iffere n ce betw een freq u en cy of stu tte rin g j
|
judgem ents m ade fro m audio cu es alone v e rs u s those m ade fro m audio
j
and v isu al cues com bined, it is re a so n a b le to assu m e th at judgm ents
fnade by IS during the study w e re e ss e n tia lly s im ila r to those m ade by
i
jE w hile listening to the s a m e u tte ra n c e s through the tape r e c o r d e r . (It
i
should be noted that d u rin g the study .E fo rm e d judgm ents of freq u en cy
of stu tterin g and fluency la rg e ly fro m audio cu es alo n e.)
U tterance D uration and L aten cy
af R esponse
The two p h y sical m e a s u r e s of u tte ra n c e duration and la ten cy of
------------------------------------------------------------------------------------------------------------------35]
! i
j ;
[response w ere m e a s u re d by an aly sis of B ru el and K ja e r G raphic L evel
R e c o rd e r (Type 2305) tra c in g s . The graphic level re c o r d e r resp o n d s I
! i
l
to changes in sound p r e s s u r e lev el as a function of tim e . R ecordings
-
a r e m ade by m e a n s of ink on lined p a p e r. F o r this study, the re c o rd e rj
i
w as s e t with w ritin g speed a t 250 m s e c , and paper speed at 30 m s e c .
The p re tre a tm e n t and p o s ttre a tm e n t tap es w ere played through the
g rap h ic level re c o r d e r by m e a n s of an A m pex 351 fu ll-tra c k tape r e - '
c o r d e r . The e x p e rim e n te r m o n ito red the tap es w ith S harpe HA 10-A
earp h o n e s and m a rk e d a line on the tra c in g s coincident with the o n set
of sp eech . _E also identified the w ord being spoken by w riting on the j
tra c in g paper as it w as p roceeding through the graphic level r e c o r d e r .
F ig u re 4 illu s tra te s a ty p ical re c o rd in g . The tra c in g of the 1 kHz tone
I
iwhich w as re c o rd e d on the m a s te r tape coincident w ith the p re se n ta tio n
j
pf the stim u lu s word is identified at point A. P o in t B indicates the
| j
beginning of the r is e in sound p r e s s u r e level as the u tteran ce is in iti- |
jated and point C is the point w here sound p re s s u re is on its final d eclin e,
indicating te rm in a tio n of the u tte ra n c e .
i
I L atency of resp o n se w as taken as the d istan ce in m illim e te rs
I
betw een o n se t of the tone (A) and the initiation of sp eech (B). Utterance!
jduration was taken as the d ista n c e in m illim e te rs fro m initiation of j
i
Bpeech (B) to the la s t dow nw ard m o v em en t of the tra c in g (C).
A
Latency
« ----- of -----► « “
Response
B n
U tterance
Durat ion *
1 c
N
n
( u
'( 1
" v " r , r , T
| 7
Figure k. I l l u s t r a t i o n s o f events r e la t in g to i d e n t i f i c a t i o n and measurement of
laten cy of response and u tte ra n c e d u ra tio n .
[Evaluation
1
A nalysis of V arian c e w as used to evaluate d iffere n ces am ong
[the th ree tre a tm e n t m e an s for each of the dependent v a ria b le m e a s u r e s ;
frequency of s tu tte rin g , u tte ra n c e duration, and latency of re s p o n s e .
The level of confidence w hich would be accepted as evidence th at d if
fe re n c e s am ong m e an s had o c c u rre d w as set at .0 5 . F o r the e v a lu a
tion of u tte ra n c e d u ratio n and latency of resp o n se with 2 and 700+ df,
an F^ ratio of 3. 00 o r g r e a te r w as needed to achieve the d esig n ated
significance le v el. F o r the evaluation of frequency of s tu tte rin g with
2 and 16 df, an IT ratio of 3. 63 o r g re a te r w as needed.
When a sig n ifican t IT ra tio was obtained, then te s ts would be
used to co m p are d iffe re n c e s betw een any two tre a tm e n t m e a n s. B e
cause the c o m p a riso n betw een two m ean s w as not independent, it was I
! I
decided th at a m o r e strin g e n t lev el should be selected to indicate a [
[ [
singificant d iffe re n c e . Thus, the .01 level of confidence was s e t as the
level w hich would be accepted as evidence of a s ta tis tic a l d ifferen ce
; i
betw een two tre a tm e n t m e a n s, values g r e a te r than ±2. 58 w ere
needed to e s ta b lis h significant d iffe re n c e s.
C H A PTE R IV j
' I
i
RESULTS AND DISCUSSION \
I
F re q u e n c y of S tuttering
The raw data for judged freq u en cy of stu tte rin g and fluency for
each su b ject a re p re se n te d in Appendix D. P re lim in a ry ad ju stm en ts
of th e se data had to be p e rfo rm e d b efo re the tre a tm e n t m ean s could be 1
com p ared by analysis of v a ria n c e . This w as due to the fa c t th at the
I
(
judgm ents of frequency of stu tte rin g w ere obtained by a g re e m e n t fro m
i
two of th re e judges th at an u tte ra n c e was spoken eith e r fluently o r
|
s tu tte re d . As a re s u lt, th e se data w ere d istrib u te d in a bim odal
I
fashion and could not m e e t the re q u ire m e n t fo r hom ogeneity of v a r i-
I
Jance. Follow ing the reco m m en d atio n of M endenhall (1967) for such an
i
in s ta n c e , th e se freq u en cies w e re n o rm a liz e d by in v e rse sine t r a n s
fo rm a tio n . The difference betw een the frequency of stu tte rin g fo r p r e -
itreatm ent and p o sttre a tm e n t p erio d s w ere then obtained by su b tractin g
i
Ithe sine converted s c o re for p re tre a tm e n t fro m the sine co n v erted
i
|
jscore for p o s ttre a tm e n t. In th is way, a (+) value w as obtained if the
p o sttre a tm e n t freq u en cy w as g r e a te r than that for the p re tre a tm e n t
38
freq u en cy , and c o n v e rse ly , a (-) value if the p o sttre a tm e n t fre q u e n c y
i i
|was le s s . The obtained m e an d iffe re n c e s (in sine converted s c o re s ) j
i I
fo r the th re e tre a tm e n ts w e re : No F eed b ac k , -0. 14; F luency Feedback^
-0 .0 8 ; and S tu tterin g F eed b ack , -0 .0 2 . i
An a n a ly sis of v a ria n c e for re p e a te d m e a su re s was then p e r -
i I
fo rm ed co m p arin g the m e a n d iffe re n c e s am ong the tre a tm e n ts . A
isum m ary of th a t an aly sis is shown in T able 3. As can be seen fro m j
the tab le, the obtained F ratio failed to re a c h the designated lev el (.05);
bf significance. Thus, H ypothesis 1 w hich stated that the m e a n d iffe r- !
ence of freq u en cy of stu tte rin g fo r the th re e tre a tm e n ts d iffere d sig n ifi
cantly fro m each o th e r, w as not su p p o rted . In other w o rd s, the m e a n j
j
freq u en cies of stu tte rin g for the th ree tre a tm e n ts differed by no m o re
j
(than the am ounts th at could be a ttrib u te d to chance v aria tio n . I
! T hese findings su g g est th at w hen s tu tte r e r s are provided w ith J
!
additional in fo rm atio n about th e ir sp eech p erfo rm a n ce by liste n in g to a
re c o rd e d rep la y of th e ir u tte ra n c e s it d o es not se e m to m ak e any d if-
j I
fe re n c e w ith r e s p e c t to su b seq u en t freq u en cy of stu tterin g o r fluency,
w hether o r not they h e a r th e ir fluent p ro d u ctio n s, or th e ir s tu tte re d
i
productions, o r n e ith e r.
j
With r e g a rd to the se rv o m e c h a n is m m odel, as applicable to
stu tterin g , th e se findings su g g est that the sp e a k e r w as not able to use
the in fo rm atio n he re c e iv e d fro m feedback in h is effort to re d u c e the
freq u en cy of his stu tte rin g . H ow ever, it should be understood th a t the
TA B L E 3
SUMMARY OF TH E ANALYSIS OF VARIANCE OF MEAN
D IFFER EN C ES IN STU TTERING FREQUENCY FOR
THREE TREATM ENTS: NO FEED B A C K , FLUENCY
FEED BA CK , AND STU TTERIN G FEEDBACK
1 -------------------------------
Source df SS MS F
Subjects 8 .47 .0 6 -
T re a tm e n ts
2 . 06 . 03
R esid u al 16
00
. 06
-
Total 26 1.51
i--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------rr\
! I
! !
(m easure of frequency of stu tte rin g is a g ro s s tool in the m e a s u re m e n t i
i
of sp eech p e rfo rm a n c e and m ay not d e te c t ap p ro x im atio n s tow ard the
; i
reduction in e r r o r signal. N e v e rth e le ss , this m e a s u re w as included in
the study on the po ssib ility th at the tre a tm e n t effects would be g re a t
enough to be refle cted by th is tool.
F u r th e r c o n sid e ra tio n should be given to another potential in t e r
p reta tio n of th ese r e s u lt s - - f r o m the fram e w o rk of the o p e ra n t co n d i
tioning m o d el. One m ight intuitively c la im that feedback of fluency
would co n stitu te re w a rd and feedback of stu tte rin g would constitute
p unishm ent. F u r th e r m o re , r e s e a r c h e r s have a ttrib u ted rein fo rc in g
q u alities to both fluency and stu tte rin g . Van R iper (1954, p. 370) has |
w ritte n of the se lf-p e rp e tu a tin g n a tu re of stu tte rin g , Sheehan (1953) hast
; _ |
W ritten of the need to stu tte r a fte r p erio d s of h e a rin g o n eself be fluent,
and S ham es and S h e rric k (1965) have w ritten of the positive and negative
I
s e lf-re in fo rc in g c h a r a c te r is tic s of the s tu tte r e r 's own v e rb a l re s p o n s e s .
[The r e s u lts of this study, h o w ev er, do not su p p o rt the hypothesis that |
; I
; i
(the e s s e n tia l n atu re of eith er stu tte rin g o r fluency is rew ard in g o r
punishing. The lack of su p p o rt for this hypothesis m ay be e ith e r:
|(1) n e ith e r feedback of stu tte rin g n o r fluency in h eren tly h a s rew ard in g
b r punishing c h a r a c te r is tic s , o r (2) feedback of fluency o r stu tterin g
I
!
jnay have re in fo rc in g c h a r a c te r is tic s but they w ere not d e m o n stra te d in
s study b eca u se the p ro c e d u re s did not sufficiently follow those of
he o p e ra n t m odel. T hese re a s o n s would su g g est the need to develop a
- - ^
I
; (
study w hich would independently e s ta b lis h the rew ard in g o r punishing
! i
c h a r a c te r is tic s of th e se two types of au d ito ry feedback. j
The findings of this study a re also re le v a n t to the clin ical te c h - j
I
nique of having s tu tte r e r s liste n to tape re c o rd in g s of th e ir u tte ra n c e s , i
The value of th a t clin ical p ro c e d u re is questioned by this study. How
e v e r, th is questioning is m ade cautiously becau se this study em ployed
feedback of sin g le -w o rd u tte ra n c e s only r a th e r than feedback of con- j
nected sp eech and the u su a l clin ica l p ro c e d u re h as been to have the
clin ician and s tu tte r e r to g e th er listen to portions of connected speech.
A gain, it should be reco g n ized th at the freq u en cy of stu tterin g as a |
m e a s u r e is g ro s s in n atu re and m ight fail to re fle c t other speech p e r
fo rm a n c e ch an g es. Inform ation re le v a n t to this question is provided by!
i
i
ithe additional two an aly se s that w e re p e rfo rm e d on each u tte ra n c e . i
I i
i i
jThese two a n a ly se s a re p re se n te d below . j
; i
j I
U tteran ce D uration
The second m e a s u re of p e rfo rm a n c e change was u tte ra n c e d u r a
tion. The p re tre a tm e n t and p o s ttre a tm e n t u tte ra n c e d u ratio n s w ere
I
(m easured to the n e a r e s t whole m illim e te r. The d iffere n ces w ere then
balcu lated by su b tra c tin g the p r e tre a tm e n t d u ratio n s fro m the p o s t
tre a tm e n t d u ratio n s for each w ord. T hese d iffere n ce s c o re s tog eth er
with th e ir m e a n s and stan d ard d eviations a re p resen te d in Appendix E.
It should be noted th at a (+) in d icate s th a t the du ratio n in c re a s e d fro m
jthe p re tre a tm e n t to the p o sttre a tm e n t and h a s becom e le s s when th e |
j J
lvalue noted is a (-). ,
; !
Next, an an aly sis of v aria n ce (see T able 4) w as p e rfo rm e d on j
the d iffere n ce s c o re s in o r d e r to d e te rm in e w hether the th re e tre a tm e n t
m e a n s w ere significantly d ifferen t. With 2 and 783 df, an F ratio of
3. 00 w as needed fo r significance at the . 05 le v el of confidence. T he F
ratio of .40 for tre a tm e n t effects w as thus not significant. This in d i-
: I
: i
cates that for this study, feedback of this kind did not contribute signifi-j
cantly to u tteran ce d u ratio n . That is , it did not m ake a difference
w h eth er or not the sp e a k e r w as given inform ation about h is p e rfo rm - j
: !
ance through feedback of h is fluent u tte ra n c e s , feedback of his stu ttered
u tte ra n c e s , o r if he did not receive any feedback a t all. T h e re fo re ,
jthis study failed to provide support fo r H ypothesis 2. A ccording to this
i
jhypothesis, th e re w as to hav e been a significant d iffere n ce in the m ean
jd ecrease in u tte ra n c e duration am ong the th r e e tre a tm e n ts . B ecause
bf the n o n -sig n ifican ce of the F te st, th e re w as no need to p e rfo rm
.separate tre a tm e n t c o m p a riso n s.
I It m ay be, how ever, that the p ro c e d u ra l d ifficu lties that w e re
jencountered during th e running of su b jects a r e of im p o rta n c e and should
be c o n sid e re d . T h at is, it w as n e c e s s a ry to re ru n an e x p e rim e n ta l
condition for two su b jects. This w as due to fa ilu re of the p e rm a n e n t-
re c o rd in g equipm ent to function p ro p e rly . A s a re s u lt, p erm a n en t
r e c o rd s of the o rig in a l p re tre a tm e n t and p o s ttre a tm e n t p e rfo rm a n c e s
TABLE 4
SUMMARY OF TH E ANALYSIS OF VARIANCE OF MEAN
D IFFE R E N C E S IN U TTERANCE DURATION F O R TH R E E
TREATM ENTS: NO FEED BA CK , FLUENCY FEED B A C K ,
AND STUTTERING FEEDBACK
Source df SS MS F
Subjects 8 428089.40 53511.17
T re a tm e n ts 2 4011.28 2005.64
R esid u al E r r o r 783 3924383.59
5011.98
T otal 793 4356484.27
w e re not available for m e a s u re m e n t and s ta tis tic a l a n a ly sis . D uring
!
;the o rig in a l p e rfo rm a n c e s of the fluency feedback tre a tm e n t fo r both
su b jects, _E had tabulated a sizab le d e c re a s e in freq u en cy of stu tte rin g j
fro m the p re tre a tm e n t to the p o sttre a tm e n t p e rio d s. D uring the re ru n j
p e rfo rm a n c e s, how ever, the su b jects p e rfo rm e d quite d iffere n tly . The:
freq u en cy of th e ir stu tte rin g , and consequently th e ir u tte ra n c e d u r a
tio n s, w e re notably red u ced during the re ru n p re tre a tm e n t period as j
co m p ared to the o rig in a l p re tre a tm e n t p erio d . T hus, becau se of the
reduced freq u en cy of stu tte rin g at the s ta r t of the re ru n se ssio n , th e re !
w as not the sam e opportunity fo r reduction in stu tte rin g as th e re had J
been during the o rig in a l se s s io n . T h ese o b se rv a tio n s can be co n stru ed |
I
to m ean that for c e rta in su b jects the effect could have been significant, j
I
i A lso of c o n s id e ra b le im p o rtan ce is the o b se rv a tio n fro m th ese
two in sta n ces that the effects la ste d fo r at le a s t five day s, the length of
jtime w hich s e p a ra te d the r e r u n fro m the o rig in a l se s sio n . These
o b se rv a tio n s, though unfortunately not docum ented in the p re s e n t in
stance, a r e m o s t pro v o cativ e and su g g est a need fo r fu rth e r in v e stig a
tion.
j F u rth e r, it should be noted th at tre a tm e n ts and su b jects w e re
janalyzed as the m a in v a ria b le s contributing to the e r r o r te rm ; the
e r r o r attrib u ted to o rd e r effects is included in the re s id u a l e r r o r .
H ow ever, the o rig in a l plan of the study w as to c o n tro l o r d e r effects
through a Latin Square d esig n . B ecause of the p ro c e d u ra l p ro b le m
4?n
i
i
noted above, it w as im p o ssib le to analyze the d a ta fo r th ese effects.
I
(Accordingly, the re s id u a l e r r o r is la rg e r than it m ig h t o th erw ise have i
Ibeen. B eca u se the re s id u a l e r r o r functions as the d en o m in ato r of the
j
|F ra tio , it b eco m es in c re a sin g ly difficult to obtain a sig n ifican t ratio
when the e r r o r te r m is la rg e . T hus, it is po ssib le th a t the la rg e
I
d en o m in ato r of the F ra tio , com bined with the lo ss of the data during
the o rig in a l s e s sio n s of the two p rev io u sly m entioned su b jects, m ay
have o b sc u re d d iffe re n c e s th at m a y actually e x ist.
While the r e s u lts of the an aly sis of the u tte ra n c e duration v a r -
i
iab le did not su p p o rt any of the hypotheses as sta te d , conclusions
d eriv e d fro m th e se findings should be guarded. The p ro c e d u ra l d iffi-
!
pulty en co u n tered in the study, as d escrib e d above, m a y have o b scu red
[ :
r e a l effects.
L atency of R esponse
j i
| The th ird in d icato r of sp eech change w as the m e a s u re m e n t of j
jthe laten cy of re s p o n s e . T h e re w e re ap p ro x im ate ly 30 p re tre a tm e n t j
and 30 p o s ttre a tm e n t m e a s u re m e n ts m ade for eac h of the nine subjects
p erfo rm in g each of the th r e e feedback tre a tm e n ts . The actual num ber
varied fro m the m a x im u m of 1, 620 because o c c a sio n a lly a su b ject did
not u tte r a stim u lu s w ord. When a latency included a fra c tio n of a
m illim e te r, the m e a s u re m e n t w as alw ays rounded off to the n e a r e s t
whole v alu e. The d iffe re n c e s in m illim e te rs betw een the p re tre a tm e n t
; ' 47
i
i
jlatency and the p o s ttre a tm e n t latency of resp o n se for each stim u lu s
w ord as spoken by each su b ject w as then calcu la ted . The p re tre a tm e n t
latency w as su b tra c te d fro m the p o sttre a tm e n t latency so th at a (-)
1
iwould indicate a red u ctio n in latency and a (+) would indicate an in
c re a s e in la ten cy . The d iffere n ce s c o re s , m ean s and stan d ard d e v ia
tions a re liste d in Appendix F .
An a n a ly sis of v a ria n c e (see Table 5) w as then p e rfo rm e d to
co m p are the m e a n s of the th re e tre a tm e n ts . W ith 2 and 772 df, a ratio
of 3.0 0 o r above w as needed fo r significance at the .0 5 level of co n fi
dence and a ra tio of 4. 61 o r above w as needed fo r significance a t the
.01 level of confidence. The obtained F ratio for tre a tm e n ts of 9.51
w as significant beyond the .01 level of confidence. T his indicated th a t
i
(latency of re s p o n s e v a rie d as a function of the type of tre a tm e n t the j
| i
sp eak er w as given. T hus, H ypothesis 3, which stated th a t the m ean i
I
d e c re a s e in laten cy of re s p o n s e would differ significantly am ong the j
i ,
th re e tre a tm e n ts of au d ito ry feedback, w as su p p o rted . !
, I
In an e ffo rt to d e te rm in e w hich of the tre a tm e n ts cau sed the
i
i
significant d iffere n ce in the latency of resp o n se m e a s u re , individual
c o m p ariso n s am ong the th re e tre a tm e n t m eans w ere p e rfo rm e d . The I
th re e tre a tm e n t m e a n s w e re : No F eed b ack , .57; F luency F eed b ack , |
-11 .6 1 ; and S tu tterin g F eed b ac k , 3 .9 8 . A tw o-failed te s t w as used
in w hich the m e a n sq u a re e r r o r of T able 5 w as used as an e s tim a te of
population v a ria n c e . B ecau se th re e c o m p a riso n s w e re p e rfo rm e d in
TA B LE 5
SUMMARY O F ANALYSIS OF VARIANCE OF MEAN
D IFFE R E N C E S IN LATENCY O F RESPONSE FOR
T H R E E TR EA TM EN TS: NO FEED BA CK ,
FLU EN CY FEED BA CK , AND
STU TTERING FEED BA CK
Source df SS MS F
Subjects 8 63652.95 7956.62 4. 54*
T re a tm e n ts
2 33334.47 16667.23 9.51**
j R esidual E r r o r 772 1352856.95 2752.40
Total 782 1449844.17
, --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 49^
I
i
I
j
iwhich the sa m e e r r o r te rm w as used as an e s tim a te of population v a r i - 1
jance, the th re e co m p a riso n s w e re not independent. It w as thus decided!
Ithat a m o re strin g e n t level of significance should be s e le c te d . T hus, j
[the .01 level of confidence w as set as the lev el of a cc ep tan c e of signifi-j
cance for d iffe re n c e s betw een m e a n s. V alues of z above +2.58 or
below -2 .5 8 w e re significant.
The f ir s t c o m p a riso n m ade w as betw een the co n tro l tre a tm e n t i
of no feedback and the e x p e rim e n ta l tre a tm e n t of fluency feedback.
With a m ean of .57 fo r no feedback and -11 .6 1 for fluency feedback, the
! ;
d iffe re n c e betw een m e an s w as -12. 18, yielding a s c o re of -3. 16. j
i
T his d ifferen ce w as significant beyond the .01 level. T hus, Hypothesis!
;3a (that the latency of re sp o n se would be significantly s h o rte r when the 1
sp eak er rece iv ed fluency feedback fo r his fluent u tte ra n c e s than when
he rece iv ed no feedback for any of his u tte ra n c e s) w as supported. j
; The second c o m p a riso n m ade w as betw een the c o n tro l tre a tm e n t
!and the stu tte rin g feedback tre a tm e n t. With a m ean of . 57 fo r the no
j
feedback tre a tm e n t and a m e an of 3. 98 fo r the stu tte rin g feedback
I
jtreatm ent, the d iffere n ce betw een the m ean s w as 3.41 yielding a
i
s c o re of + .8 8 . T his d ifferen ce w as not significant a t the p re s c rib e d
level of .0 1 . T hus, H ypothesis 3b w as not su p p o rted . A pparently, it
m ade no d ifferen ce w ith re g a rd to latency of re s p o n s e w hether the
sp e a k e r w as given feedback of h is stu tte re d re s p o n s e s or w hether he
w as not given any feedback. When given additional in fo rm atio n about
Jus e r r o r s , as w as the c a se in stu tte rin g feedback tre a tm e n t, the j
jspeaksrs app aren tly did not use the inform ation to th e ir advantage by
red u cin g th eir e r r o r s as they had done for the fluency feedback t r e a t- j
jment. !
j
The final c o m p ariso n m ade w as betw een the m e a n s of the
fluency feedback tre a tm e n t w ith a m ean of -1 1 .6 1 and the stu tte rin g
feedback tre a tm e n t with a m ean of 3 .9 8 . The d ifferen ce of +15.50 led ;
i
to a sco re of +3. 10 w hich w as significant a t the .01 level of confi-
I
dence. Thus, H ypothesis 3c (that th e re would be a sig n ifican tly g re a te r
reduction in latency of re s p o n s e following fluency feedback, than fo l- |
I
lowing stu tte rin g feedback) w as supported. O nce again, fluency fe e d - j
back tre a tm e n t re su lte d in significantly s h o rte r laten cies of re s p o n s e , j
!
T hese findings su g g est th a t w hen given inform ation about o n e 's fluent j
i i
I
.productions, as was the case in the fluency feedback tre a tm e n t, the j
; |
sp eak er w as able to use the in fo rm atio n to h is advantage, and change !
i
ihis subsequent p e rfo rm a n c e in the d irec tio n of reduced e r r o r signal
p erfo rm a n c e and consequently red u ced latency of re sp o n se .
T hese findings in d icate that latency of resp o n se is a sp eech
I
behavior which can be changed as a function of p resen tin g additional
{information through au d ito ry feed b ack . F u rth e rm o re , the type of
i
inform ation w hich the s p e a k e r can u tilize is th a t concerning h is c o r r e c t
p e rfo rm a n c e . When the s tu tte r e r w as given feedback of his fluent
u tte ra n c e s he w as able to re d u c e the latency of his re s p o n s e s ; this w as
----------------------------------------------------------------------------------------------------------------------- 5TT
not the c a se when the s tu tte r e r w as given feedback of h is s tu tte re d
re sp o n s e s, or w hen he w as not given feedback of any of his u tte ra n c e s .
It is in te re s tin g th a t fluency feedback re su lte d in significant
reduction in latency of re s p o n s e but did not re s u lt in any changes in
the other two m e a s u r e s : u tte ra n c e d u ratio n or frequency of s tu tte rin g .
[A possible explanation for this finding m ig h t be that latency of resp o n se
is a sp eech behavior w hich is m o re re a d ily am enable to change than
j
jare the o th e r two m e a s u r e s . It m ight ju s t be th at latency of re sp o n se
!was but the f ir s t of the th re e m e a s u r e s to change. T hat is, the other
I
two m e a s u re s m ig h t have changed also had the study gone on fo r a
i
longer period of tim e and had m o re in sta n c e s of feedback been p r e
sented to the s p e a k e rs .
The m a jo r finding fro m this study, that is, the red u ced latency
of resp o n se as a function of fluency feedback, h as im plications fo r the
clinical technique of having the s tu tte r e r listen to tape re c o rd in g s of
i
h im se lf speaking. It would ap p ea r that he would re c e iv e g r e a te s t
jbenefit if those ta p es w ere edited so th a t he listen ed to h is c o r r e c t
productions only.
C H A PTE R V
SUMMARY, CONCLUSIONS AND IMPLICATIONS
1 S u m m ary
i
The P ro b le m
! -----------------------------------
This study w as c o n ce rn ed w ith the speech behavior of people
Who exhibit d isflu en cies w hich a re desig n ated as s tu tte rin g . The p u r
pose w as to investigate the effect of receiv in g auditory feedback of the
jsp eak er's own disflu en t sp e e c h o r of his own fluent sp eech upon h is
subsequent sp eech p e rfo rm a n c e . T his approach w as d e riv e d fro m a
feedback m odel of the sp e e c h m e c h a n ism which a s s e r ts th at speech
production is actu ally co n tro lled by feedback fro m the output. An e rro r
signal is produced when the a ctu al output does not m atch the intended
output. Then ad ju stm en ts in output a re m ade so as to re d u c e the
Amount of the e r r o r signal.
|
The q u estio n s r a is e d in re la tio n to the m odel, then, w e re :
, 1. If the sp eak er w e re given inform ation about h is e r r o r signal
j p e rfo rm a n c e by w ay of receiv in g auditory feedback of his
j stu tte re d productions only (stu tterin g feedback), would he
show evidence of using the inform ation to ad ju st his p ro d u c
tion of su b seq u en t sp eech and th ereb y achieve le s s e r r o r
s ig n a l?
52
53
2. If the sam e sp eak er w e re given inform ation about his
c o r r e c t productions by re c e iv in g only feedback of his
fluent productions (fluency feedback), would this type of
in fo rm atio n influence h is production of subsequent sp e e c h ?
The independent v a ria b le s w e re two types of au d ito ry feedback:
feedback of fluent prod u ctio n s, and feedback of stu tte re d p ro d u ctio n s.
i
The dependent v a ria b le s , changes in sp eech p e rfo rm a n c e , w e re m e a s
u red by th re e p a ra m e te rs of speech: freq u en cy of stu tterin g , u tte ra n c e
duration, and latency of re sp o n se .
1 The m a jo r h ypotheses g e n e ra te d fro m the questions above w e re :
j
! 1. The d e c re a s e in freq u en cy of stu tte rin g would be sig n ifi
cantly g r e a te r (p < .05) w hen a sp eak er was given eith er
I fluency feedback or s tu tte rin g feedback, as co m p ared to a
tre a tm e n t in which he re c e iv e d no feedback (control).
2. The d e c re a s e in the freq u en cy of stu tte rin g would be signifi-
j cantly g r e a te r (p < .01) when the sp eak er rece iv ed feedback
i of his fluently spoken u tte ra n c e s than when he re c e iv e d
feedback of his stu tte re d re s p o n s e s .
With re g a rd to the second m e a s u re , u tte ra n c e du ratio n , w hich w as
defined as the te m p o ra l difference betw een the initiation of a re s p o n s e
and the com pletion of the re sp o n se , the following hypotheses w ere
jmade:
3. The d e c re a s e in d u ratio n of an u tteran ce would be sig n ifi
cantly g r e a te r (p < .05) a fte r a sp e a k e r rece iv ed e ith e r
fluency feedback or .stu tterin g feedback than a fte r he
! rece iv ed no feedback.
i
4. His d u ratio n s would be significantly s h o rte r (p < .01) when
he rece iv ed fluency feed b ack than when he rece iv ed s tu tt e r
ing feedback.
54
F in ally , a th ird set of hypotheses re g a rd in g latency of re sp o n se
w ere developed. L atency of re sp o n se was defined as the tim e fro m the
[presentation of the stim u lu s w ord to the initiation of the re s p o n s e . It
I
was hypothesized that:
5. T h e re would be a significantly g r e a te r (p < . 05) d e c re a s e
in the latency of re sp o n se when the sp eak er rece iv ed
fluency feedback o r stu tte rin g feedback, as co m p ared to
the co n tro l tre a tm e n t of rece iv in g no feedback.
j 6. T h e re would be a significantly g r e a te r d e c re a s e (p < .01)
in laten cy of resp o n se when the sp eak er rece iv ed feedback
of his fluently spoken u tte ra n c e s than when he re c e iv e d
1 feedback of his s tu tte re d u tte ra n c e s .
Method
Nine s tu tte r e r s se rv in g as th e ir own co n tro ls re a d w ords aloud
under th re e tre a tm e n ts of aud ito ry feedback: fluency feedback, s tu tt e r
ing feedback, and no feedback (control). The fluency feedback t r e a t
m e n t c o n siste d of the im m ed iate re p la y of each u tteran ce that w as
j
|judged to be spoken fluently. The stu tte rin g feedback tre a tm e n t w as
the im m ed iate rep lay of eac h u tte ra n c e that w as judged to have been
[stuttered. F o r the co n tro l tre a tm e n t, none of the s p e a k e r's u tte ra n c e s
I
w e re re p la y e d . A s e r ie s of stim u lu s w ords w ere spoken p re tre a tm e n t
jand then again p o sttre a tm e n t. M e a su re m e n ts w ere m ade of th re e
‘speech p e rfo rm a n c e v a ria b le s : freq u en cy of stu tte rin g , u tte ra n c e
d u ratio n , and latency of resp o n se d u rin g the p r e tre a tm e n t and p o s t
tre a tm e n t p e rio d s. C o m p a riso n s betw een the p re tre a tm e n t and
^
p o sttre a tm e n t p erio d s of th e se re sp o n se s w ere then m a d e . The m ean
d ifferen ces w ere c a lc u la te d . T h ese d iffere n ces w ere then co m p ared
for the th re e tre a tm e n ts by m e a n s of an aly sis of v a ria n c e . C o m p a ri
sons betw een two tre a tm e n t m e an s w ere evaluated by te s ts .
R esults
The f ir s t c o m p a ris o n w as am ong the m e a n d iffere n ces in
stu tte rin g freq u en cy for the th re e d iffere n t feedback tre a tm e n ts . The
obtained F ra tio w as not sig n ifican t at the designated (.05) level of
fconfidence.
| The next c o m p a riso n w as am ong the m ean d iffere n ces in u tte r -
I
^nce d u ratio n for the th re e d iffe re n t feedback tre a tm e n ts . Again, the
|obtained F ratio w as not sig n ifican t at the .05 lev el of confidence.
I
i
The th ird c o m p a riso n , that for the m ean d ifferen ce in latency
Df re sp o n se for the th re e tre a tm e n ts , yielded an F w hich w as sig n ifi
cant a t the .01 le v el of confidence, s s te s ts rev ea led d iffe re n c e s a t the
.01 level of confidence betw een the fluency feedback and the co n tro l
tre a tm e n ts and also betw een the fluency feedback and stu tte rin g f e e d
back tre a tm e n ts . In both c a s e s fluency feedback re su lte d in sig n ifi
cantly s h o rte r la te n c ie s,
j
C onclusions
The r e s u lts of this investigation su p p o rt the following co n clu
sio n s:
56
2.
4.
! Im plications
The finding th a t the s p e a k e rs took significantly s h o rte r tim e to
begin th e ir u tte ra n c e s (latency of resp o n se) a fte r having listen ed to
feedback of th e ir fluent sp eech su g g e sts definite rele v an ce of the s e r v o -
fn ech an ism m odel of sp e e c h production to the p ro b lem of stu tte rin g .
T hat is, the sp e a k e r ap p are n tly sam p led fro m the fed -b ack u tte ra n c e s
and adjusted h is su b se q u en t u tte ra n c e s by sh ortening the tim e he spent
in p re p a rin g to speak.
i
i
Although feedback of fluency, as w ell as stu tte rin g , seem e d to
have little effect on the o th e r m e a s u r e s ex p lo red in the study (frequency
of stu tte rin g and d u ratio n of u tte ra n c e s), this does not n e c e s s a rily
im ply th at the feedback m o d el la ck s applicability to th e se two m e a s u r e s
R eceiving au d ito ry feedback of e ith e r fluently spoken or
s tu tte re d u tte ra n c e s seem s to have little effect on the
freq u en cy of su b seq u en t stu tte rin g , at le a st w hen single
w ord u tte ra n c e s a re involved.
R eceiving au d ito ry feedback of e ith e r fluently spoken or
stu tte re d u tte ra n c e s does not appear to have an effect upon
the am ount of tim e taken to say subsequent w o rd s, at le a s t
for sin g le -w o rd u tte ra n c e s .
R eceiving aud ito ry feedback of u tte ra n c e s w hich a re judged
to be spoken fluently, re s u lts in a significantly s h o rte r
latency of re sp o n se on subsequent u tte ra n c e s than rece iv in g
no auditory feedback of such u tte ra n c e s .
R eceiving aud ito ry feedback of u tte ra n c e s w hich a r e judged
to be spoken fluently r e s u lts in a s h o rte r latency of r e
sponse on su b seq u en t u tte ra n c e s than receiv in g auditory
feedback of u tte ra n c e s w hich a re judged to have been
s tu tte re d .
5 T
[t m ight be that the latency m e a s u re is m o re resp o n siv e to change than
are the o th e r two m e a s u r e s . F re q u e n c y of stu tte rin g and u tte ra n c e
duration m ight have been changed also if a g re a te r num ber of feedback
p resen ta tio n s had been given. A nother m a jo r co n sid era tio n is the
lim itation of the study to the inv estig atio n of single-w ord u tte ra n c e s .
C hanges in freq u en cy of stu tte rin g and u tte ra n c e duration m ig h t have
[been obtained also if the m ethodology had allowed fo r feedback of c o n
tinuous sp eech ra th e r than of sin g le -w o rd u tte ra n c e s .
R efinem ents in the study m ight have d em o n strated th at auditory
feedback can affect frequency of s tu tte rin g and u tteran ce du ratio n as
w ell as latency of re s p o n se . F o r ex am p le, a m a jo r im p ro v em en t would
I
have been m ade in the e x p e rim e n ta l d esig n by using eith er a m uch
la rg e r sam ple of su b jects (at le a s t 30) o r by em ploying single subject
design. Although the a n a ly sis w as p e rfo rm e d upon a larg e num ber of
resp o n ses fro m each of the nine su b je c ts, n o n eth eless, the re sp o n se s
I
to feedback of the v a rio u s s p e a k e rs w ere quite individualized and r e
sulted in la rg e e r r o r te r m s .. C onsequently, possible effects upon f r e
quency of stu tte rin g and u tte ra n c e d u ratio n m ay have been o b sc u re d .
F u rth e r, because of the v a ria tio n betw een su b jects, it beco m es difficult
!
to g e n e ra liz e the findings to a population of s tu tte r e r s . In selectin g
j
e ith e r a lte rn a tiv e m ethod for fu rth e r investigation (la rg e r sam p le o r
single su b ject design) one would w ant to study the sp e a k e rs in te rm s of
th e ir se v e rity of s tu tte rin g and in te rm s of th e ir type of stu tterin g
----------------------------------------------------------------------------------------------------------------------- 5B“
(silent s tu tte r e r s v e rs u s v o c a l s tu tte r e r s ) . In th is way, one would be
able to d e te rm in e w hether o r not the effects of au d ito ry feedback a re
m ore effective w ith c e rta in disfluent s p e a k e rs . In o th e r w o rd s, is it
jpossible to identify an in te ra c tio n betw een type of feedback and type of
jdisfluent behavior ?
O th er im p ro v e m e n ts could be m ade p ro c e d u ra lly . F o r exam ple,
i
the a p p a ra tu s u sed for the p re s e n t study could be red esig n e d e le c tr o n
ic ally so th at m o re p re c is e co n tro l of u tteran ce re p la y would be p o s
s ib le . In addition, this would allow for the feedback of p h ra s e s o r
sen ten ces ra th e r than single w o rd s.
; With re g a rd to im p lic atio n s for therap y , only guarded s ta te -
j
m ents can be m a d e. Given a clin ical situation som ew hat s im ila r to the
p ro c e d u re s of this study, the efficacy of p resen tin g e ith e r fluency o r
stu tte rin g feedback of sin g le -w o rd u tte ra n c e s s e e m s of questionable
value if o n e 's purpose is to a lte r the freq u en cy o r d u ratio n of s tu tte rin g ,
T his m a y c a s t som e doubt, th e re fo re , on the p ra c tic e of having clien ts
liste n to th e ir sp eech being played back on a tape r e c o r d e r . H ow ever,
this is not the c a s e if it is co n sid ere d th e ra p e u tic a lly advantageous to
jreduce the tim e th a t the c lie n t takes in p re p a rin g fo r the initiation of
i
|the sp eech a ct. Such an in sta n c e would be that of the s tu tte r e r who
p e rfo rm s ex ten siv e r e h e a r s a l or p re p a ra tio n as a c h a r a c te r is tic a sp ect
of his stu tte rin g .
59
In s u m m a ry , although an effect due to au d ito ry feedback w as not
found for freq u en cy of stu tte rin g and u tte ra n c e d u ra tio n s, it was found
for the period of tim e that the sp eak er took in p re p a rin g for his u t t e r
ance. T hat is, the sp eak er took le ss tim e fro m the p re se n ta tio n of the
stim u li to the initiation of his resp o n se subsequent to receiv in g auditory
feedback of his fluently spoken u tte ra n c e s as co m p ared to rece iv in g
!
feedback of his s tu tte re d re s p o n s e s , o r to rece iv in g no feedback. Thus
In p a rt, the h ypotheses g en erated fro m the s e rv o m e c h a n is m m odel of
Ispeech production w ere supported. It was also pointed out th a t even
i
i
g r e a te r ap p licab ility of the m odel to the p ro b lem of stu tte rin g m ight be
jdem onstrated given those p ro c e d u ra l refin e m en ts th a t w ere suggested.
A P P E N D I X E S
I
60
A PPEND IX A
STIMULUS WORDS
61
~5z
PR A C TIC E WORDS
ice count rh e u m a tism
fo rtificatio n dem o n stratio n ju risp ru d e n c e
led definitely labor
fa b ric a tio n b less sad
zoologist favor reh ab ilitatio n
bitum inous path re frig e ra to r
am ount stim u latio n baby
noon p ro fe s s o rs h ip n e v e rth e le ss
knee neighbor deriv ativ e
thick th e rm o m e te r nation
re m a rk a b ly soil point
cloud m a rk e t re c o n stru c tio n
w est y ard queen
p ro s p e rity te rm in a tio n p articip a n t
clothes technology fact
m o m e n ta rily significantly bew ilderm ent
sa t w ash fat
fig u re w eight h o rtic u ltu ra l
cham pionship v ocational biographical
fo rce se n tim e n ta lity fourth
c o m p ariso n fix fertiliz atio n
ra n m agnificent c h a ra c te ris tic a lly
shoulder ta il
joy
tir e c h a rita b le fa v o ritism
d issip atio n ocean benediction
e lse rep ly n o m enclature
tra n q u ility
S elected fro m : E . L. T horndike and I. L o rg e, The T eacher*s
W ord Book of 30,000 W ords (New Y ork: T e a c h e rs C ollege, C olum bia
U n iv ersity , 1944).
63
P R E -P O S T WORDS I
s e m ic irc u la r
cent
p assionately
lift
duty
religiously
m a te ria lis tic
sy m p ath izer
dare
lord
cosm opolitan
w aterm elo n
busy
building
superhum an
v e te rin a ry
cloth
w h atso ev er
battle
cup
deal
a r t
qualification
tech n icality
gate
h u rt
bottom
te leg rap h ic
significant
v isitatio n
S elected fro m : E. L. T horndike and I. L o rg e, The T eacher*s
W ord Book of 30,000 W ords (New Y ork: T e a c h e rs C ollege, C olum bia
U n iv ersity , 1944).
Z4
TREA TM EN T WORDS I
re p re se n ta tio n size adult
guide w orth captain
discontented sociological celeb ratio n
justification ventilation p a ra p h e rn a lia
b u reau c ra cy painstakingly leg islatio n
dependency notoriety tra d itio n a lly
roof tw elve yellow
tra v e l com bination touch
stood spot race
circulation v ocabulary heaven
resto ra tio n strik e dom ination
to leratio n b ru tality post
vaccination lu m in ary pen
w onderfully noticeably lib ra ria n
v eg etarian ru d im e n ta ry shoe
scientifically d isco lo ratio n nose
m a s te r tongue c la ssific a tio n
escape note bee
subject mythology luxuriously
cow wild gam e
so cialistic culum ination h u rry
w ear proud h o rizo n tally
night public tra n s a tla n tic
le g islato r choose bag
m a caro n i sp re a d b ra n c h
enjoy
revolution
su scep tib le shut
S elected fro m : E . L. T horndike and I. L o rg e, The T e a c h e r ’s
W ord Book of 30, 000 W ords (New Y ork: T e a c h e rs C ollege, C olum bia
U n iv e rsity , 1944). _________________________ _______ _
P R E -P O S T WORDS II
dilapidated
m ig ra to ry
c re d ita b le
band
join
h y d ro ch lo ric
p e c u lia rly
m oney
iron
r iv e r
lion
p erso n ific atio n
probability
te ach
no r
p o p u la rity
com ing
su ffer
p ro fe ssio n a lly
m a n a g e ria l
coalition
brow n
ju risd ic tio n
cook
m a n n e r
coincident
population
page
m o m e n ta ry
m outh
S ele cted fro m : E . L. T horndike and I. L o rg e, The T eacher*s
W ord Book of 30,000 W ords (New York: T e a c h e rs C ollege, C olum bia
U n iv e rsity , 1944).________________________________________________________
ZE
TR EA TM ENT WORDS II
s to rm quiet b a c te ria l
th e m se lv e s hippopotam us dispensation
ago bone single
prove m a rc h g ravitation
song hide shout
h o sp ita lity allow pure
sa la m a n d e r space durability
m om ent spoke debatable
m et detonation tria n g u la r
designation inch substitution
catch
m a la d ju stm e n t situation
c r itic is m yes cu ltiv ato r
blood clock p a rtic u la rly
g en eratio n
c o m p a trio t rule
w eak s a rc a s tic a lly shore
d eriv a tiv e la b o ra to ry m ism an ag em en t
d e c la ra tio n m a il cultivated
hom ogeneous m a tu re p ro clam atio n
re fo rm a tio n w hose h e rs e lf
confidential jum p p ecu liarity
w ise v io lin ist crow d
a fra id c e m e te ry heat
s e a t dinner cum ulative
contem ptible su c c e ssfu lly s ta r
sand sy ste m calculation
belong
technology m illim e te r
cancellatio n
c o m m e m o ra te
S elected fro m : E . L. T horndike and I. L o rg e, The T e a c h e r!s
W ord Book of 30,000 W ords (New York: T e a c h e rs C ollege, C olum bia
U n iv ersity , 1944)._________________________________________________________
67
P R E -P O S T WORDS III
lib e ra lis m capability
hunt equal
d ra m a tic a lly e x p re s s
p e rsu a siv e ly life
diligently b e a st
c a rn iv o ro u s decide
se n sa tio n a l bio g rap h er
height e n ter
m a n u fa ctu re b inoculars
b le ss hole
d re a m com m on
a rm y h e a rd
m e c h a n ism geographic
g e n e ro sity n atio n alism
c o n te m p o ra ry evening
S elected fro m : E . L. T horndike and I. L o rg e, The T eacher*s
W ord Book of 30,000 W ords (New York: T e a c h e rs C ollege, C olum bia
U n iv ersity , 1944).________________________________________________________
68
TR EA TM ENT WORDS III
d eco ratio n station adult
questionable stick v a rio u sly
coagulate te a r slow
dog peace s to re
s y m m e tric a l cauliflow er m u ltip licity
m e m b e r plan co u n terb alan ce
shine lite ra tu re p e rs p ira tio n
te m p o ra rily d eclam ation co st
d e sire easy sim ple
ch air p ercep tib le specification
c a r took dance
able stre n g th subdivision
p re d e c e s s o r su g ar spend
bell denom ination re sp e c tfu lly
office settle m ilita r is m
s a n ita riu m p le asa n t d estro y
although politively stab ilizatio n
n atu ra liz a tio n action s im ila rity
g rain lim itatio n p a rad o x ical
s ta tis tic a l te m p e ra m e n ta l oak
dry chief m isd e m e a n o r
su p e rfic ia l s p irit h e re d ity
pain c a fe te ria dom inican
crow n c rite rio n com bustible
m unicipal g eo g rap h e r nom ination
sim u ltan eo u sly bit offer
p olitician
S ele cted fro m : E. L . T horndike and I. L o rg e , The Teacher*8
W ord Book of 30,000 W ords (New York: T e a c h e rs C ollege, C olum bia
U n iv ersity , 1944).
A PPEND IX B
TA LLY OF STU TTERING DURING
PR A C TIC E PERIOD
69
70
FREQUENCY OF STU TTERING DURING PR A C TIC E
PERIO D FOR EACH OF TH REE SESSIONS
Subject S essio n I I I I I I I V V V I *
1 ( L i . A . ) 1
2
7
-
_
-
_**
2 ( B . C . )
J
1 3 2 0 3 1 3
2 3 2 3 3 4 3
3 4 4 4 1 5 3
3 ( L . D . ) 1 2 1 4 1 2
_
2 3 1 1 3 3 -
3 2 3 2 1 1 -
4 ( R . N . ) 1 2 0 1 0 1 0
2 0 0 0 0 1 0
3 0 0 2 1 1 1
5 ( R . R . ) 1 1 1 . 3 1
_ _
2
3
1 2 1 1 • -
6 ( V . S . ) 1 2 2 1 1
2 1 2 3 2 2 3
3 5 0 3 2 2 3
7 ( D . S . ) 1 1 1 1
_ _ _
2 2 1 2 2 5 -
3 2 I 1 I 3 -
8 ( J . W . ) 1 3 1 2 2
_
2 1 0 1 2
- -
3 2 2 1 2 3 -
9 ( R . B . W . ) 1 1 3 2
_ _ _
2 0 2 3 1 1 -
3 3 3 2 - - -
E ach block re p re s e n ts 5 w o rd s.
Indicates th at the fre q u e n c ie s w e re not tab u lated .
A PPEND IX C
INSTRUCTIONS TO THE JUDGES
71
INSTRUCTIONS TO TH E JUDGES
You a r e going to judge the u tte ra n c e s of s p e a k e rs who a re
c o n sid e re d s tu tt e r e r s , as they re a d individual w o rd s. D uring the
{original study, w ords w ere p ro je c te d onto a sc re e n and the ta s k of the
s p e a k e r w as to say th e w ord he saw . U nder one condition, the subject
jwas given feedback; th a t is , the reco rd in g of his u tte ra n c e w as replayec
i
im m e d ia te ly if he w as judged to have s tu tte re d . U nder an o th er e x p e ri
m e n ta l condition, he w as given feedback if the u tte ra n c e was judged to
I
fiave been spoken fluently. Now, your ta sk is to liste n to and judge
^ach u tte ra n c e and decide w hether you would have p re se n te d fluency
I
feedback b ecau se the w ord w as spoken fluently, o r p re se n te d stu tterin g
|
feedback because the w ord w as s tu tte re d . You a re to m a rk y our judg-
jment on the ta lly sh e e ts in fro n t of you.
A tone is p re s e n te d on the tape coincident with the p resen ta tio n
of th e w ord. T hus, you w ill know when the stim u lu s w ord w as p r e
sen ted re la tiv e to w hen the sp eak er began h is u tte ra n c e .
A re th e re any q u estio n s?
72
A PPE N D IX D
JUDGED FREQ U EN CY O F STUTTERING AND
FLU EN CY FO R T H R E E TREATM ENTS
OF AUDITORY FEEDBA CK
74
FREQ UENCY O F STU TTERIN G AND FLU EN CY
DURING P R E T R E A T M E N T AND PO STTREA TM EN T
PERIODS FO R NO FEED BA CK
1
Subjects
1
P r e tr e a tm e n t P o s ttre a tm e n t D ifference
F requency of S tuttering
1 12 11 -1
2 12 14 +2
3 16
9 -7
4 0 0 0
5 4 4 0
6 8 7 -1
7 13 10 -3
8 1 6 +5
9 14 3 -11
Total 80 64 -16
Mean 8.88 7. 11 1. 77
F requenc y of F luency
1 18 19 + 1
2 18 16 -2
3 14 21 +7
4 30 30 0
5 26 26 0
6 22 23 + 1
7 17 20 +3
8 29 24 -5
9 15 27 + 12
T otal 189
206 17
M ean 21 22.88 1.88
T 5
FREQUENCY OF STU TTERIN G AND FLUENCY
DURING PR E T R E A T M E N T AND PO STTREA TM EN T
PERIODS FO R FLU EN CY FEED BA CK
Subjects P r e tr e a tm e n t P o s ttre a tm e n t D ifference
F re q u e n c y of S tu tterin g
1 12 8 -4
2 13 17 +4
3 10 7 -3
4 0 0 0
5 14 8 -6
6 3 2 -1
7 5 2 -3
8
7
0 -2
9 6 10 +4
T otal 65 54 -11
Mean 7.22 6 -1 .2 2
!
F re q u e n c y of F luency
1 18 22 +4
2 17 13 -4
3 20 23 +3
4 30 30 0
5 16 22 +6
6 27 28 + 1
7 25 28 +3
8 28 30 +2
9 23 20 -3
T otal 204 216 +12
M ean 22.7 24 + 1.33
76
FREQUENCY OF STU TTERING AND FLUENCY
DURING PR E T R E A T M E N T AND PO STTREA TM EN T
PERIODS FOR STU TTERING FEEDBACK
J Subjects P r e tr e a tm e n t P o s ttre a tm e n t D ifference
1
F re q u e n c y of S tuttering
1 10 8 -2
2 13 18 +5
3 4 4 0
4 3 2 -1
5
9 6 -3
6 4 7 +3
7 7 4 -3
8 5 0 -5
9 11 11 0
T otal 66 60 -6
Mean
1
7. 33 6 .6 -. 66
j
F re q u e n c y of F luency
1
1
20 22 +2
2 17 12 -5
3 26 26 0
4 27 28 + 1
5 21 24 +3
6 26 23 -3
7
19 26 +7
8 25 30 +5
9 18 18 _______0
Total
199 209
+10
Mean 2.21 23.22 1. 11
APPEND IX E
D IFF E R E N C E SCORES IN U TTERA N CE DURATION
FOR TH R E E TREATM ENTS OF AUDITORY FEED BA CK
77
D IF FE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF
U TTER A N CE DURATION FOR SUBJECTS 1-9 DURING THE NO FEEDBA CK TREATM ENT
1 2 3
Subjects*
4 5 6 7 8
9
-179 +254 -143 - 4 - 4 0 + 3 + 36 + 71
+ 12 +323 + 1 + 6 + 2 + 2 - 5 + 2 + 6
+ 3 0 - 20 - 3 - 4 - 33 + 2 + 28 + 87
- 2 +225
+ 9 0 + 7 0 0 + 4 - 21
+ 3 +478 - 6 + 15 + 2 + 26
_
18 + 1 - 56
+ 1 + 11 + 32 + 2 - 12 - 3 - 61 + 2
-149
- 2 - 5 + 21 - 2 -
1
-
1
_
24 0 + 34
+ 6
+ 99 + 1 + 11 0 + 6 + 4 + 2 - 5
+ 1 +302
- 9 0 0 + 5 + 6 + 23 - 81
- 2 - 25 - 2 - 3 -
39
- 2 0 + 5 - 18
0 - 3 + 8 - 4 + 2 -
1 + 2 + 4 - 12
- 7 0 - 3 + 7 - 3 + 3 - 14 + 3 + 4
+ 3 - 68 + 22 + 1 - 7 + 1 + 1 + 68 - 2
+ 5
- 91 + 37 + 2 - 5 + 1
-
9 0 - 4
- 15 - 3 + 32 - 2 -
5 + 10
_
60
_
1 - 25
- 4 -282 + 29 - 1 - 5 - 4 + 9 0 + 2
-225 - 15 + 10 + 10
-
5 3
_
53
_
1 - 7
-122 +473 - 10 + 4 - 1 - 1 + 1 + 5 + 6
* P re tre a tm e n t duration w as su b tra c te d fro m p o sttre a tm e n t duratio n so th at a (-) in d icates a
reduction in duration and a (+) indicates an in c re a s e in duration.
— j
oo
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF
U TTERA N CE DURATION FOR SUBJECTS 1-9 DURING TH E NO FEEDBA CK T R E A T M E N T --C ontinued
1 2 3
Subjects*
4 5 6 7 8 9
- 4 - 8 + 6 - 1 + 2 - 81 0 0 - 3
-170 -124 - 16 + 9 - 2 - 4 - 3 0 - 55
+ 1 - 11 + 12 + 7 + 29 + 11 - 28 0 + 1
- 2 - 1 - 1 0 0 + 4 - 4 + 3 - 19
0 - 1 - 30 - 2 - 1 - 18 - 19 - 2 - 33
- 3 -146 - 13 - 3 - 2 - 4 + 13 + 5 + 1
+ 11 - 7 + 42 - 3 - 2 - 1 + 6 + 5 0
- 1 + 44 - 3 - 8 0 + 1 + 1 + 1 - 2
- 6 - 17 - 3 + 1 - 15 - 53 + 13 + 3
- 5 - 14 + 7 0 0 0 + 2 -105
- 8 + 8 0 + 3 0 - 3 + 5 + 96
T otal -692 +1687 + 26 + 44 - 48 - 94 -287 +218 -545
M ean -2 6 .6 1 +58.17 .90 1.47 -1 .6 0 -3 . 13 -9 .5 7 7.27 -1 8 .1 7
S. D. 64.66 258.17 26. 17 5.26 9.51 17.33 20.90 14. 19 45.43
$
P r e tr e a tm e n t d u ratio n w as su b tra c te d fro m p o s ttre a tm e n t duration so th a t a (-) in d icates a
reduction in duration and a (+) indicates an in c re a s e in duration.
v O
D IFF E R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS
OF U TTERANCE DURATION FOR SUBJECTS 1-9 DURING THE FLU EN CY FEED BA CK TREATM ENT
1 2 3
Subjects*
4 5 6 7 8
9
+ 19 - 32 0 20 - 7
+ 2 35 11 60
+ 3 +826 -
9
-
11 - 1 + 4 0 1 7
- 22
- 59 - 1 - 11 - 6 - 2 0 4 45
+ 3 +178 - 10 -
9 - 2 - 2
_
7 + 1 64
- 6 +138 + 3 -
2
-109 - 2
_
1 + 1 1
- 3 + 3 - 4 - 15 - 7 - 2
-
9 3 + 1
-108 + 2 - 4 -
5 + 3 - 2
_
17
_
11 5
+ 2 + 53 - 3 - 3 - 4 - 6 + 1 3 + 47
- 2
+ 69 -
9
-
13 -160 - 1 + 2 6 + 12
+ 11 +217 - 8 0 - 8 - 1 + 2 + 2 + 7
- 4 + 4 + 7 -
1 + 2
- 9 + 5 + 2 3
+ 3 + 8 - 4 -
9 + 79 - 4 + 1 + 1 + 62
0
- 39 + 1 - 1 + 1 - 3 0 + 6
9
+ 10 + 51 - 2 + 2 - 35 - 7 + 8 + 4 6
+ 10 - 1 + 5 + 7 + 3 - 5 0 + 2 4
+ 3 +243 + 6
-
9 + 15 - 5
_
38 1 1
- 1 +786 - 18 - 8 - 7 - 12
-
3 0 +
39
- 2 - 5 32 4 - 4 - 7
'
2 4 2
* P re tre a tm e n t duratio n w as su b tra c te d fro m p o sttre a tm e n t duration so th a t a (-) indicates a
reduction in duration and a (+) in d icates an in c re a s e in duration.
a
a
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STA N D A R B D EV IA TIO N SO F
U TTERANCE DURATION FOR SUBJECTS 1 -9 DURING THE FLU EN CY FEEDBACK TREATM ENT --C ontinued
1 2 3
Subjects*
4 5 6 7 8 9
- 1 +229 -117 0 - 2 + 22 + 2 + 4 - 8
- 1 - 1 - 58 - 5
- 9 + 2 - 7 - 7 + 65
+ 1 + 61 - 2
- 9 0 - 7 + 5 - 1 + 1
- 2 - 1 - 12 - 7 0 - 33 - 1 - 1 + 3
- 1 +148 - 8 0 + 1 - 1 - 3 0 + 1
- 11 + 4 - 2 - 2 - 50 0 + 1 - 1 - 2
- 1 + 32 + 6 + 2 + 1 - 3 + 4 0 + 1
- 5 + 1 - 16 - 8 - 1 - 8 + 1 - 1 +148
-217 +106 0 - 14 + 11 + 36 - 1 + 4 + 6
- 2
- 2
+ 3
+ 12
- 6
- 5
- 32
- 1
- 1
0
- 4
- 20
- 83
- 9
- 1
- 8
+ 35
+ 4
0
- 8
- 4
- 2
0
T otal -326 3030 -329 -160 -319 - 74 - 47 -141 176
Mean -1 1 .2 4 101 -1 0 .9 6 -5 .3 3 -1 1 .0 -2 .4 7 -1 .5 7 -4 .7 0 6.2 8
S. D. 45. 15 204.76 23.71 5.88 45.68 10.70 12.21 18. 17 39. 17
’ '‘P r e tr e a tm e n t duration w as su b tra c te d fro m p o s ttre a tm e n t duration
reduction in du ratio n and a (+) in d icates an in c re a s e in duration.
so th a t a (-) indicates a
o o
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF U TTERANCE
DURATION FO R SUBJECTS 1-9 DURING THE STUTTERING FEEDBA CK TREATM ENT
Subjects*
1 2 3 4 5 6 7 8
9
+ 1 -222 + 2 2 72
59 6 3 + 13
0 +244 - 4
-
2 0
-
16
-
22
_
5 -134
- 5 - 75 + 2 + 1 - 6 0 -
5
-
91 -329
+203 -169 - 2 + 3 - 2 - 51 + 10 - 6 + 44
0 - 84 0 - 2 + 10 - 14 -
9
- 8 + 6
- 2 - 16 - 1 + 4 + 3 + 3 + 2
-
25 - 49
+ 44
- 29 - 7 + 3 -
79 + 13 + 1
-
59 + 2
- 10 +342 - 24 + 2 - 2 + 2 + 1
-
1 + 7
0 - 2 + 1 + 1 - 15 - 1 - 8
-
1 - 3
0 - 1 + 1 + 8 0 0 + 8 0 + 50
- 5 +170 + 2 -
3
-
3 0 + 8 0 -138
-244 + 2 + 1 - 6 - 2 + 3
-
1 + 2 + 67
- 18 - 3 - 1 + 4 0 0
-
1 0 + 4
- 12 +161 -
1 - 2 + 2
-
2
-
1 + 1 0
- 1 +560 0 0
-
10 + 5 + 3 + 2 -137
- 3 + 1 + 37 + 2 -
6 + 1 + 20 + 1 - 1
- 3 +127 - 1 + 4 - 71 - 4 0 + 2 0
- 1 +347 0 + 1
-
3 0 + 1
_
1 + 6
* P re tre a tm e n t duratio n w as su b tra c te d fro m p o sttre a tm e n t duration so th at a (-) indicates a
reduction in duration and a (+) indicates an in c re a s e in duration.
00
tSJ
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF UTTERANCE
DURATION FO R SUBJECTS 1-9 DURING TH E STUTTERING FEED BA CK T R E A T M E N T --C ontinued
1 2 3
Subjects*
4 5 6 7 8 9
+ 3 - 1 - 1 0 - 23 - 4
- 9
.
3 + 32
- 4 +131 + 1 - 1 + 36 + 1 - 1 - 22
- 49
+ 3 - 62 + 3 + 10 - 2 - 3 - 2 + 2 + 36
- 6 - 47 0 - 2 - 6 - 23 - 3
-
1 + 14
- 3 - 70 - 9 - 2 - 88 - 7 + 1 + 1 + 17
+ 3 0 - 13 0 - 3 - 1 - 5 - 3 - 10
- 4 +306 - 2 0 - 4 - 1 - 4 + 1 0
- 2 0 - 3 + 2
+119 - 4 - 5 - 8 + 11
- 3 - 6 + 1 + 1 + 6 - 3 - 4 + 8
+ 9 0 + 1 + 1 + 28 - 5 - 38
+ 2 + 1 + 9 + 6 + 13 - 4
+ 96 + 3 + 4 + 2 + 3 + 13
Total - 69 1761 - 88 38 -202 -120 - 27 -225
-591
Mean -2 .65 58.70 -2 .9 3 1.26 -6 .7 3 -4 .0 -1 .0 4 -7 .50 -21.11
S . D. 61 .85 162.87 8 .29 3.49 35.52 6.22 7.52 19 .71 79.28
P r e tr e a tm e n t duration w as su b tra c te d fro m p o sttre a tm e n t duration so th at a (-) indicates a
reduction in duration and a (4-) indicates an in c re a s e in duration.
oo
u >
APPEND IX F
D IFF E R E N C E SCORES IN LATENCY O F RESPONSE
FOR TH R EE TREATM ENTS OF AUDITORY FEED BA CK
84
D IF FE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF
LATENCY OF RESPONSE FOR SUBJECTS 1-9 DURING THE NO FEED BACK TREATM ENT
Subjects*
1 2 3 4 5 6 7 8
9
+ 3 + 12 + 5 5 - 18
.
7 + 2 + 52 0
-113 - 17 - 5 - 4
- 19
-
9 + 15 + 1 -132
-182 - 3 + 2 - 3 - 36 + 3 + 1 + 34 + 28
- 68 - 4 - 4 - 3 - 22 0 + 5
+ 9 + 24
- 28 - 18 + 35 - 5 +201 + 4 + 15 + 1 + 19
- 12 - 5 - 4 - 2 + 11 - 5 + 1 - 4 + 89
+ 90 - 2 0 - 4 - 39
- 2 - 2 + 17 - .30
- 33 - 7 - 3 0 - 10 - 2 + 2 + 2 - 7
- 10 - 10 + 1 + 5 - 57 + 18 + 36 + 63 + 27
+ 37 - 11 - 2 + 1 - 27 +
9 + 1 + 1 + 16
- 11 - 9 + 1 - 1 - 12 + 4 -
6 + 27 + 22
-152 - 1 - 3 - 4 - 8 - 6 + 21 - 3 - 1
+ 4 - 5 - 11 - 1 - 23 + 10 + 37 + 56 + 15
- 4 - 9 + 24 - 5 - 14 - 17 + 12 + 11 - 32
+ 9 - 13 + 27 - 2
- 29
- 6 + 34 - 28 - 1
+ 27 - 10 - 7 - 1 - 30 -
19
-
3 + 15 + 27
+169 - 6 - 3 - 1 - 1 -
19 + 17 + 29 - 39
+ 57 - 5 - 7 0 - 26 - 7 + 4
+ 19 + 33
* P re tre a tm e n t latency w as su b tra c te d fro m p o s ttre a tm e n t latency so that a (-) indicates a
reduction in latency and a (+) indicates an in c re a s e in latency.
o o
ui
D IF FE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF
LATENCY OF RESPONSE FOR SUBJECTS 1-9 DURING TH E NO FEEDBACK T R E A T M E N T --C ontinued
1 2 3
Subjects*
4 5 6 7 8 9
-253 + 13 + 3 - 1 - 12 - 16 + 9 - 7 - 9
- 57 - 8 - 3 + 1 - 49 - 1 + 23 + 3 + 87
+ 35 - 39 - 1 - 2 - 13 + 4 - 37 + 96 + 10
+ 25 - 4 - 15 - 6 + 10 0 + 3 - 59 0
+ 53 - 3 - 23 - 1 - 62 + 3 + 12 + 63 - 40
- 1 - 16 + 48 + 1 - 6 - 18 + 28 + 60 + 1
- 70 - 39 + 5 + 1 + 25 - 7 - 7 + 34 - 4
+ 50 - 6 - 2 + 1 - 7 - 2 - 5 + 18 - 7
+ 51 - 2 - 3 - 2 + 8 - 4 + 49 - 97 + 1
- 46
- 6
- 33
- 10
+ 4
- 12
+ 8
+ 23
- 4
0
- 4
- 25
- 17
- 17
- 6 + 6
+ 1
+ 14
+112
+ 4
+ 17
+ 30
T otal -436 -276 + 74 - 51 -563 -140 178 529 144
Mean -1 5 .0 3 -9 .2 0 2.47 -1 .7 0 -1 8 .7 9 - 5 .0 5. 93 18.24 4.97
S. D. 82. 11 11.61 14.76 2.41 20.24 8.79 18.09 40.91 38.96
* P re tre a tm e n t latency w as su b tra c te d fro m p o s ttre a tm e n t latency
reduction in latency and a (+) indicates an in c re a s e in latency.
so th at a (-) indicates a
oo
O '
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF LATENCY
OF RESPONSE FOR SUBJECTS 1-9 DURING TH E FLU EN CY FEED BA CK TREATM ENT
1 2 3
Subjects*
4 5 6 7 8 9
- 81
- 12
+
+
3
2
- 21
34 +
0
10
- 12
- 4 +
0
6
+ 41
+ 13
+ 2
+ 46
-104
0
- 29 + 10 - 20 - 4 + 13 + 7 + 20 - 77 + 53
-232 + 5 + 5 + 3 - 12 + 8 + 6 + 16 + 7
- 62 + 12 - 8 - 4 + 34 + 2 + 4 + 4 - 32
-148 + 4 - 13 + 3 - 16 0 + 22
- 19 - 34
-145 - 4 - 17 + 2 + 81 - 2
+ 9 + 14 - 10
-125 - 3 + 6 - 1 + 13 - 1 - 25 + 15 - 46
+ 11 + 1 + 2 + 2 - 60 -
7 + 26 - , 7 - 2
-295 + 7 0 - 1 - 4 + 1 + 16 - 4 + 16
- 49 + 7 - 3 + 1 + 2 - 17 + 17 + 4 - 10
- 27 + 5 - 7 + 3 + 28 - 14 - 40 - 4 + 29
- 84 + 7 - 2 + 1 - 34 + 2 + 48 + 15 - 8
+ 75 + 7 - 6 - 6 - 6 - 6 + 7 + 63 - 23
- 10 - 1 - 7 + 2 + 26 + 3 + 17 - 21 - 46
- 52 + 78 + 12 + 3 - 6 - 4 - 44 - 2 - 3
- 21 4 3 3 + 11 5 + 3 - 4 - 40
* P re tre a tm e n t latency w as s u b tra c te d fro m p o sttre a tm e n t latency so th a t a (-) indicates a
reduction in latency and a (+) indicates an in c re a se in latency.
oo
-* J
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF LATENCY
OF RESPONSE FO R SUBJECTS 1-9 DURING THE FLU EN CY FEED BA CK T R E A T M E N T --C ontinued
1 2 3
Subjects
4
*
5 6 7 8 9
+ 4 + 3 - 6 4 + 6 - 1 + 3
.
7 - 73
- 75 + 1 - 32 - 3 + 51 - 4 -
13
-
20 - 50
+ 10 0 + 10 -
8
- 24 - 5 + 9 + 2 - 12
-255 + 57 - 5 + 1 + 25 - 1
-
9 + 16 + 18
-140 + 34 - 10 + 1 + 25 - 3 + 29
-
1 - 22
-147 + 7 + 2 + 1 + 5 - 12
-
62
-
3 - 4
- 7 + 3 - 5 -
7
- 13 + 3 0 + 22 - 9
- 92 + 1 - 14 0
-
11 + 8 + 2
-
2 - 5
-250 + 3 - 8
-
3 + 6 + 6 -
42 - 1 - 5
0 + 4 - 1 -
3
- 50 + 8
- 25
-185 + 7 - 17 - 3 - 23 0
+ 40 0 - 5
T otal -2 .6 7 0 299 -208 - 28 - 111 - 33 14 40 -448
M ean -9 2 .0 7 9. 97 -6 .9 3 -. 93 2.8 -1 . 14 .48 1. 44 -1 6 .5 9
S. D. 95. 11 18. 10 10. 10 3. 79 33.60 6. 18 25. 35 3. 62 31. 01
* P re tre a tm e n t latency w as s u b tra c te d fro m p o sttre a tm e n t latency
reduction in latency and a (+) indicates an in c re a se in latency.
so that a ( -) indicates a
00
0 0
D IFF E R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF LATENCY
OF RESPONSE FO R SUBJECTS 1-9 DURING THE STUTTERING FEED BA CK TREA TM EN T
1 2 3
Subjects*
4 5 6 7 8
9
+ 56 + 3 - 14 7 + 6 2 + 30 + 18 17
+ 32 + 13 + 21 - 2 - 4 + 3
_
4 - 2 32
+ 69 - 36 + 4 - 10 - 66 + 2 + 30 + 10 5
- 66 0
+ 9 - 1 + 4 + 1 + 13 + 11 + 4
- 94 - 11 + 3 - 3 - 13 + 3
_
5 + 7 5
+ 16 - 2 + 17 - 14 - 37 -
4 +
19 - 30 + 3
+ 42 + 6 + 8 - 3 - 33 + 24 + 14 0
_
17
+106 + 4 + 24 + 3 +112 + 11 + 12 - 1 26
+ 22 - 1 0 - 3 + 37 + 5 + 8 + 4 21
+ 89 - 4 0 - 5 + 33
-
4 + 7 + 26 + 84
+142 + 15 + 15 - 6
+ 9
-
4 + 12 0 + 55
+ 29 - 5 + 7 - 6 + 7
-
7 + 19 - 35 + 63
+160 - 1 + 2 - 14 - 6
-
5 + 7 + 44 + 71
- 54 + 7 + 21 - 2 + 32 -
1 +
9 + 11 + 2
- 37 + 7 0 - 3 - 8 + 8 + 11 + 4 4
+ 56 - 4 0 - 6 - 5 -
6
_
3 + 11 0
+131 + 3 + 15 - 2 - 41 + 4
-
2 0 92
+ 37 6 + 2 • 4 - 21 “ 12 - 3 + 86 + 40
* P re tre a tm e n t latency w as su b tra c te d fro m p o sttre a tm e n t latency so that a (-) Indicates a
reduction in latency and a (+) indicates an in c re a se in latency.
o o
v O
D IFFE R E N C E SCORES (IN M ILLIM ETERS), MEANS AND STANDARD DEVIATIONS OF LATENCY
OF RESPONSE FOR SU BJECTS 1-9 DURING TH E STUTTERING FEED BA CK T R E A T M E N T --C ontinued
1 2 3
Subjects
4 5 6 7 8 9
- 58 - 4 + 7 - 3 + 63 - 8 - 20 - 34 + 14
+ 50 + 7 - 7 - 6 + 114 - 5
-
33 + 18
-
14
-264 - 44 + 11 - 5 - 57 0
-
44 + 6
-
6
+105 - 3 + 15 - 7 - 30 - 7 -
4 + 33
-
19
+ 23 - 1 + 24 0 - 3 - 6 -
36
-
4
-
42
+ 28 + 2 - 11 - 3 - 95
- 6 + 2 + 5 + 15
+ 16 + 21 + 9 0 - 15
- 7 -
24 + 1
-
2
+319 - 5 - 11 - 6 - 31
- 7 - 11 + 12 + 6
+ 2 + 2 + 13
-
4 + 16
-
1 - 1
-
26
- 93 + 5 - 16 -
4 + 2 - 5 - 7
+ 11 + 18 -
9 - 15 - 8 + 9
+ 4 + 12 - 1 - 14
T otal 461 - 17 198 -136 - 80 - 44 6 204 221
Mean 16.48 -. 56 6. 60 -4 . 53 -2. 66 -1. 52 .23 6.97 8. 19
S. D. 93.21 12.54 7.4 9 3.69 44.42 7. 10 18. 72 22.38 38. 82
P r e tr e a tm e n t latency w as su b tra c te d fro m p o sttre a tm e n t latency so that a (-) indicates a
reduction in latency and a (+) indicates an in c re a se in latency.
1 06
R E F E R E N C E S
91
R E FE R E N C E S
Black, J. W. The effect of delayed sid e-to n e upon vocal ra te and
in ten sity . Jo u rn a l of Speech and H earin g D is o rd e rs , 1950,
J ^ , 56-60.
jBloodstein, O. N. H ypothetic conditions under w hich stu tte rin g is
red u ced o r absent. J o u rn a l of Speech and H earin g D is o rd e rs ,
j 1950, 15, 142-153.
jBrown, G. S. , & C am pbell, D. P. P rin c ip le s of s e rv o m e c h a n is m s .
New York: W iley, 1948. C ited by H. L ane and B. T ra n e l,
The L o m b ard sign and the ro le of h earin g in sp eech . J o u rn a l
of Speech and H earin g R e s e a rc h , 1971, 14, 677-709.
I
Brow n, S. R. T he loci of stu tte rin g s in the sp eech sequence. Jo u rn al
of Speech D is o r d e r s , 1945, 10, 181-192.
I
i
C h ase, R. A. An inform ation-flow m odel of the o rg an izatio n of m otor
i activity: I T ran sd u ctio n , tra n s m is s io n , and c e n tra l control
of se n so ry in fo rm atio n . Jo u rn a l of N ervous and M ental D is e a s e ,
1965, 140, 239-251. (a)
C h ase, R. A. An info rm atio n -flo w m odel of the o rg an iz atio n of m otor
activity: II Sam pling, c e n tra l p ro c e ssin g , and u tiliz atio n of
s e n so ry in fo rm atio n . J o u rn a l of N ervous and M ental D ise a se ,
1965, 140, 334-350. (b)
C h e rry , C. , & S a y e rs, B. E x p e rim e n ts upon the total inhibition of
s ta m m e rin g by e x te rn a l control and som e c lin ic a l re s u lts .
Jo u rn a l of P sy c h o so m a tic R e s e a r c h , 1956, _ 1 _ , 233-246.
^ o h e n , E. A c o m p a riso n of o ra l read in g and spontaneous sp e e c h with
j s p e c ia l re fe re n c e to the adaptation and co n siste n cy effect.
| S peech M o n o g ra m s, 1955, 2 0 , 144.
C u rle e , R. F . , & P e rk in s , W. H. The re lia b ility of ju d g m en ts of in
sta n c e s of stu tte rin g . P a p e r p resen te d to ASHA convention,
1970.
92
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------93“
C u rle e, R. F . , & P e rk in s , W. H. C o n v ersatio n al ra te control th e ra p y
fo r s tu tte rin g . J o u rn a l of Speech and H earing D is o r d e r s , 1969,
34, 245-250.
jEdw ards, A. L. E x p e rim e n ta l design in psychological r e s e a r c h . New
York: H olt, R in e h a rt and W inston, 1965.
I
I
F a irb a n k s , G. S y stem a tic r e s e a r c h in e x p e rim e n ta l phonetics: 1. A
| th e o ry of the sp e e c h m e c h a n ism as a s e rv o s y s te m . J o u rn a l of
Speech and H earin g D is o r d e r s , 1954, 133-139.
F a irb a n k s , G. Selective vocal effects of delayed au d ito ry feedback.
Jo u rn a l of S peech and H earin g D is o rd e rs , 1955, 2£, 333-346.
F a irb a n k s , G. , & G uttm an, N. E ffects of delayed aud ito ry feedback
upon a rtic u la tio n . Jo u rn a l of Speech and H earin g R e s e a r c h ,
1 9 5 8 ,_1, 12-22.
G oldiam ond, I. S tu tterin g and fluency as m anipulable o p e ra n t resp o n se
c la s s e s . In L. K ra s n e r and L. P. U llm an (Eds. ), R e s e a rc h in
beh av io r m o d ificatio n . New York: H olt, R in e h a rt and W inston,
1965. Pp. 106-156.
[G ruber, L. S en so ry feedback and stu tte rin g . Jo u rn a l of Speech and
| H earing D is o rd e rs , 1965, 30, 378-280.
!
(H ardy, W. G. On language d is o rd e rs in young children: A re o rg a n i-
I zation of thinking. Jo u rn a l of Speech and H earin g D is o r d e r s ,
! 1965, 30, 2-16.
H ays, W. L. S ta tistic s fo r p sy c h o lo g ists. New York: Holt, R in e h a rt
and W inston, 1963.
Je n se n , P. J. E ffects of re w a rd and punishm ent on stu tte rin g in
c h ild re n . P r o je c t No. 5-8377, C o o p erativ e R e s e a rc h P r o
g ra m of the Office of E ducation, 1966.
Johnson, W. S tu tterin g . In W. Johnson, S. B row n, J. C u rtis , C.
Edney, J. K e a s te r (Eds. ), Speech handicapped school c h ild re n .
New York: H a rp e r and B ro th e rs , 1956. Pp. 202-300.
L ane, H. , & T ra n e l, B. The L o m b ard sign and the ro le of h e a rin g in
sp eech . J o u rn a l of Speech and H earin g R e s e a r c h , 1971, 14,
677-709.
94
L ee, B. S. A rtific ia l s tu tte r. J o u rn a l of Speech and H earin g D is-
o r d e r s , 1951, _16, 53-55.
jLove, L. R. , & J e f f r e s s , L. A. Identification of b rie f pau ses in the
fluent sp eech of s tu tte r e r s and non s tu tte r e r s . Jo u rn a l of
Speech and H e a rin g R e s e a r c h , 1971, 14, 229-240.
i
M a ra is t, J. A. , St Hutton, C. E ffects of aud ito ry m asking upon the
| sp eech of s tu tte r e r s . J o u rn a l of Speech and H earing D is o r d e r s ,
1957, 22, 385-389.
M endenhall, W. Introduction to p robability and s ta t is tic s . B elm ont,
C alifo rn ia: Woods w orth P ublish in g Co. , 1967.
M a rtin , R. , 8t Siegel, G. M. The effects of resp o n se contingent shock
on stu tte rin g . Jo u rn a l of Speech and H earin g R e s e a rc h , 1966,
I 9, 340-352. (a)
M a rtin , R. , St Siegel, G. M. The effects of sim ultaneously punishing
stu tte rin g and re w a rd in g fluency. J o u rn a l of Speech and H e a r
ing R e s e a rc h , 1966, 9^, 466-475. (b)
M ysak, E. D. Servo theory and s tu tte rin g . Jo u rn a l of Speech and
| H earin g D is o r d e r s , i960, 25, 188-195.
i
M ysak, E. D. A serv o m odel fo r sp eech th e ra p y . J o u rn a l of Speech
j and H earin g D is o r d e r s , 1959, 2 4 , 144-149.
!
iN eelley, J. A study of sp eech b eh av io r of s tu tte r e r s and n o n
s tu tte r e r s under n o rm a l and delayed au d ito ry feedback. Jo u rn al
of Speech and H earin g D is o r d e r s , M onograph S upplem ent, 1961,
7, 63-82.
S h am e s, G. H. , & S h e rric k , E. C. A d isc u s sio n of non fluency and
s tu tte rin g as o p e ra n t b eh av io r. In D. A. B a r b a r a (Ed. ), New
d ire c tio n s in stu tterin g : T h eo ry and p ra c tic e . S pringfield,
Illinois: C h a rle s C. T h o m as, 1965.
I Shane, M. L. S. E ffect on s tu tte rin g of a lte ra tio n in auditory fe e d
back. In W. Johnson (Ed. ), S tu tterin g in ch ild ren and a d u lts .
M inneapolis: U n iv ersity of M in n eso ta P r e s s , 1955, pp.
286-297.
93-
S h e a re r, W. M. C y b ern etics in the tre a tm e n t of voice d is o rd e rs .
J o u rn a l of Speech and H earin g D is o r d e r s , 1959, 24, 280-282.
Siegel, G. M. P u n ish m en t, s tu tte rin g , and disfluency. Jo u rn a l of
Speech and H earin g R e s e a r c h , 1970, 13, 677-714.
Sheehan, J. G. T heory and tre a tm e n t of stu tte rin g as an a p p ro a c h -
avoidance conflict. Jo u rn a l of P sy ch o lo g y , 1953, 27-49.
Sheehan, J . G. C onflict th eo ry of stu tte rin g . In J. E ise n se n (Ed. ),
S tuttering: A sym posium . New York: H a rp e r and B ro th e rs ,
1958.
JSkinner, B. F . V erbal b e h a v io r. New York: A pp leto n -C en tu ry -
| C ro fts , 1957.
S o d e rb e rg , G. A study of the effects of delayed auditory sid e-to n e on
! fo u r asp e c ts of s t u t t e r e r s ' sp eech during o ra l reading and
spontaneous speaking. U npublished d o c to ra l d is se rta tio n ,
Ohio State U n iv ersity , 1959.
(Soderberg, G. The re la tio n s of s tu tte rin g to w ord length and w ord
frequency. J o u rn a l of Speech and H earing R e s e a r c h , 1966,
9, 584-589.
S o d e rb e rg , G. D elayed auditory feedback and stu tte rin g . J o u rn a l of
Speech and H earing D is o r d e r s , 1968, 33, 260-267.
T horndike, E. L. , & L o rg e, I. The te a c h e r 's w ord book of 30,000
w o rd s . New York: T e a c h e rs C ollege, C olum bia U n iv ersity ,
1944.
Van R ip e r, C. Speech co rre c tio n : P rin c ip le s and m e th o d s. E n g le
wood C liffs, New J e rs e y : P re n tic e -H a ll, 1954.
Van R ip e r, C. The p re p a ra to ry s e t in stu tte rin g . Jo u rn a l of Speech
D is o r d e r s , 1937, 2 , 149-154.
|W ien er, N. C y b e rn e tic s. New York: W iley, 1948.
W iener, N. The hum an use of hum an b e in g s. Boston: Houghton M iff
lin, 1950.
w .
W illiam s, D. E. , W ark, M. , & M inifie, F . R atings of stu tte rin g by
audio, v isu al, and au d io -v isu al cues. Jo u rn a l of Speech and
H earin g R e s e a r c h , 1963, 6^ 91-100.
iYates, A. J. The re la tio n s h ip betw een th eo ry and th e ra p y in the
J clin ical tre a tm e n t of s tu tte rin g . I n B. B. G ray and G. England
| (E d s .), S tu tterin g and the conditioning th e ra p ie s . M onterey,
I C alifornia: M onterey Institute for Speech and H earin g , 1969.
Zipf, G. K. The psycho-biology of lan g u ag e. Boston: Houghton M iff
lin, 1935.

Asset Metadata

Creator Jensen, Flora Michelle Wark (author)

Core Title An Experimental Study Of The Effects Of Fluency Feedback And Stuttering Feedback On The Subsequent Frequency Of Stuttering Utterance Duration, And Latency Of Response

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Communicative Disorders

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag health sciences, speech pathology,OAI-PMH Harvest

Language English

Advisor Haney, Russell (committee chair), Slucki, Henry (committee member), Wood, Nancy E. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-788613

Unique identifier UC11363458

Identifier 7309314.pdf (filename),usctheses-c18-788613 (legacy record id)

Legacy Identifier 7309314

Dmrecord 788613

Document Type Dissertation

Rights Jensen, Flora Michelle Wark

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA