Psrceprual and Motor Skills, 1981, 5 3 , 187-194.

@ Perceptual and Motor Skills 1981

AWARENESS TRAINING AND REGULATED-BREATHING

METHOD IN MODIFICATION OF STUTTERING1

ROBERT LADOUCEUR LBONCE BOUDREAU

unive~sitkLaulrl U n i v m b t de Moncton
SYLVIE T G B E R G E
Universitk Laud

Summary.-Awareness has been shown to be an important variable in vari-

ous rypes of learning in humans. Its role in the modification of different clin-
ical disorders is promising. It is hypothesized that systematic awareness uain-
ing prior t o the introduction of regulated-breathing method would significantly
improve fluency in stutterers. 16 stutterers (mean age 25.1 yr.) were randomly
assigned to one of two groups, awareness training plus regulated-breathing
method or regulated-breathing method only. All clients were seen individually
during two 90-min. sessions and were informed that their speech was being
recorded. The percentage of stuttering and the rate of speech were analyzed.
Awareness training significantly reduced stuttering compared to a control pro-
cedure. But the most significant improvement appeared after the introduc-
tion of the regulated-breathing method. At a 1-mo. follow-up, although the
frequency of stuttering was significantly less than during baseline, the level
of disfluency was around 5%. From a clinical perspective, such results are
far from satisfactory and no further follow-ups were conducted. The theoretical
and practical implications of these results are discussed to improve the ef-
ficacy of our therapeutic methods used to counteract stuttering.

Recent research has shown that awareness is of paramount importance in

different types of learning in humans (see Brewer, 1974).2 Furthermore, the
modification of various clinical disorders is significantly improved when the
patient's attention is focused on his maladaptive habit. For example, nervous
habits (Ladouceur, 1979), obesity (Plimer & Iuppa, 1978), tics (Ladouceur &
Hillel, 1974), and pain tolerance (Ladouceur & Carrier, 1979) were signifi-
cantly improved when the therapeutic intervention was mainly aimed at in-
creasing a subject's attention to his performance. Such data support the recent
trend in behavior therapy towards emphasizing the role of cognitions in the
maintenance and in the modification of clinical disorders (Wilson & O'Leary,
1980).
Stuttering has been considered to be particularly resistant to therapy. This
problem becomes more serious when one considers its high incidence of occur-
rence. Four percent of the general population have been stutterers at some

'Requests for reprints should be sent to Robert Ladouceur, Ecole de Psychologie, Uni-
versitk Laval, Quebec, Canada G l K 7P4.
'R. Ladouceur, & P. Mercier, Awareness: a neglected factor in behavior therapy. (Manu-
script submitted for publication, 1980)
188 R. LADOUCEUR. ET AL.

point in their lives, with one percent continuing to stutter during adulthood
(Lanyon, 1978). Within the realm of behavior therapy, much effort has been
devoted to the understanding and the modification of this debilitating prob-
lem. A variety of treatment procedures have been developed, including such
methds as delayed auditory feedback, rhythmic speech, biofeedback, masking,
and shadowing. But unfortunately, the results have been quite modest. In ex-
tensive reviews of the literature, Ingham and Andrews (1973) and Lanyon
(1978) concluded that few procedures demonstrated significant clinical re-
sults. When positive outcomes were obtained, the maintenance and generaliza-
tion of the therapeutic benefits were less than satisfactory.
Among the published studies on the treatment of stuttering, the most strik-
ing and spectacular success is that reported by Azrin and his colleagues (Azrin
81 NUM, 1974; Azrin, Nunn, & Frantz, 1979), using an original method based
on regulated-breathing training. Thick package treatment involves several ele-
ments, such as awareness training, anticipation of stuttering, relaxation train-
ing, learning incompatible activities, corrective and preventive learning, coven
rehearsal, etc. After a single 2-hr. session, a decrease of 94% in stuttered words
is reported, this decrease reaching 98% at a 4-mo. fodow-up. But unfortu-
nately,,both of Azrin's studies suffer from serious methodological weaknesses.
No operational definition of stuttering is provided, and more importantly all
measures of stuttering are based on self-report. Behavioral and objective meas-
ures are of paramount importance in evaluating the efficacy of therapeutic in-
terventions (Nay, 1979; Nelson & Hayes, 1979). Replications in our labora-
tory of regulated-breathing method experiments, with more methodological rigor
have consistently provided statistically significant reductions in stuttering (Bou-
dreau, Jeffrey, Leblanc, & Dupuis, 1976; C8ti & Ladouceur, 1980; Ladouceur,
C6t.6, Leblond, & Bouchard, in press; Ladouceur & Boudreau, 1980), but from
a clinical standpoint the results are much more modest than the ones reported
by Azrin and his colleagues (Azrin & Nunn, 1974; Azrin, et al., 1979). After
treatment and follow-up, the percentage of stuttering was generally around
5%. Such high frequency of disfluent words questions the clinical utility of
this procedure, especially when the normal level is operationally defined as or
below 3% disfluent words (see Webster, 1979).
A perusal of the literature on this clinical problem indicated that in many
instances, providing immediate feedback to the stutterer concerning his dis-
fluencies has brought interesting results (Berecz, 1973; Cooper, Candy, & Rub-
bins, 1970; La Croix, 1973; Shaw & Shrum, 1972; Siege1 & Martin, 1968; Win-
gate, 1959). These data show that, when the patient becomes aware of his
stuttering, its frequency tends to decrease. But, from a clinical perspective, the
maintenance and the generalization of this procedure alone was noc satisfactory.
It may well be that such a specific and limited intervention is not p e ~se suf-
ficient to modify stuttering adequately, but its addition to a more complete and
 MODIFICATION OF STUTTERING 189

articulated treatment package may produce significant results. It is hypothe-

sized that the regulated-breathing method, coupled with extended awareness
training, could provide a powerful tool to reduce stuttering.
METHOD
Subjects
The clients were 16 stutterers ( 3 females and 1 3 males) who responded to
a radio advertisement offering treatment for speech disfluencies. Their ages
varied from 15 to 47 yr. (average 25.1 yr.). All reported stuttering from
childhood and seven of them had previously received therapy without success.
According to the definition proposed by Young ( 1961), stuttering included
four kinds of disfluencies: ( a ) hesitation before completing a word or a syl-
lable, ( b ) prolongation of a syllable, ( c ) repetition of a word or part of a
word, ( d ) blocking.
Procedure
The clients were randomly assigned to one of two groups. One female
graduate student in clinical psychology who had received intensive training
with this method served as therapist. All clients were seen individually during
one 90-min. evaluation session and one 2-hr. treatment session. In the group
- -

given awareness training plus regulated-breathing method (two females, six

males), the first session was devoted to a short behavioral analysis. Clients
were informed that their speech was being recorded. These data served to es-
tablish the pre-treatment level of disfluency. Subjects were then taught to be-
come aware of their stuttering by pressing a golf counter for each stuttered
word or syllable. The therapist provided an operational definition of stuttering
(Young, 1961) and gave an example of each category. She also monitored
stuttering in the same manner as did the subject. This phase mas continued
until the client accurately perceived and identified 80% of the stuttered words.
Following this awareness mining phase, the behavioral analysis was resumed
and a second recording was made to evaluate the frequency of stuttering. At
the beginning of the second session, another assessment was performed, which
was followed by the administration of the regulated-breathing technique. This
method outlined by Azrin and Nunn (1974) includes inconvenience review,
training of awareness and relaxation, learning of incompatible activities, cor-
rective and preventive training, symbolic rehearsal, and positive practice. This
session was terminated by again recording a 10-min. conversation with the pa-
tient, which served as posttest measure. The group with only regulated-breath-
ing (one female and seven males) followed the same procedure as described for
the first group, except that they were not given awareness training. The length
of this session was yoked for each in the second group with a corresponding
subject of the first group. Measures were again taken at a 1-mo. follow-wp
during a 30-min. conversation.
190 R. LADOUCEUR, ET AL.

During the five assessment periods, the rate of speech (total number of
syllables emitted per minute) and the percentage of stuttering (number of
stuttered syllables/total number of stuttered syllables X 100) were calculated
by one observer who was not aware of the purpose of the research and the
phases of the recording. A second observer independently assessed 30% of
the total recording in order to establish an inter-judge coefficient of reliability.
RESULTS
Reliability checks indicated a high percentage of agreement between in-
dependent assessors. The mean percentages of agreement for stuttered syllables
and total syllables were 87.5% and 94.7%, respectively. Also, the stuttered
words identified by the therapist were also evaluated. The independent as-
sessor confirmed that 82.2% of stuttered syllables were correctly identified.
The Mann-Whitney test performed on baseline data indicated no signifi-
cant differences among groups on the percentage of stuttered syllables (z =
.94, p > .05). When this test was applied to the data of the second phase, the
awareness-trained group reduced frequency of stuttering significantly more
than the second group ( z = -2.0, p < .05). At no other phase were such
comparisons significant ( p < .05). As depicted in Fig. 1, both groups fol-
lowed a similar trend except for the post-awareness-training period.
Friedman analyses of variance (Siegel, 1956) were conducted to deter-
- ~*oreneroand regulated
b r e a m l halnlng.
~
I
l2i ItaguMedbrWlhlnghalnlng.
&......a

FIG. 1. Percentage of stuaered syllables for the group given in Session 1 aware-
ness training plus the regulated-breathing method and for the group given regulated-
breathing only during each phase of the study
 MODIFICATION OF STUTTERING 191

mine the changes in the percentage of stuttering across the various phases of
the experiment. Significant differences were obtained for Group 1 ( ~ = 2
25.3,df= 4 , p < .001) andGroup 2 (x," 14,df = 4 , p < .05); seeFig.
1. Wilcoxon matched-pairs signed-ranks test further ~ieldedsignificant dif-
ferences for both groups from baseline data to posttest ( p < .05) and follow-
up periods ( p < .05). Interestingly, no significant changes were observed
between the latter two phases, which implies that the positive butcomes ob-
tained at posttest were maintained at follow-up. Concerning the rate of speech,
all comparisons among the various phases of the experiment were statistically
nonsignificant ( p > . I ) .
DISCUSSION
The differential results obtained by each group after the second phase in-
dicate the i m p o r a c e of awareness training in the modification of stuttering.
Indeed, when stuttering is measured after this training but prior to the introduc-
tion of the regulated-breathing method, we observe a significant reduction in
the percentage of stuttered syllables for subjects trained to identify each stut-
tering episode, relative to that for subjects submitted to a baseline condition.
But within-group comparisons do not indicate statistically significant differ-
ences from baseline to awareness-training phase for Group 1. Such data par-
tially confirm the first aspect of our hypothesis, which stated that systematic
awareness training would bring a significant reduction in stuttering. Various
explanations could account for such results. First, the length and the intensity
of our awareness-training procedure may explain the results. Our subjecrs were
trained to identify skillfully 80% of their stuttering during one 5-min. period.
Bearing this fact in mind, it is quite surprising that such a short training period
produces a statistically significant reduction in stuttering, a clinical disorder
particularly known to be resistant to any form of therapy. Second, it may
prove to be important to train stutterers to become aware not only of the fre-
quency of their stuttering, but also of the different types of stuttering, such as
interjection of sounds, syllables or words, repetitions, prolonged sounds, tension
pauses, etc. (see Williams, Darley, & Spriestersbach, 1978). Finally, in addi-
tion to learning to identify each stuttered episode, the stutterer may reduce this
problem if he becomes aware of other behavior closely associated with stutter-
ing, such a s muscular tension in the masseter muscle or irregularity in the
respiratory cycle. Such a position is very close to chat recently put forward by
Lanyon ( 1978).
The present data further support the efficacy of the regulated-breathing
method. But the results are far from being as satisfactory as the ones reported
by Azrin and his colleagues (Azrin & Nunn, 1974; Azrin, et al., 1979). The
latter reported a 98% reduction after one or two sessions. In their second
study a 97% reduction was maintained at a 3-mo. follow-up. In the present
192 R. LADOUCEUR, ET AL.

experiment, the percentage of stuttering after treatment was still around 4%

and even greater at a 1-mo. follow-up (see Fig. 1 ) . The reduction in stutter-
ing from baseline to the latter two periods was about 50%. The discrepanq
between Azr&'s results and the present ones is large. This absence of a clin-
ically significant reduction of stuttering becomes more puzzling if we consider
that client's expectancy for improvement was probably high, that dam were
gathered by the original therapist and in the original situation in which the
stutterer could have been adapted. Our earlier studies of regulated-breathing
procedure did not replicate Azrin's spectacular results (Boudreau, et d l . , 1976;
C8t6 & Ladouceur, 1980; Ladouceur & Boudreau, 1980). The main difference
between the two groups of researchers concerns the procedures of assessment of
stuttering. Azrin and his colleagues relied only on self-evaluation methods.
Clients carried recording sheets with them and monitored each stuttered word
as soon as possible. They totaled the frequency at the end of the day. Such an
assessment procedure is far from adequate. The results obtained by our sub-
jects when they were asked to monitor each stuttered episode before awareness
training showed that the average response was around 30%. Their perform-
ance was improved to 80% after systematic training. Such data question the
validity of self-evaluation as the unique means of therapeutic outcome assess-
ment of stuttering.
On the other hand, although the regulated-breathing method produced a
statistically significant decrease in stuttering ( a change from about 9% to 4 %
from baseline to posttest), from a clinical perspective, such a level of stuttering
is not satisfactory. At the 1-mo. follow-up, the frequency of stuttering was not
clinically improved. A wend towards an increase of stuttering was even ob-
served (see Fig. 1). Although the importance of long-term follow-ups (6-mo.
or more) have been recently discussed (Ladouceur & Auger, 1980), the a b
sence of clinical improvement reported 1-mo. after the end of treatment led to
no further follow-ups. This level of stuttering is quite high. In such studies,
the normal range is ". . . disfluent word counts were at or below 3% and any
disfluencies which occurred tended to be relatively effortless and short in dura-
tion" (Webster, 1979, p. 234). A study recently conducted in our laboratories
confirmed that normal speakers emitted an average of 3% of stuttered syllables
either in a one to one conversation or during the presentation of a short speech
in front of a small a~dience.~As recently discussed by Kazdin (1977), it is
one thing to obtain a statistically significant difference but quite another to
produce a clinically acceptable change. Unfortunately, the emphasis put on
awareness training did not increase the efficacy of the regulated-breathing
method to an acceptable clinical level. Also, such modest results become more
salient in the absence of a central group to determine the expected adaptation
'R. Ladouceur, & L. St-Laurent-Simard. Normative data in the evaluation of sruttering
therapy. (Unpublished manuscript, UniversitC Laval, QuCbec, Canada)
 MODIFICATION OF STUTTERING 193

effects to the therapist and to the assessment procedures. The elements com-
bined with the regulated-breathing method did not even provide satisfactory
outcomes.
In conclusion, at this stage, we may consider Azrin and Num's method
as a promising technique in the modification of sturtering. But further work
should be carried out before claiming that the method can eliminate stuttering.
Taking into consideration the positive trend in the results obtained in this re-
search with brief awareness training, further work should be conducted in using
various forms of awareness training over a more extended period. Such work
'is being conducted in our laboratory.

REFERENCES
AZRIN,N., & NUNN,R. C. A rapid method of eliminating stuttering by a regulated-
breathing method. Behavior Research and Therapy, 1974. 12, 279-286.
AZRIN,N. H., NUNN,C. R.;& FRANTZ,S. E. Comparison of regulated-breathing versus
abbreviated desensitization on reported stuttering episodes. Journal o f Speech
and Hearing Disorders, 1979, 44, 331-339.
BBREU,J. M. The treatment of stuttering through precision punishment and cognitive
arousal. Journal o f Speech and Hearing Disorders. 1973, 38, 256-267.
BOUDREAU, L., JEFFREY.C., LEBLANC, R., & DUPUIS,W. Le Egaiement trait6 par
le renversement d'habitudes. Revue de Modification d u Compo~rement,1976, 6 ,
34-50.
BREWER, W. F. There is no convincing evidence for operant and classical conditioning
in humans. In W. B. Weiner & D. S. Palermo (Eds.). Cognition and symbolic
processes. Hillsdale, NJ: Lawrence Erlbaum, 1974. Pp. 1-42.
COOPER, E. B., CADY,B. B., & ROBBINS, C. J. The effect of the verbal stimulus words
wrong, right, and free on the disfluency rates of stutterers and nonstutterers.
Journa~of Speech and Hearing Research, 1970, 13, 239-244.
C~TB,C., & LADOUCEUR, R. Effects of social aids on regulated-breathing treatment
for stutterers. Paper presented at the World Congress on Behavior Therapy,
Jerusalem, 1980.
INGHAM, R. J., & ANDREWS, G. Behavior therapy and stuttering: review. Journd of
Speech and Hearing Disorders, 1973, 6, 405-441.
KAZDIN, A. E. Assessing the clinical or applied importance of behavior change through
social validation. Behavior Modification, 1977, 1, 427-452.
LA CROIX,2. E. Management of disfluent speech through self-recording procedures.
]ousnd of Speech and Hearing Disorders, 1973, 38, 272-274.
LADOUCEUR, R. Habit reversal treatment: learning an incompatible response or in-
creasing the subject's awareness? Behavior Research and Therapy, 1979, 17, 313-
316.
LADOUCEUR, R., & AUGER,J. Where have all the follow-ups gone? The Behavior
- . 1980.. 3.. 10-11.
TheraDist.
LAD~UCEUR, R., & BOUDREAU, L. Effects of social support on the measurement and
treatment of stuttering by the use of regulated-breathing. Paper presented at the
World Congress on Behavior Therapy, Jerusalem, 1980.
LADOUCEUR, R., & CARRIER, C. Awareness and pain control. Paper presented at the
Association for the Advaocement of Behavior Therapy, San Francisco, 1979.
LADOUCEUR, R., C ~ T B ,C., LEBLOND, G., & BOUCHARD, L. Evaluation of regulated-
breathing method and awareness training in the treatment of stuttering. Journal
of Speech and Hearing Disorders, in press.
LADOU~EUR, R., & HILLBL,J. Traitement d'un tiqueur par therapie Mhaviorale. Luval
Medical, 1971, 42, 1033-1040.
194 R. LADOUCEUR, ET AL.

LANYON,R. I. Behavioral approaches to stuttering. In M. Hersen, R. M. Eisler, &

P. M. Miller (Eds.), Progress i n behavior modification. Vol. 6. New York:
Academic Press, 1978. Pp. 47-83.
NAY, R. W. Multimethod clinical arsessment. New York: Gatdner Press, 1979.
NELSON,R. O., & HAYES,S. C. Some current dimensions of behavioral assessment.
Behavioral Assessment, 1979, 1, 1-16.
PLIMER, P., & IUPPA,G. Effects of increasing awareness on food consumption in obese
and normal weight subjects. Addictive Behaviors, 1978. 3, 19-24.
SHAW,C. K., & SHRUM, W. F. The effects of response-contingent reward on the con-
nected speech of children who stutter. Journal o f Speech and Hearing Disorders,
1972, 37, 75-88.
SIEGPL,G. M., & -TIN, R. R. The effects of verbal stimuli on disfluencies during
spontaneous speech. Journal o f Speech and Hearing Disorders, 1968, 11, 358-
364.
SIEGEL,
S. Nonparametric statistics for the behavioral sciences. New York: McGraw-
Hill. 1956.
WEBSTER. R. L. Empirical considerations regarding stuttering therapy. In H. H.
Gregory (Ed.), Controversies about stuttering therapy. Baltimore: Univer. Park
Press, 1979. Pp. 209-239.
WILLIAMS, D. E., DARLEY, F. L., & SPRIESTERSBACH, D. C. In F. L. Darley & D. C.
Spriestersbach (Eds.), Diagnostic methods in speech pathology. New York:
Harper & Row, 1978. Pp. ?-??
WILSON,G. T., & O'LEARY,K. D. Principles o f behavior therapy. Englewood Cliffs,
NJ: Prentice-Hall. 1980.
WINGAG, M. E. calling attention to stuttering. Journal o f Speech and Hearing Re-
search, 1979, 2, 326-335.
YOUNG.M. A. Predicting ratings of severity of stuttering. Journal o f Speech and Hear-
ing Disorders, 1961, 7, 31-54.

~ c c e p i e dJune I S , 1981.