u .‘..‘:.\W|‘l ‘l
,.......x;._. a .,
ix: 23%.”.3

flan. “Ea

r

3*“:

. a,
a“.

V :.
7.1.1.:

A.»

 

 

 

 

 

.1:
‘xl‘. 1...,

0.32.1“:

 

'13..

 

 

 

 

figfiafinfiw 1 a».

 

 

 

f 0;}? LTBfi—A RY
" " Mich? State
UniverSIty

 

 

 

This is to certify that the
thesis entitled

The Effectiveness of Melodic Intonation Therapy and
Therapeutic Singing on Functional Communication Skills
for Adults with Expressive Aphasia

presented by

Yea Ju Rhee

has been accepted towards fulfillment
of the requirements for the

Master of Music degree in Music Therapy

 

 

/'l

F4

 

Ola'jor’Professor’s Signature

Date

MSU is an Affinnative Action/Equal Opportunity Employer

 

 

‘U-O-I-Une-a-g-

- —.-o-I---o---o-q-—o-a-c-e-I-----o-o->-.—-—

II--— —i-—--—--A----—---—~—.o--

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Proj/Acc&Pres/ClRC/Date0ue indd

TITLE: THE EFFECTIVENESS OF MELODIC INTONATION THERAPY AND
THERAPEUTIC SINGING ON FUNCTIONAL COMMUNICATION SKILLS
FOR ADULTS WITH EXPRESSIVE APHASIA

By

Yea Ju Rhee

A THESIS

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of

MASTER OF MUSIC
Music Therapy

2009

ABSTRACT
By
Yea Ju Rhee

The study of music and speech for persons with expressive aphasia by Sparks and
Holland attracted a fair amount of critical attention in the 19705. The effectiveness of
Therapeutic Singing (TS) and Melodic Intonation Therapy (MIT) in improving the
speech production of persons with expressive aphasia has been established in subsequent
research since that time. Based on the present research, the reframing study of Modified
Melodic Intonation Therapy (Baker, 2000), which is the systematic combination of
Melodic Intonation Therapy and Therapeutic Singing, gives convincing answers to some
of this study's research questions regarding whether or not the addition of TS to MIT
helps persons with aphasia improve their speech production. Two case studies were
conducted to investigate the effectiveness of MIT and TS to improve functional
communication skills. Eight variables of spoken language were compared: total length of
samples, total number of utterances, total number of words, average length of utterances,
average number of words per minute, total number of items named, total number of
prompts per item named, and scores for MIT performance. The results showed that MIT
and TS interventions helped the subject who had low functioning speech to improve
functional communication skills in both the long term and short term. On the other hand,
the subject who had high functioning speech skills showed positive numerical gains on
some of the variables in functional communication, but there were not statistically

supported differences between pre and post-tests in the short-term.

Copyright by
Yea J u Rhee

2009

ACKNOWLEDGEMENTS

The author gratefully acknowledges all those who helped in the realization of this
study - to participants and their wives for delightful cooperation, their enthusiasm, and
for making this study existed. I would like to thank committee members, Professor
Frederic Tims, for his grateful feedback, and Professor Hwan Jung for all his guidance in
the statistical analysis of my data to help this study show reasonable data sets. A special
thanks to my consultant, Professor Katie Strong, for providing expert advice in the
speech and language pathology fields and the help in recruiting participants in the
Aphasia Support Group around Lansing. I am forever grateful to my supervisor, Roger
Smeltekop, for his immeasurable support throughout my six years as a music therapy
student and for his ongoing lessons as the role model in my life. Finally, very special
thanks to my husband for all his support through love and patience to finish my music
therapy degree, and to my parents for the special care of my twins with uncountable

grace, and to my son and twins for waiting for mommy’s time.

iv

TABLE OF CONTENTS

 

 

 

 

 

 

LIST OF TABLES ........................................................................................................... vi
LIST OF FIGURES ....................................................................................................... viii
CHAPTER I 1
INTRODUCTION ....................................................................................................................... 1
STATEMENT OF PURPOSE ............................................................................................................. 4
CHAPTER II 5
LITERATURE REVIEW ............................................................................................................ 5
CHAPTER III 18
METHOD .................................................................................................................................. 18
Subjects .................................................................................................................. 1 8
Setting and Apparatus ............................................................................................ 22
Research Design .................................................................................................... 23
Procedure ............................................................................................................... 24
Measurement and Analysis .................................................................................... 40
CHAPTER IV 45
RESULTS .................................................................................................................................. 45
Findings and Analysis .................................................................................................... 45
Spontaneous Speech Measurement ........................................................................ 46
Naming Measurement ............................................................................................ 64

MIT Measurement .................................................................................................. 70
CHAPTER V 71
DISSCUSSION .......................................................................................................................... 71
Summary and Conclusion .............................................................................................. 7]
Implications for Interventions ........................................................................................ 80
Implications for Future Researchers ............................................................................. 83
APPENDICES 86
Measurement F orms & M T Hierarchy ......................................................................... 87
Picture Stimuli Sheets for Naming Test ........................................................................ 97
F lyer & Consent Form ................................................................................................. 113
ASHA levels of Comprehension, Expression, Fluency ................................................. 116
Raw Data & Statistics Data ......................................................................................... 119
REFERENCES- 127

 

LIST OF TABLES

Table 1. Subjects Description .......................................................................................... 18
Table 2. L’s Linguistic Contents ..................................................................................... 26
Table 3-1. MIT Hierarchy Level I ................................................................................... 28
Table 3-2. MIT Hierarchy Level II ................................................................................. 31
Table 3-3. MIT Hierarchy Level III ................................................................................ 33
Table 3-4. MIT Hierarchy Level IV ................................................................................ 34
Table 4—1. Scoring Hierarchy of Spontaneous Speech Measurement ............................. 41
Table 4-2. Scoring Hierarchy of Naming measurement ................................................. 42
Table 4-3. Scoring Hierarchy of MIT Measurement ....................................................... 43
Table 5-1. & 5-2. L’s Paired t-test on Total Length of Samples ..................................... 47
Table 5-3. & 5-4. R’s Paired t-test on Total Length of Samples .................................... 49
Table 6—1. & 6-2. L’s Paired t-test on Total Number of Utterances ............................... 50
Table 6-3. & 6-4. R’s Paired t-test on Total Number of Utterances ............................... 52
Table 7-1. & 7-2. L’s Paired t-test on Total Number of Words ...................................... 54
Table 7-3. & 7-4. R’s Paired t-test on Total Number of Words ..................................... 55
Table 8-1. & 8-2. L’s Paired t-test on Average Length of Utterances ............................ 57
Table 8-3. & 8-4. R’s Paired t-test on Average Length of Utterances ............................ 59
Table 9-1. & 9-2. L’s Paired t-test on Average Number of Words per Minute .............. 60
Table 9-3. & 9-4. R’s Paired t-test on Average Number of Words per Minute .............. 62
Table 10-1. & 10-2. L’s Paired t-test on Total Number of Items named ........................ 64
Table 10-3. & 10-4. R’s Paired t-test on Total Number of Items named ........................ 66

vi

LIST OF TABLES

(continued)
Table 11-1. & 11-2. L’s Paired t-test on Average Number of Prompts per Item ........... 67
Table 11-3. & 11-4. R’s Paired t-test on Average Number of Prompts per Item ........... 69
Table 12. Summary of the data result in t-test ................................................................ 72
Table 13-1. & 13-2. The second-half period in L’s paired t-test .................................. 107
Table 14-1. & 14-2. The first-half period in L’s paired t-test ....................................... 108
Table 15-1. & 15-2. Independent t-test for L’s Goal 111 ............................................... 110

vii

LIST OF FIGURES

Figure 1. One Group pre-test-post-test Design ................................................................. 23
Figure 2. Subject L’s Sessions Diagram ........................................................................... 24
Figure 3-1. Total Length of Sample in L’s Spontaneous Speech Test ............................. 47
Figure 3-2. Total Length of Sample in R’s Spontaneous Speech Test ............................. 48
Figure 3-3. Total Number of Utterances in L’s Spontaneous Speech Test ....................... 50
Figure 3-4. Total Number of Utterances in R’s Spontaneous Speech Test ...................... 52
Figure 3-5. Total Number of Words in L’s SP Test .......................................................... 53
Figure 3-6. Total Number of Words in R’s SP Test ......................................................... 55
Figure 3-7. Average Length of Utterances in L’s Spontaneous Speech Test ................... 57
Figure 3-8. Average Length of Utterances in R’s SP Test ................................................ 58
Figure 3-9. Average Number of Words per Minute in L’s SP Test .................................. 60
Figure 3-10. Average Number of Words per Minute in R’s SP Test ................................ 62
Figure 4-1. Total Number of Items Named in L’s Naming Test ...................................... 64
Figure 4-2. Total Number of Items Named in R’s Naming Test ...................................... 65
Figure 4-3. Average Number of Prompts per Item Named in L’s Naming Test .............. 67
Figure 4-4. Average Number of Prompts per Item Named in R’s Naming ...................... 68
Figure 5—1. MIT Scores of Percent in L’s MIT Test ......................................................... 70

viii

CHAPTER I

INTRODUCTION

Music is a method of communication that is both nonverbal and verbal. Music
helps people who have difficulty speaking their minds mentally or physically, in order to
give them a voice in a nonverbal musical medium. On the other hand, music also helps
the people who have a functional difficulty in verbal expression, in order to improve their
ability to communicate by training them with musical cues, such as rhythm and melody.
When I visited the meeting of the aphasia support group as a volunteer, an intriguing
observation was that persons with expressive aphasia could sing with steady rhythmic
patterns and speech production without long pauses between words, although they
couldn’t speak when they tried talking to someone. This research will seek to provide
important information to help the persons who have fimctional difficulty in producing
language expression by demonstrating the effectiveness of music in functional
. communication through singing.

There are different conditions that contribute to functional language difficulty,
and stroke is one of the major causes. According to data from the American Heart
Association and the National Institute of Health, 600,000 to 750,000 strokes occur in the
US each year. A stroke, which causes damage to the posterior part of the frontal lobe in
the left hemisphere of the brain, results in language disorders including Broca’s aphasia
(expressive) and Wemicke’s aphasia (receptive). Expressive aphasia is characterized by
disturbed prosody, effortful and slow speech with long pauses between words, poor
sentence construction, difficulty naming objects, word-fmding difficulties, and poor word

repetition (Baker & Tamplin, 2006).

Speech and music share neural networks in the brain because music and language
both use structured patterns of pitch (melody), duration (rhythm), and intensity
(dynamics) (Patel, Peretz, Tramo, & Labreque, 1998). According to the research
(Borchgrevink, 1982; Zatorre, Evans, & Meyer, 1994), pitch in melody is processed only
by the right hemisphere, but rhythm is processed by the left hemisphere, so music
including rhythm and pitch stimulates both right and left hemisphere in the brain. This
rationale would support the idea that singing which includes pitch and rhythm could
activate both hemispheres in the brain; music would be beneficial to persons who have
Broca’s aphasia with damage to expressive speech functions by stimulating the left side
of their brains.

Neurologic Music Therapy (NMT) is one of the dominant music therapy
approaches, which is based on a neuroscience model of music perception and production
and the influence of music on functional changes in nonmusical brain and behavior
fimctions (Thaut, 2005). NMT has several standardized techniques in three different areas
of functional training: 1) Sensorimotor Training, 2) Speech and Language Training, and
3) Cognition Training. The dominant domain is the Sensorimotor area in NMT training,
which is well established by plenty of research, live demonstration with patients, and
collaboration with physiological experts. In recent years, several studies have attempted
to find the reasons for the treatment effect of MIT and establish standardized techniques
in Speech and Language Training and Cognition Training. According to the present

research by speech and language pathologists, rhythmic stimulus can be an important

musical element to activate the human nervous system. For this reason, rhythmic
instruments were used to stimulate the subject by rhythmic beats in the present study.

In NMT, Melodic Intonation Therapy (MIT) is one of the most important
techniques of Speech and Language Training. MIT was established first from the speech
pathologist, Sparks, in thel97 Os, and later the speech pathologist, Galloway, made the
first contribution that connected music therapists and speech therapists by investigating
the effectiveness of MIT in speech functions (Galloway & Kraus, 1982). This technique
proved the power of musical elements, such as rhythm and melody, in brain functions
(Sparks, Helm, & Albert, 1974).

While I participated in the NMT training in 2009 at Colorado State University,
music therapists mentioned that research on the Speech and Language domain is still in
its early stages, as the brevity of the bibliography attests in the Music Therapy field. I
have observed from small workshops to national conferences in music therapy that many
music therapists use NMT techniques without clear recognition of what techniques they
are using, such as MIT, Musical Speech Stimulation (MU STIM), Rhythmic Speech
Cueing (RSC), Vocal Intonation Therapy (VIT), Therapeutic Singing (TS), Oral Motor
and Respiratory Exercises (OMREX), Developmental Speech and Language Training
through Music (DSLM), and Symbolic Communication Training through Music
(SYCOM). This study will be intended to broaden our horizons of understanding of the
effectiveness of functional communication through music in the Speech and Language

Training area of music therapy.

STATEMENT of PURPOSE

With the intent of identifying effective music therapy techniques, the purpose of
this study is to ascertain the effectiveness of Therapeutic Singing and Melodic Intonation
Therapy in improving the functional communication skills of adults with expressive
aphasia.

The goals for subject L’s case study, who had a lower level of fimctional
communication skills, are as follows: (1) to give specific information whether
Therapeutic Singing (TS) combined with Melodic Intonation Therapy (MIT) helps adult
clients who are diagnosed with expressive aphasia as measured by the quality of their
functional communication skills; (2) to gain some information as to whether MIT
treatment itself helps clients improve the quality of their functional communication skills;
(3) to compare the effectiveness of TS combined with MIT and MIT alone in increasing
fimctional communication skills; and (4) to investigate whether TS enhances the benefits
of MIT treatment.

The goal for subject R’s case study, who had a higher level of functional
communication skills, is as follows: 1) To acquire information as to whether Therapeutic
Singing (TS) itself helps clients improve the quality of their functional communication

skills.

CHAPTER II

LITERATURE REVIEW

This literature review reflects information in the following areas: 1) music and
brain; 2) stroke and aphasia; 3) music and speech; 4) melodic intonation therapy (MIT)
and therapeutic singing (TS); 5) modified melodic intonation therapy (MMIT); and 6)

measurement tools for functional communication skills.

Music and Brain

Throughout brain science research over the past two decades, the element of
music has emerged as a factor in answering the question of what stimulates brain
functionality. Sacks (2006) addresses this question by stating that music provides one of
the most powerful sources of auditory stimulation in the human brain. Also, evidence
from developmental studies indicates that music can increase auditory cortical functions
of the brain (Engineer, et al., 2004).

The belief that music is processed only in the right hemisphere has been rejected
by researchers (Hachinski & Hachinski, 1994). They found musical activity involved
almost every region of the brain that was known, and almost every neural subsystem.
They also insisted that different neural locations throughout the cortex handled different
aspects of the music. Actually, the brain uses functional segregation for music
processing, and employs a system of feature detectors whose job is to analyze specific
aspects of the musical signal, such as pitch, rhythm, tempo, timbre, and harmony
(Levitin, 2006). These cerebral locations in the brain are not restricted to only the right

hemisphere as past information suggests (Prior, Kinsella, & Giese, 1990).

Stroke and Aphasia

According to data from the American Heart Association and the National Institute
of Health, 600,000 to 750,000 strokes occur in the US each year (Schlang, Marchina, &
Norton, 2008). Moreover, according to the World Health Organization, Cerebral Vascular
Accidents (CVA) affect 15 million people worldwide each year, with five million people
left permanently disabled. A CVA, commonly known as a stroke, results in the rapidly
increasing loss of brain functions when a part of the brain is deprived of blood flow, and
brain cells subsequently die. Overall, CVA affects 11% of females and 8.4% of males
(Mackay & Mensah, 2004). The most well-known Neurogenic speech disorder, aphasia,
is most often caused by a cerebral vascular accident.

People who have experienced a CVA are known as having a left CVA or right
CVA, depending on the site of the lesion. The patients with a left CVA experience
problematic physical functions and communication. The left CVA causes the most
apparent impairments of the motor control and sensory experience of the right side of the
body and causes a right hemi-paresis. Furthermore, a left CVA also causes language
disorders including Broca’s aphasia (expressive) and Wemicke’s aphasia (receptive)
(Baker & Tamplin, 2006).

Expressive aphasia is caused by damage to a region of the inferior left frontal
lobe, and it is also called motor aphasia, non-fluent aphasia, or Broca’s aphasia. This
results in the disorders of speech production characterized by slow, laborious, and non-
fluent speech. People with expressive aphasia have difficulty saying basic words with

semantic meaning. On the other hand, people with receptive aphasia, otherwise known as

sensory aphasia, fluent aphasia, or Wemicke’s aphasia, show poor speech comprehension
and production of meaningless speech (Goodglass & Kaplan, 1972).

According to several examinations of auditory comprehension, a person with
Broca’s aphasia shows understanding and retention of spoken language that is essentially
normal in a variety of contexts (Sparks & Deck, 1986). Sparks also indicates that the
examination of verbal expression provides a profile of the good candidate for Melodic
Intonation Therapy (MIT). He states:

1. “Almost no responses occur in confrontation naming, responsive naming, word
and phrase repetition, or sentence completion. However, an occasional response
will be poorly articulated but accurate enough to indicate correct encoding of the
target wor .”

2. “Articulation of stereotype phrases will be precise and easily transcribed into
phonetic symbols. On the other hand, any attempt to duplicate or initiate
propositional language results in slurred articulation and phonemic substitutions.”

3. “Effort at self-correction is often vigorous. This is to be expected in an aphasic
who is acutely aware of making errors in his verbal output. Unfortunately, the

product is not improved by this effort.” (p. 255)

There are useful selection criteria for adult aphasia patients, with the best
candidates for MIT including: “(a) no evidence of bilateral brain damage, (b) average-to-
high receptive abilities, (c) poor repetition of single words, ((1) non-fluent verbal
production with inadequate articulation agility and effortful ignition of verbal
communication, and (e) a well motivated, emotionally stable patient with an appropriate
attention span” (Benson, Dobkin, Rothi, Helm-Estabrooks, & Kertesz, 1994, pp. 566-

568)

However, Sparks et a1. (1986) also show that the other three kinds of aphasia are
not receptive to MIT as a form of treatment. There are three types of poor candidates for
MIT treatment: Wemicke’s, T ranscortical, and Global aphasia. People with these types of

aphasia are very difficult to engage in MIT and achieve minimal therapeutic success.

Music and Speech

According to Sacks (2008), speech is not just a succession of words in the proper
order. It contains inflections, intonations, tempo, rhythm, and melody. Speech and music
both are dependent on phonology and articulatory mechanisms that are elementary in
other primates, and for their appreciation both are dependent distinctly on human brain
mechanisms dedicated to the analysis of complex, segmented, rapidly changing streams
ofsound.

Brain processing of music and speech overlaps, and there has been extensive
research to figure out how music stimulates speech. Belin (1996) insists that musical
intervention has been used as part of speech therapy in stroke rehabilitation. There is
evidence that music therapy practice is effective in alleviating speech disorder problems,
such as Broca’a aphasia (Naeser & Helm-Estabrooks, 1985). King (2007) also gives
convincing information of the effectiveness of music in brain functionality. This is
important for music therapists to know and use in reviewing the definitions and
characteristics of adult onset aphasia, as well as apraxia and dysarthria.

Jackson (1871) characterized propositional speech from so called automatic
speech. He stated that even when propositional speech became seriously damaged,
automatic speech could be preserved in aphasia. Hobson (2006) gives a rationale for

applying music therapy methods to communication disorders, as well as definitions and

descriptions of several speech disorders relevant to music therapy practice. He notes that
people with aphasia are able to sing familiar songs but are not able to engage in
conversations. The classical interpretation of this observation is that singing familiar
songs would depend on right hemisphere functions, whereas propositional speech would
depend on left hemisphere functions. But it 'has not been substantiated by quantitative
data yet (Hebert, Racette, Gagnon, & Perets, 2003). Sacks (2008) pointed out that singing
familiar songs was a form of automatic speech as well as cursing or reciting a poem.
With this apt observation, Sacks (2008) reformulated the question of music therapy:

A person with aphasia may be able to sing or curse or recite a poem but not to
utter a propositional phrase. The question of whether singing has any use in the
recovery of speech, then, can be formulated another way: can language embedded

in unconscious automatism be “released” for conscious, propositional use? (p.
218)

This shows an important role of music, especially singing, for speech therapy as a
“trigger”. In his view, it could be possible that by re-experiencing language even though
it is a kind of wholly automatic language embedded in music, cortical areas previously

inhibited but not damaged can be de-inhibited (Sacks, 2008).

Melodic Intonation Therapy

Melodic Intonation Therapy (MIT) was introduced in 1973 through successful
research using three chronic non—fluent aphasics (Albert, Sparks, & Helm, 1973). MIT is
a treatment technique for the rehabilitation of expressive (Broca’s) aphasia. There is also
support for its use with apraxia. Fitting candidates for MIT are clients with left-side

stroke lesions in Broca’s area, which is responsible for encoding speech production, or

clients with lesions interrupting the nervous connections between Wemicke’s and
Broca’s area. MIT utilizes a patient’s unimpaired ability to sing to make speech
production possible. The rationale is based on a hemispheric transfer of speech functions
from the left-hemisphere (Wemicke’s area) to its right-hemisphere homologue during
encoding of speech into singing (Albert, et al., 1973; Helfrich-Miller, 1994; Keith &
Aronson, 1975; Overy, Norton, & Ozdemir, 2005; Popovici, 1995; Sparks, et al., 1974).
Several published reports over the years have outlined a programmed technique
for MIT,(Helm-Estabrooks & Albert, 1991; Sparks & Holland, 1976) which include
relatively strict criteria for patient selection. Functional sentences or utterances in MIT
are translated into a song by translating the speech inflection patterns into musical rhythm
and intonation. Therefore, it is technically more accurate to refer to the sung utterances
not as songs, but as melodic intonations. Singing is reduced to Sprechgesang (speech

singing) and finally to normal speech pattern in later steps of the therapy.

The important characteristics are the rhythmic and melodic elements for the
therapeutic stimulus (Thaut, 2005). The range of musical notes is limited to representing
only slightly greater inflection than the pattern of normal speech. MIT works in a gradual
progression and is designed to lead the client through a sequence of therapy steps, which
gradually increases the length of intoned sentences, decreases dependence on the
therapist by fading hand tapping and singing, and increases the client’s independence
from intonation. The client is guided through a sequence of four progressive levels of
difficulty, each containing a series of discrete steps. A criterion of 90% correct responses
at each level is needed to ensure gradual progression, as well as the advancement up to

the next level. In each step, target sentences or phrases are practiced until the level is

10

completed or failure occurs for that particular sentence (Sparks & Deck, 1986). Based on
the supportive literature and clinical observation, the song strategy proves helpful for
patients with left frontal lobe damage or bilateral damage that leaves singing centers of
the right temporal lobe relatively intact (Lucia, 1987) .

According to research from Emory University, neurological development is
steady, but critical periods of rapid cognitive growth are known to occur. Musical
techniques such as MIT could be beneficial to children suffering speech and language
problems due to some subtle cerebral dysfimction, since the child’s brain possesses
resilience and plasticity to some developmental point (Galloway & Kraus, 1982).
According to the research (Albert, etal., 1973), MIT is effective for 75% of adults with
non-fluent aphasia, and for 25% of people with severe non-fluent aphasia with little
change in their ability to communicate verbally in meaningful ways. There are currently
very few treatment techniques for in use of language therapy, and many are not uniformly
practiced to allow for consistent measurements. MIT can satisfy consistency

requirements to offer beneficial variation for research-level studies (Benson, et al., 1994).

Therapeutic Singing

Both singing and speech are essential tools for human expressive communication
in both nonverbal and verbal ways. Singing has been described as an “automatic speech
skill” which generally leads to functional speech treatment (Lucia, 1987). According to
Thaut (2005), therapeutic singing (TS) refers to an unspecific use of singing activities in
groups to facilitate initiation, development, and articulation in speech and language, as

well as to increase functions of the respiratory apparatus, used with a variety of

neurological or developmental speech and language dysfunctions. (S. Jackson, Treharne,
& Boucher, 1997; Thaut, 2005)

Singing has also been recommended as a therapeutic intervention specifically for
people with Broca’s aphasia. (Benton, 1977; Gerstrnan, 1964; Gleason & Goodglass,
1984; Keenan, 1987). According to the research, while examining the preservation of
singing in 24 patients with Broca’s aphasia, 87.5% of the subjects produced accurate
melodies, and 57% sang with correct diction (Yamadori, Osumi, Masuhara, & Okubo,
1977). Singing was. recommended for the rehabilitation of expressive deficits as early as
1953 (Vargha & Gereb), when it was observed that many aphasics could sing the words
of previously learned songs better than they could speak.

Speech therapists are looking for new interventions to use with speech-disordered
patients. Singing can be a valuable tool in speech rehabilitation because of its similarities
to speech production. Singing is a musical behavior which shares with the speech
elements of fundamental frequency, fundamental frequency variability, vocal intensity,
rhythm or rate, and diction (Cohen, 1992). The use of singing has been successfully
demonstrated to facilitate speech development with speech-disordered clients (Crocker,
1958; Darrow & Starker, 1986; Lathom, Edson, & Toombs, 1965; Leung, 1985; Ogden,
1982; Vanderark, 1986).

Another research study (Keith & Aronson, 1975) was conducted with an adult
woman with severe expressive aphasia. After a month of traditional speech therapy, her
progress had been minimal, so the researchers decided to give her a one-hour singing

therapy session once a week. After a short period of treatment, the patient was able to

initiate speech with song-like phrases. Two months later, she had progressed to the point

where she could eliminate the singing entirely from her speech.

Differences between MIT and TS

At this point the author needs to give convincing answers to the questions
regarding the differences between MIT and TS. Therapeutic singing (TS) has distinct
melodies and melodic intonation therapy is based on the spoken prosody of verbal
utterances, so MIT has a limited range of sung notes and is comfortable for the untrained
voice of adults.(Sparks & Holland, 1976). MIT, when using an intoned utterance which
resembles a familiar song, may produce disastrous results. The familiar melody will
stimulate a recall of the words of that song. According to the article by Sparks (1974),
researchers selected for their client the target phrase, “meat and potatoes.” Then
researchers asked the client to use a melody that resembled a popular song, “I love my
baby, my baby loves me”. When the target phrase was presented and the aphasia client
was asked to join in, the client altered it to, “Meat and potatoes, my baby loves me”. For
this reason, MIT prohibits the use of familiar or popular melody lines in order to sing
target phrases. Therapists must create a new melody prosody. On the other hand, for TS it
is valuable to use popular or common songs which the clients already know and prefer to
sing in order to improve their utterances of speech by the possible stimulation of the brain

regions for long term memory.

 

Modified Melodic Intonation Therapy

Modified Melodic Intonation Therapy (MMIT) makes use of the strong
association between MIT and TS. Baker (2000) has developed a modified version of MIT
technique for working with persons with non-fluent expressive aphasia. In this technique,
target phrases are more melodic than Sprechgesang (speech song), and the natural pattern
of speech inflection is revised to create a more musical structure. The phrases chosen in
modified MIT (MMIT) should be meaningful for functional communication (Baker &
Tamplin, 2006).

MMIT phrases are composed of a limited range of pitches within an octave to
ensure that the patients are able to sing them comfortably. Two criteria prescribe the
choice of musical phrases: first, the phrase is essentially different from the phrases
preceding and following it; and second, the creation of a melodic phrase that can be
easily encoded in memory and then produced on the researcher’s requests (Baker &
Tamplin, 2006).

Baker designed a procedure with six steps involved in the process:

1) Client sings familiar songs; 2) Therapist sings and plays functional phrases to

clients; 3) Therapist sings the phrases; 4) Therapist withdraws participation; 5)

Client practices independent production of learned words and phrases. (Musical

cues may be provided initially and later faded); and 6) Client practices use of the

phrases out side of music therapy context (pp. 148-149).

In the first step clients are encouraged to sing familiar, well-known songs. It
provides clients with immediate positive feedback of verbal output, motivation, and

distraction when progress is slow and difficult. The second step involves the introduction

of a small number of meaningful phrases set to simple melodies. However, it is important

 

that phrases used should not resemble familiar melodies, as this might result in the
accidental generation of non-propositional speech comprising the original lyrics of the
song. The third is that clients are directed to sing the phrases with the therapist. And then,
a therapist gradually withdraws participation and clients encourage to sing the phrases
unaided in the fourth step. In the fifth step a therapist asks clients questions to test
independent word generation. When this step has been completed, new phrase or short
sentence can be introduced with new melodic phrases. The final step involves the use of
the target words in normal conversation outside of the music therapy session (Baker &

Tamplin, 2006).

Measurement tools for functional communication skills

Lezak (Lezak, Howieson, Loring, Fischer, & Hannay, 1995) suggests that a
review of client’s language and speech functions to indicate whether communication
problems are present will include examination of the following aspects of verbal
behavior: 1) Spontaneous speech; 2) Repetition of words; 3) Speech comprehension; 4)
Naming; 5) Reading; and 6) Writing.

The Multimodal Communication Screening Test for Aphasia (MCST-A) is one of
3 criterion-referenced assessment tools in the Augmentative and Alternative
Communication (AAC) Assessment Battery for Aphasia (AAC-ABA) developed by K.
Garrett & J. Lasker. The MCST-A is designed to systematically assess whether people
with severe aphasia can use alternative or augmentative modalities to communicate via
alternative pictorial symbols. The second tool is Systems Trials protocol for Aphasia

(AAC-STA), and the third tool is AAC Categorical Assessment (AAC-CAT) which will

 

aid in determining whether the person with aphasia can benefit from partner dependent or
independent alternative communication strategies (Garrett & Lasker, 2005). MCST-A
will be used to modify the Naming measurement picture sheets in order to investigate the
effectiveness of functional communication in this research.

Another clinical application for measurement is the Aphasia Diagnostic Profiles
(ADP) by Nancy Helrn-Estabrooks. ADP is a systematic method of assessing language
and communication impairment associated with aphasia. Designed to meet the needs of
medical settings, it creates profiles that explain critical areas of the client’s performance.
Nine brief subtests create composite scores. It indicates overall severity and specific
strengths and weaknesses. The severity profile identifies the client’s strongest response
modalities. Error profiles identify the communication value of a client’s response.
Behavioral profile indexes the client’s overall social-emotional state during testing. This
assessment takes 40 to 50 minutes, but the tool will be revised to 10 minutes for pre-post

tests in the current study.

With the intent of identifying effective music therapy techniques, the purpose of
this study is to ascertain the effectiveness of Therapeutic Singing and Melodic Intonation
Therapy in improving the functional communication skills of adults with expressive
aphasia. The two case studies were conducted with subjects who were in different level
of functional speech. The goals of for subject L’s case study are as follows: (1) to
determine whether Therapeutic Singing (TS) combined with Melodic Intonation Therapy
(MIT) helps adult clients who are diagnosed with expressive aphasia as measured by the

quality of their functional communication skills; (2) to determine whether MIT treatment

16

 

itself helps clients improve the quality of their functional communication skills; (3) to
compare the effectiveness of TS combined with MIT and MIT alone in increasing
functional communication skills; and (4) to investigate whether TS enhances the benefits
of MIT treatment. On the other hand, the goal for subject R’s case study is as follows: 1)
To give information whether Therapeutic Singing (TS) itself helps clients improve the

quality of their functional communication skills.

 

CHAPTER III

METHOD
Subjects

Two adults with diagnoses of expressive aphasia with different levels of
functional speaking were selected to participate in this study. One was described by the
speech pathologist as high level and the other as low level in functional speaking. The
subjects were recruited with the help of speech-language pathologists from the Aphasia
Support Group in the Lansing area, explaining the research study in the group meeting.
Announcements were also placed in the e-mail newsletters of the Aphasia Support Group
(see Appendix 9- Flyer). Three restrictions were imposed on selecting the subjects: 1)
Diagnosis of expressive aphasia caused by CVA; 2) Adult males; 3) No other therapy
received, such as speech or physical therapy, during the course of this study.

There are some differences in the two subjects. L is considered to have a the low
level of functional communication skills, while R is considered to have a high level of
functional communication skills, which means is that R did not need to receive MIT
treatment because he was capable of success in the final level of MIT treatment. (see

Table 1)

Table 1 Subject Description

 

Subjects Treatments Age Gender Diagnosis 32:38:): *SLC *SLE Fluency

Expressive Aphasia

L TS & MIT 62 Male 1 year Level Level

Mild Apraxia 6 3 Lev“ 2
Expressive Aphasia
Level Level
R TS 53 Male Right Dysfunction 7 years 7 5 Level 5

 

(* SLC — Spoken Language Comprehension, * SLE - Spoken Language Expression)

Subiect L - Low Functional Communication Skill

On May 09, 2008, L had a sudden onset cerebrovascular accident (CVA) from a
blood clot in the left parietal lobe. It resulted in expressive aphasia, mild apraxia of
speech and the loss of memory fimction, both long-tenn and short-term. His spouse stated
that his stroke was caused by high cholesterol and the distribution of stress. He was
hospitalized after onset of the stroke and had undertaken speech therapy and physical-
occupational therapy, but he wasn’t receiving any other treatments while this research
treatment was being provided.

In the area of Speech and Communication, L showed some limited characteristics
in functional communication skills: 1) non-fluent, effortful, slow, halting and uneven
speech; 2) limited word output, with short phrases and sentences; 3) misarticulated and
distorted sounds; 4) agrammatical speech; 5) impaired repetition of words and sentences;
6) impaired naming; and 7) poorer production of speech than comprehension.

The Spoken Language Comprehension (SLC) level was rated at a high level 6.
According to the FCMs in America Speech and Hearing Association (ASHA), the level 6
in SLC states that “the individual is able to understand communication in most activities,
but some limitations in comprehension are still apparent in vocational, avocational, and
social activities. The individual rarely requires minimal cueing to understand complex
sentences. The individual usually uses compensatory strategies when encountering
difficulty” (See APPENDIX 11). On the other hand, the Spoken Language Expression (SLE)
level was rated at a low level 3. According to the FCMs in ASHA, a level 3 in SLE
means that “the communication partner must assume responsibility for structuring the

communication exchange, and with consistent and moderate cueing, the individual can

 

produce words and phrases that are appropriate and meaningful in context” (See
APPENDIX 12). Furthermore, L was rated at the low level 2 in Fluency. According to the
FCMs in ASHA, a level 2 in Fluency refers to Speech that is “functional most of the
time, but labored in many day-to-day situations due to extended disruptions of speech
flow which sometimes renders the individual difficult to understand. Participation in
vocational, avocational, and social activities requiring speech is reduced overall. Listener
discomfort is evident throughout conversational interactions” (See APPENDIX 13).

In the area of Sensorimotor skills, his fine motor function on the right side was
not fully stable, but he was able to ride a bike by himself. There also was limited control
and ability in the facial muscles because he was diagnosed with mild apraxia. Therefore,
the limited usage of tongue and mouth muscles caused limited accuracy of pronunciation
in vowel sounds and especially in consonant sounds.

In the area of cognitive function, there was a slight loss of memory functions. His
wife stated that during the year after onset of the stoke, his short-terrn memory was

coming back, but some of his long term memories were still blank.

Subject R - High Functional Communication Skill

R had a second stroke (CVA) seven years ago, with his first stroke (CVA)
occurring twelve years ago. He was diagnosed with acute expressive aphasia and right
side dysfunction. He had received speech, physical, and occupational therapies after the
onset of the stroke, but he hasn’t been given any other therapy for several years, and there

was no input from other therapies during the music therapy intervention.

20

In the area of Speech and Communication, L showed some limited characteristics
in functional communication skills: 1) fluent, effortful, slow; 2) limited word output; 3)
no impaired repetition of words and sentences; and 4) poorer production of speech than
comprehension.

The Spoken Language Comprehension (SLC) level 7 was higher than subject L’s.
According to the F CMs in ASHA, a level 7 in SLC states, “the individual’s ability to
independently participate in vocational, avocational, and social activities are not limited
by spoken language comprehension. When difficulty with comprehension occurs, the
individual consistently uses a compensatory strategy” (See APPENDIX 11). On the other
hand, the Spoken Language Expression (SLE) level 5 was higher than L’s. According to
the F CMs in ASHA, a level 5 in SLE indicates “the individual is able to successfully
initiate communication using spoken language in structured conversations with both
familiar and unfamiliar communication partners. The individual occasionally requires
minimal cueing to frame more complex sentences in messages. The individual
occasionally self-cues when encountering difficulty” (See APPENDIX 12). Furthermore, The
Fluency level of 5 for subject R was higher than that of that of subject L. According to
the FCMs in ASHA, a level 5 in Fluency indicated that “Speech is functional for
communication, and fluency can be maintained in some situations. Self-monitoring is
inconsistent. The frequency and severity of disruptions of speech flow within problem
situations is distracting some of the time. Speech difficulties are noticeable when they
occur, and sometimes limit vocational, avocational, and social activities requiring speech
in problem situations. Listeners are occasionally aware of fluency difficulties relative to

particular situations” (See Appendix 13).

21

In the area of Sensorimotor skills, there was a weakness in right side vision and
limited horizontal range of vision in the right eye. R had difficulty in reading letters line
by line because of the loss of vision. The fine motor skill was functioning quite well, but
the tactile sensation in his right hand somewhat dysfunctional.

In the area of cognitive function, his memory had returned since the onset of the
stroke. His long term and short term memory was satisfactory, but there was a memory
loss of specific professional skills in his occupation.

This study was explained to each subject and the general content was reiterated to
the subjects. Test measurements were presented briefly by using a Macintosh notebook.
After subjects verified their agreement, their guardians and subjects were asked to sign

the informed consent. (See Appendix 10 — Consent form)

Setting and Apparatus

The study was conducted in an available meeting room to which subjects had
access, such as the living room or dining room at the subject’s homes. The materials and
equipment used in the study were: response sheets, lyric sheets, pencils, castanets, a
keyboard, a stop watch, a iMovie software, and PowerPoint software in the Macintosh
notebook.

While testing with the Spontaneous Speech Measure, iMovie software in the
Macintosh notebook was used to record the subject’s production of speech, and while
testing with the Naming Measure, PowerPoint software was used to randomize 90
pictures to form each sixteen-picture test. A Samsung stop watch was used to measure the

given minutes. During MIT Treatment, castanets were used to help subjects and

22

therapists make rhythmic stimulus with one hand, instead of working on hand tapping

with both hands. Lyric sheets and castanets were also used for the TS treatment.

Lam—Wise

A pre-experimental design consisting of a one-group pre-test-post-test study was
used to administer the procedure. The subjects were pre-tested, received treatments, and
then post-tested in each session (Albert, et al., 1973). A benefit of this design is the
inclusion of a pre-test to determine baseline scores (2005), a useful design when there are

not enough subjects to provide a control group. (See Figure 1)

Figure 1. One Group pre-test-post-test Design

vmrfilfi»y¢n3za “law-r.

Pretest

       

www— '—

 

(Source: Dooley, 1990)

There were eight dependent variables for subject L who had low functioning
expressive aphasia, and seven dependent variables for subject R, who had high functional
speech because this subject did not need to receive MIT treatment: 1) Total length of
samples; 2) Total number of utterances; 3) Total number of words spoken by subjects; 4)
Average length of utterances; 5) Average Number of Words per Minute in the
Spontaneous Speech assessment of the Functional Communication Measure; and in the
naming test: 6) the number of items named by a client; and 7) the number of prompts per
item named given by a therapist in the Naming Assessment of the Functional

Communication Measure (FCM); and in MIT test; 8) the percent scores collected by the

23

Melodic Intonation Therapy Measure; these variables were obtained from the pre-test,
post-test, and MIT test, revised with the support of a speech-language pathologist and
based on Aphasia Diagnostic Profiles (ADP) and Functional Communication Measures

(FCMs) from the Adult National Outcomes Measurement System (NOMS).

ELM

The intervention period for the two subjects consisted of sixteen individual
sessions, three to five times a week for 80 minutes, including pre and post-tests. The
period for subject L (low function) was divided into two different terms of interventions
of eight sessions each (16 sessions in all): 1) MIT in the first period of 8 sessions; 2) TS
combined with MIT in the second period of 8 sessions (see Figure 2. Subject L’s Sessions
Diagram). On the other hand, Subject R (high function) received TS itself for all 16
sessions because he could complete the final level of MIT at the outset and did not
require MIT treatment. He could repeat items in normal speech patterns from the
beginning of the sessions. When the researcher gave signals, he replied with appropriate

and accurate sentences.

Figure 2. Subject L ’s Sessions Diagram

 

       
 

session ltl scx‘sinl‘. 15 ~.t‘\\.il)il lb
sr-ssron 2 session it ‘

MIT

'15 '18

erl'

MIT

 

 

24

Subject L ’s Intervention

The intervention strategies consisted of the original MIT protocol (Albert, et al.,
1973) in the first period of sessions and TS protocol preceding MIT treatment in the each
of second period of sessions. After pre-testing the Spontaneous Speech Test for 13 min.
and the Naming Test for 5 min. before every session, a form of Melodic Intonation
Therapy (MIT) was introduced, and L worked on the steps of each level of MIT for
around 30 min. for eight sessions (from session 1 to 8). After half way through, by the
eighth session of the procedure, Therapeutic Singing (TS) treatment was added from
sessions nine to sixteen after the pre-test measures. Therefore, subject L received the pre-
test, TS, MIT, and the post-test during the second period of sessions. There were two
techniques used during his intervention: 1) Melodic Intonation Therapy (MIT); 2)

Therapeutic Singing (I S), described in the following paragraphs.

1) Melodic Intonation Therapy (MIT)

The primary goal of MIT for non-fluent aphasics is the basic recovery of verbal
utterances with the accurate use of language. There is also research evidence for its use
with apraxia (Galloway & Kraus, 1982). Therefore, Subject L is a good candidate to
practice MIT, as a subject who suffered from both non-fluent expressive aphasia and mild
apraxia, to improve functional communication skills. There are four levels of MIT
hierarchy and it includes three to four steps in each level. Subject L started his first
sessions on level I and II and was capable of improving to level III at the end of the
session. Before interventions began, the relevant sample materials of MIT needed to be

collected.

25

Linguistic Contents

The selection of linguistic content is important, depending on the severity of the
aphasic’s communication inability and the usefulness of the verbal material (Sparks &
Deck, 1986). Subject L’s sentences were chosen by the suggestions of his wife and
himself about things as basic as family relationships, personal needs, and his background
information. Content was subsequently selected regarding the complexity of syllables in
each level, but some level II sentences were used in level III. According to the Sparks’
article (1986), meaningful stimulus samples for the aphasic who has no impairment other
than phonological errors should focus on producing more intelligible speech. Subject L’s
collection of contents, however, was more focused on the usefirlness to his daily life.
Regarding the ability to produce utterances, less important words, such as prepositions,

were eliminated from the contents. The phrases and sentences are illustrated in Table 2.

Table 2. L ’s Linguistic Contents

 

Subject L’s Sample linguistic contents for Level II

 

What time? My name is L My wife is Marlene
Take Sadie walk I had a stroke Today is Friday

Good girl Sadie I had a dog What time want eat?
One, two, three I love you, Marlene I have trouble speaking
Cup of coffee Thanks for being I work at GM

Go to sleep One, two, three, four I can understand you
Jump Sadie My lovely wife Happy birthday to you
Stay with me I love hunting The love in my life
Thank you I love my country Thank you so much
One O’clock God bless you

1 love you Time for lunch

 

 

Subject L’s Sample linguistic contents for Level IH

 

Happy birthday to you Go to bed at night 1 love to go hunting

My name is L I love my wife, Marlene There will be a baby deer
My wife is Marlene 1 have had a stroke 1 saw a deer this morning
One or two babies I can understand you I’ve been walking with Sadie
I saw female deer I love to take a bike I go to the gas station

I mow this morning

26

What follows is the original MIT hierarchy from level I to level IV (see Appendix
2)(Sparks & Deck, 1986), and also the subject L’s MIT description for a non-fluent

aphasia.

MIT Hierarchy Level I
The goal of the first level is the establishment of the client’s comfortable

adaptation to the technique by learning intoning, hand tapping, and response to hand-
signal controls of the therapist. The linguistic contents of level I were used from the
sample materials in level 11 depending on the subject’s comfortable adaptation. When the
researcher gave a stimulus by humming the sentence prosody and asking subject L to join
humming in unison, he produced some unintelligible output instead of simply humming,
but the researcher accepted his production as a humming response because the muscle
movement of his tongue and face was restricted for speaking due to the symptoms of
mild apraxia. For a convenient provision of the rhythmic stimuli, a castanet was used to
produce rhythmic beats instead of hand tapping. In the beginning of the session, subject L
utilized his more comfortable left-hand because he had a right-side weakness. On the first
day of the sessions, the researcher repeated this level as long as L felt comfortable
intoning, hand tapping, and responding to hand-signals.

Subject L took about 5 to 10 minutes in the beginning of MIT treatment, then
proceeded onto the next level as soon as he was comfortable with the first level goals. No

scoring took place at this level. See more detailed procedures in table 3-1.

27

Table 3-1. MIT Hierarchy Level I

 

LEVEL I — learn intoning and hand tapping, response to hand-signals

 

 

Stimulus (Therapist) (T) Hums melody twice with (HT) (HT) Hand Tapping
. Unison humming with (HT). Unison (Together)

Response (Client) (C) Fades participation but continues (HT)

Score and Progression No score. Proceed to next step 11.

MIT Hierarchy Level II

The goal of the second level is the aphasic subject’s immediate repetition of the
target intoned sentence and the response to a question. Level II is a four-step procedure.
He had remained at this level II protocol for thirteen sessions, and then proceeded to the
next level III for sessions fourteen to sixteen. Subject L spent 20 to 30 minutes for level 11
of MIT technique in each session. For more detailed information on each of the four steps,
scoring procedures are in Table 3-2.

The researcher needed to modify the melody pattern on the target sentence
whenever the subject was unable to produce verbal utterance of the melody. In other
words, once the researcher modified the melody pattern on the target sentence, the
change of melody stimulated the subject to produce clearer utterances. According to the
book, Rhythm, Music, and the Brain (Thaut, 2005), musical elements as reinforcement
are classified by three aspects; 1) Spatial (Pitch, Dynamic, Sound duration, Harmony); 2)
Temporal (Tempo, Meter, Rhythmic pattern, Form); and 3) Force cueing (Dynamic,
Harmony, Timbre, Tempo). This modification of melody pattern is included in the area of
spatial cueing, such as pitch. There is research evidence based on the effect of musical
cues for persons with aphasia (Cohen & Ford, 1995). Namely, when the researcher

modified the melody with high pitch intonation on the last word, subject L produced

28

clearer utterance, so the change of pitch stimulated his speech production. For example,
the melody for the phrase (go to sleep) was changed, so that the last word (sleep) had a
higher pitch than before. For more information about the modification of melody
patterns, refer MIT article (Sparks & Deck, 1986).

The linguistic contents of level II was selected from the contents of subject L’s
daily-based usage. The numbers of syllables were in the range of two to six syllables for
a sentence or a phrase. His first sentence was selected from the sentence with the most
usage, such as Take Sadie Walk, which was reduced from full sentences such as I love to
take Sadie out for a walk, depending on the level of his functional communication skills.
Six to eight sentences were used for this level II procedure from the sample materials for
linguistic contents (see Table 2).

There are five specific techniques of level II technique: 1) Hand Tapping; 2)
Control by hand signals; 3) Fading Participation; 4) Unison Repetition; and 5) Verbal

Cueing.

Hand Tapping with a Castanet

Rhythmic stimulus has proved to be an effective musical element to activate the
htunan nervous system through several decades (Sacks, 2006). The castanets were used to
stimulate the subject by rhythmic sound. While humming the melody, the researcher
started making the rhythmic beats with the castanet, and then asked the subject to join
making the rhythm first with the castanet alone and then in unison with the intoned
sentence with the castanet. Subject L used his more functional left hand for the castanet at
the beginning of the sessions, and the researcher also encouraged him to gradually use his

right hand depending on his adjustment. The researcher asked the subject to turn a little

29

bit in his chair, so that the performance was visible to each other, the researcher and the
subject.

Control by Hand Signals

Using a castanet for hand signals was an effective way to control the subject’s
response as a nonverbal means. It was used in many steps when the researcher asked the
subject to join in humming or intoning sentences by hand signals. It was also used to
guide the subject on when to take turns, such as the subject needed to listen or to take
turns with the researcher for the repetitions in step 3. Also the castanet was held up and
forward toward the subject to produce one strong rhythmic sound. This way the signals
were useful in enforcing latency and inserted delay for responding, if used consistently
(Sparks & Deck, 1986).

Fading Participation & Immediate Repetition

Level II of the hierarchy involved fading the participation of the researcher in step
2 and the immediate repetition by the subject after the researcher had just presented in
step 3. If the researcher faded her participation in order to allow the ‘solo’ performance,
subject L sometimes stopped producing the intoned sentence at early stages in sessions,
but he did gradually keep going to perform solo by making beats with his castanet. When
the subject stopped his vocal production, the researcher needed to rejoin in unison
because it was evident that he was not ready to proceed to step 3 of solo repetition on his
own.

Verbal Cuing

The use of verbal cuing is shown in step 3 and step 4 in Level II, and the cue has

an effect on counting scores (see Table 3-2). If he wasn’t able to repeat the intoned

30

sentence in step 3 as a solo, the researcher gave verbal cuing of the initiation of the

utterance in the target sentence. L needed the cuing in most sentences in steps 3 and 4.

Table 3-2

MIT Hierarchy Level II

 

LEVEL 2 — Ability repeating intoned sentences immediately,

Response to a question

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

1.Hums melody with (HT),
2.1ntones sentence with (HT),
3.Signals (C) to join in unison intoning of sentence

Intone sentence together with (HT) - Unison

Acceptable- 1 point. Proceed to Step 2, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

Same as Step 1
But fades participation except (HT)

Intone sentence with (HT) gradually Solo

Acceptable- 1 point. Proceed to Step 3, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

1.Signals (C) to listen

2.1ntones sentence with (HT)

3.Signals (C) to repeat,

Cueing for initiation of utterance if necessary

Repeat intoned sentence with (HT) immediately

Acceptable— 2 point without cue, 1 point with cue
Proceed to step 4, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

1.1ntones question, “What did you say?”
2.Signal (C) to repeat
Cueing for initiation of response if necessary

Repeat intoned sentence with (HT)

Acceptable— 2 point without cue, 1 point with one
Proceed to step 1, for next sentence
Unacceptable - Discontinue progress for sentence.

 

(Table.3-2. MIT Hierarchy Level 11)

31

MIT Hierarchy Level III
The goal of the third level is the aphasic’s ability to delay responses and respond
to less specific questions. Through this level of training, the subject was able to respond

with the intoned sentence even if the researcher signaled to wait for 2 to 3 seconds.

Backup
The important technique, Backup, was introduced in level III and IV, which was

the means of attempting indirect correction of errors by repeating the previous step. This
manner of correcting errors is meaningful, as the subject may or may not be aware of the
purpose of this procedure, but his failure is not drawn to his attention (Sparks & Deck,
1986). If the subject presented an unacceptable response in step 2, he was guided to
return to the previous step of listening to the intoned sentence and following in unison,
followed by a retrial of the repetition after a l or 2 second delay. This backup procedure
was used in level III and IV.

There are three steps in the procedures for level III. The subject received this level
III protocol from the 14th to l6tb sessions and was unable to precede to the last level of
MIT. L took 20 to 30 minutes for the level HI techniques in each session. He showed
comfortable adaptation in intoning sentences with gradual solo performance after the
researcher removed her unison participation, but he was challenged by repeating intoned
sentence after the 1 or 2 second delay in step 2. For more detailed information on each of

the three steps in Level III procedures and scoring are in Table 3-3.

32

Table 3-3. MIT Hierarchy Level III

 

LEVEL 3 — Ability to the delay of responses, Response to specific questions

 

Stimulus (Therapist) (T)

Step Response (Client) (C)

Score and Progression

1.1ntones sentence with (HT),
2.Signals (C) to join in unison sentence
3. Fade participation except (HT)

Intone sentence with (HT) gradually Solo

Acceptable- 1 point. Proceed to Step 2, same sentence
Unacceptable - Discontinue progress for sentence.

 

Stimulus (Therapist) (T)

Step Response (Client) (C)

Score and Progression

1.Signals (C) to listen

2.1ntones sentence with (HT)

3.Signals (C) to repeat after 1 or 2 second Delay
4. If (C) fails, Backup step 1 (B) Back up

Repeat Intoned sentence with (HT) afier Delay
If fail, backup step 1 and retrial

Acceptable- 2 point without (B), 1 Point with (B)
Process to Step 3, same sentence
Unacceptable - Discontinue progress for sentence.

 

Stimulus (Therapist) (T)

Step Response (Client) (C)

Score and Progression

l.Intone a related question (e. g. what kind of, How many)
2.Signals (C) to answer
3. If (C) fails, Backup step 2 (B) Back up

Appropriate Answer (Intoned or Spoken)
If fail, backup step 2 and retrial

Acceptable- 2 point without (B), 1 Point with (B)
Process to Step 1 for next sentence
Unacceptable - Discontinue progress for sentence.

 

MIT Hierarchy Level IV

(Table.3-3. MIT Hierarchy Level 111)

The goal of this last level of MIT is the aphasic’s ability to return to normal

speech prosody, which is facilitated by a technique called Sprechgesang (speech-song). In

this technique the melodic line remained the same as the intoned sentence of the

preceding step, but the pitches of intoned words were replaced by the variable pitch of

natural speech, retaining the tempo, rhythm, and stress of the target sentences. More

33

complex and longer sentences are used as the target materials in this level. There are four

procedural steps in level IV, but subject L was unable to precede to this level. For more

detailed information on each four-step procedure and scoring, see Table 3-4. More

descriptive explanations of MIT procedures can be found in the literature (Sparks &

Deck, 1986).

Table 3-4. MIT Hierarchy Level IV

 

Stimulus (Therapist) (T)

Step
1 Response (Client) (C)

Score and Progression

LEVEL 4 — Ability of Normal Speech

1.Signal (C) to listen

2.1ntone sentence with (HT)

3.Present it twice in Sprechgesang

4.Signal (C) unison Sprechgesang of sentence (HT)
5. If (C) fails, Backup to present in Sprechgesang

6. Retrial No. 4

Sprechgesang of sentence with (HT)

Acceptable- 2 point with Sprechgesang.

- 1 point with Backup

- Proceed to Step 2, same sentence
Unacceptable - Discontinue progress for sentence.

 

Stimulus (Therapist) (T)

Step Response (Client) (C)

Score and Progression

1.Signal (C) to listen

2.Present sentence in Sprechgesang with (HT)
3.Signals (C) to repeat after 2 or 3 second Delay
4. If (C) fails, Backup to Step 1 same sentence

Sprechgesang of sentence with (HT) after 2 or 3 second Delay.
Acceptable- 2 point without Backup
- 1 point with Backup

- Proceed to Step 3, same sentence
Unacceptable - Discontinue progress for sentence.

 

(Continued)

34

Table 3-4. MIT Hierarchy Level IV (Continued)

 

Stimulus (Therapist) (T)

Step Response (Client) (C)
3

Score and Progression

1.Signal (C) to listen

2.Present sentence once in Sprechgesang

3.Present sentence twice in Normal Speech Prosody
4.Signals (C) to repeat after 2 or 3 second Delay

5.If (C) fails, Backup to Step 2 same sentence

Repeat sentence in Normal Speech Prosody
after2 or 3 second Delay, no (HT)

Acceptable- 2 point without Backup

- 1 point with Backup

- Proceed to Step 4, same sentence
Unacceptable - Discontinue progress for sentence.

 

Stimulus (Therapist) (T)

Step

4 Response (Client) (C)

Score and Progression

1.Ask question about substantive information on the same
sentence.

2. If fail, Backup step 3 same sentence and retrial

3. Ask question about associative information

4. If fail, No backup and retrial if fail one.

Answer appropriate responses

3 point — bonus point, one or more associative responses
2 point — without Backup, substantive content

1 point — with Backup.

Proceed to next sentence.

 

Seven Principles

There are seven important principles of speech-language therapy involved in MIT

that the researcher should remember while leading MIT hierarchy (Sparks & Deck,

1986): 1) gradual progression of increased length and difficulty of task; 2) indirect

attempts to correct by the ‘backup’ technique; 3) repetition; 4) timing and controlled

latency; 5) avoidance of practice effect; 6) attention to the purpose of the subject’s verbal

utterance; 7) frequency of sessions.

35

2) Therapeutic Singing (TS)

According to Thaut’s book (2005), Therapeutic Singing (TS) is a reinforcing and
supporting technique for speech and language development, used with a variety of
neurological or developmental speech and language dysfunctions. This technique is able
to be used for a broad range of therapeutic goals, and also to support the goals of other
specific therapeutic techniques, such as Melodic Intonation Therapy (MIT), Vocal
Intonation Therapy (V IT), musical speech stimulation (MUSTIM), and so on. Modified
Melodic Intonation Therapy (MMIT) is the modified technique of MIT by combining the
Therapeutic Singing technique in MIT procedure.

The Therapeutic Singing (TS) technique was added to subject L’s intervention
before conducting MIT in the second period of sessions, from the 9th to the 16’” Subject
R, who was in high level of functional communication skill, utilized only TS without
MIT. There are different hierarchies of treatment outlined by the researcher, depending

on the level of speech and language for each subjects L and R.

Therapeutic Singing Hierarchy [or subject L

1. Stimulate memory function
- Therapist (T) play a familiar song with keyboard
- Client (C) listen to the song
- Acceptable — (C) respond to the song as familiar
- Unacceptable — (C) don’t respond at all. Use next song

2. Initiation of Utterance
- (T) sing the song by accompanying on the keyboard.
- (T) invite (C) unison singing to the song.
- Acceptable — (C) produce utterances (unclear, humming acceptable)
- Unacceptable — (C) don’t respond at all. Retrial step 1.

3. Adjustment of syllables and Repetition
- (T) sing in unison by adjusting to a comfortable beat of song for (C)
- (T) repeat the song twice in unison
- (C) sing the song in unison with 20% or more accurate syllables produced.
- Acceptable - (Full participation required, unclear production acceptable)

36

- Unacceptable - (C) did not respond at all. Retrial step 2.
- If Second retrial fails, proceed to next song material

4. Fade (T) participation
- (T) withdraw participation and encourage (C) to produce utterances
- (T) check (C) accuracy of production while (C) sing SOLO
- Acceptable -— (C) sing alone with 20 % or more production of lyrics
- Unacceptable — (C) terminate participation, Retrial step 3.
5. Correct with articulated speech
- (T) distinguish the unclear errors between consonant and vowel errors
- (T) ask (C) to imitate the oral movement of intoned vowel sounds on unclear words
- (T) ask (C) to imitate the oral movement of consonant sounds on unclear words
- (C) correct his utterance of words by training the oral motor movement
6. Unison again
— (T) encourage (C) to sing in unison again with instrumental accompaniment
This hierarchy was designed by the researcher and focused on the goals based
on his weaknesses in speech skills. Because of the symptoms of acute expressive aphasia
and mild apraxia, two primary goals were founded; l) to increase verbal utterance
through the song lyrics; 2) to improve the accurate speech articulation by focusing on a
number of syllable, consonant, and vowel sounds. This procedure was generally utilized
for subject L’s TS based on his goals. An average of four songs were used in each
session, and he had the repeated usage of song materials with one or two new songs
added in each session. As a result, subject L increased the production of a number of
words through song lyrics and also gradually improved the accuracy of verbal utterances
of syllables, consonant and vowel sounds. By repeating the song materials from sessions
the 9th to the 16th, he accomplished 100% of word production with accurate verbal
syllables and vowel sounds in lyrics, except for accuracy of consonant sounds. His

improvement was usually ordered: first, gradually producing the accurate number of

syllables by singing on vowel sounds. Second, gradually improving the clarity of vowel

37

sounds in matching rhythm and beat of the target song materials, and finally, producing
the exact consonant sounds. While he participated in this hierarchy, his slurred speech
was greatly reduced.

The list of song selections was taken from the books: The Great Family
Songbook (W eisssman, 2007); America’s All-Time Favorite Songs (Stone, 2009); The
Best Songs Ever (Leonard, 1985); Rise Up Singing (Patterson, 2004). The song materials
were given to subject L to assist in making his own song profiles, which were used for
the second period of the sessions. Songs were listed such as Home on the Range, God
Bless America, My Country ’ T is of Thee (America), If You 're Happy, I ’ve Been Working

on the Railroad, MyBonny Lies Over the Ocean, etc.

Therapeutic Singing Hierarchy (or subject R

1.Reading lyrics
- Therapist (T) ask Client (C) to read the lyric on the song sheet.
- Client (C) read the lyrics on the sheet.
- Acceptable — (C) reads the lyrics with (T)’s prompts (Verbal or Physical)
- Unacceptable -— (C) stops to read with frustration

2.Listening Stimulation
- (T) ask (C) to listen to the song with sight-reading by matching
the rhythmic beat with castanets simultaneously.
- (T) make the rhythmic beat with a castanet while listening.
- (C) listen to the song with sight-reading and playing the castanet simultaneously.
- Acceptable — (C) listening with sight-reading and making rhythmic beats
with the castanet.
- Unacceptable — (C) don’t respond with the sight-reading and making beats

3.Adjustment to the song
- repeat #. 2 '
- (T) focus on (C) verbal outputs

4.Unison singing with accompaniment on the keyboard
- (T) encourage (C) unison singing by accompanying on the keyboard.
- (C) sing in unison with sight-reading and matching the rhythmic beat
with the castanet

38

- Acceptable — (C) sings the lyric with 80% of accuracy
- Unacceptable — (C) shows frustration and stops singing

5.Fade (T) participation

- (T) withdraw participation and encourage (C) to sing alone

- (T) check (C) accuracy of production while (C) sings SOLO

- (C) sing alone by matching the rhythmic beat of the song with the castanet

- Acceptable — (C) sings alone with 80 % or more accuracy of lyrics

- Unacceptable - (C) terminates participation

This hierarchy was designed by the researcher and focused on the goals based

on his areas of weakness in both speech fimctions and other dysfunctional areas. Because
of his loss of range of vision, he had feelings of frustration whenever he tried to do any
reading. Two primary goals were established in the TS intervention: 1) to improve the
reading skill without frustration; 2) to improve accurate speech articulation by matching
the right rhythmic beats of the song. This procedure was generally utilized for subject R’s
TS intervention based on his goals. An average of four songs was used in each session,
along with repeated usage of the same song materials. Several new songs were gradually
added. As much as repeating the materials were important, steps 1 and 2 were skipped
and the intervention procedure and time consumption for the song material was reduced.
If new song material was added, the new material started from step 1. As a result, subject
R improved the reading skills without having frustration. Through repeating the song
materials from sessions 1 to 16, be accomplished steady reading skills without any
prompts or terminations while performing the song materials. While he participated in
this hierarchy of treatment, his slow speech with unmatched rhythmic beats and his
reading failures were reduced.

Subject R’s song materials were collected from the Beatles CD Album to make

his own song profile, which were used for the full period of sixteen sessions. Songs were

39

listed such as Let it Be (Beatles), A Hard Days Night (Beatles), A Day in The Life
(Beatles), Venus (Frankie Avalon), Look What You ’ve Done to Me (Boz Scaggs), Imagine
(John Lennon), Woman (John Lennon), Hey, Jude (John Lennon), Yesterday (John

Lennon), Everybody Talking (John Lennon), and so on.

WI—m—WM

There were two measuring procedures used for data collections: Functional
Communication Measure (F CM) used in both the pre-tests and post-tests and Melodic
Intonation Therapy Measure (MITM). On the pre and post-test, F CM was used to
examine whether two musical interventions (TS & MIT) had effectiveness in improving
functional communication skills. FCM was composed of two different parts: Spontaneous
Speech Measurement (SPM) and Naming Measurement (NM). See the measurement 1
forms in Appendix 4 (SPM) and Appendix 6 (NM).

According to aphasia assessment (Lezak, et al., 1995), there are six examination
areas that are aspects of verbal behavior: 1) Spontaneous Speech; 2) Repetition of words;
3) Speech Comprehension; 4) Naming; 5) Reading; and 6) Writing. A review of
language and speech functions indicates whether communication problems are present in
those six areas (Lezak, et al., 1995). In order to reduce the measuring spans during each
session, two examinations, l) Spontaneous Speech and 2) Naming, were chosen from
among the six examination areas above to make up the Functional Communication

Measurement (FCM).

4o

Spontaneous Speech Measurement & Naming Measurement in F CM

In the Spontaneous Speech Measurement, five variables were collected through
each pre-test and post-test: 1) Total Length of Samples; 2) Total Number of Utterances;
3) Total Number of Words; 4) Average length of Utterances; and 5) Average Number of
Words per Minute. For more information about the Scoring Hierarchy of the Spontaneous
Speech, see Table 4-1. The video recorded data was analyzed by utilizing the i-movie
program in the Macintosh notebook for the 2 to 3 hours of time used in each session.
Before the pre-test, subjects were asked to choose the questions they would like to
answer, but the set time to answer was no longer than 13 minutes. The same
questionnaire was conducted in both the pre-test and post-test. The list of questions was
made from the materials in the Aphasia Needs Assessment (Garrett & Beukelman, 1995)

and the Aphasia Diagnostic Profiles questions (ADP)(Helm-Estabrooks, 1992).

Table 4-1. Scoring Hierarchy at Spontaneous Speech Measurement

 

Scoring Hierarchy of Spontaneous Speech Measurement in F CM

 

1. Total Length of Samples Sum of Seconds, including both intelligible and unintelligible
words
The unspoken times were deleted after 10 second without subject’s
responses.

2. Total Number of Utterances Including paraphasic errors
The effort in seconds of words produced was included in the
utterances. If the subject stops to speak by saying ‘1 don’t know’,
then the utterance was
ended as counting an utterance.

3. Total Number of Words Counting Total Number of Words produced for 13 minutes.
2-1. Average Length of Average number of words for three longest intelligible phrases
Utterances e.g.) l. I like a ball (4 words)

* .me too (2 words)

2. I want to go home (5 words)
3. I love peanut (3 words)

*. Lovely (l word)

= 4+5+3/3 = 4 avera e len of utterances

3-1. Average Number of Words Total number of words are divided the average length of samples.
per Minute It means the Speed of SpeakirE

41

On the other hand, the Naming Test content was composed of the picture stimulus
sheets, which was edited from the picture materials in the Boston Naming Test (Kaplan,
Goodglass, & Weintraub, 2001), the Aphasia Diagnostic Profiles (ADP) (Helm-
Estabrooks, 1992), and the Test of Adolescent/Adult Word Finding (TAWF) (German,
1992). A total of 90 pictures were selected from the references with the assistance of a
speech and language pathologist. A total of sixteen different picture sets were
manipulated by randomizing the order of ninety pictures, which was made by using the
PowerPoint slideshow program in the Macintosh Notebook. A total of five minutes was
given to the test, and the same ordered set of pictures was used both pre and post-test in
each session. In the naming test, two variables were measured: 1) Number of items
named by subjects; 2) Average Number of prompts per item named. Subjects L and R

each had a different scoring hierarchy, according to the level of speaking. See table 4-2.

Table 4-2. Scoring Hierarchy at Naming measurement

 

Scoring Hierarchy of Naming Measurement in F CM

 

1. Number of Items * Subject L —- Counted the total number of items named, including writing
named by subjects letters
* Subject R — Counted the total number of items named with verbal speaking.

2. Average Number of Counted the total number of prompts given by the researcher, then divided into
Prompts per item #.1 above (Total number of Items named)
named "‘ Subject L - Included the prompts in Appendix 7 (Hierarchy of Word
Retrieval Prompts for the Client with Aphasia) and written prompts by
correction of the letter errors.
Unacceptable — if no response after four prompts given, pass to the next

picture.

* Subject R — Included the verbal prompts in Appendix 7
Unacceptable - If no response after 5 seconds, given a prompt, then if no
response, given another prompt, then after 5 seconds, pass to the next picture

On the other hand, MIT measure was conducted while the subject was

participating in MIT treatment procedure. There are brief guidelines for the scoring

42

hierarchy in Table 4-3 to illustrate management of errors in the subject’s response

(Sparks & Deck, 1986).

Table 4-3. Scoring Hierarchy at MIT Measurement

 

Scoring Hierarchy of Melodic Intonation Therapy (M11) Measurement

 

 

Level 11

Scoring Examples description Step 1 Step 2 Step 3 Step 4
First sentence: Subject succeeds in all steps. Maximum scores
attained l l 2 2
Second Sentence: Succeeds in all steps but requires a cue to initiate
response in Steps 3 and 4. 1 1 1 1
Third Sentence: Succeeds in Steps 1 and 2, requires a cue to initiate
response in Step 3, and fails Step 4 because of an unacceptable
response 1 l 1 0
after backup
Fourth Sentence: Succeeds in Steps 1 and 2, requires a cue to initiate
response in Step 3, and requires a backup to initiate Step 4. 1 l l 1

Fifth Sentence: Succeeds in Step 1, fails in Step 2. Progression

stopped 1 0 _ -
and no scores given for Steps 3 and 4.

 

Total 18/24 (75%) 5/5 4/5 5/8 4/6

 

 

 

(Continued)
Level III
Scoring Examples description Step 1 Step 2 Step 3

First sentence: Subject succeeds in all steps. Maximum scores attained 1 2 2
Second Sentence: Succeeds in all steps but requires backups for Steps 2

l 1 l
and 3.
Third Sentence: Succeeds in Steps 1, requires a backup to initiate Step 2,
Succeeds in Step 3 without backup. I l 2

Fourth Sentence: Succeeds in Steps 1, requires a backup in Step 2 because
Of a a a a l l 0
inaccurate response, fails to initiate response in step 3 after a backup.

Fifih Sentence: Succeeds in Step 1, fails to repeat accurately in step 2, and
fails
again after a backup. Progression stopped and no score may be given for l 0 -

step 3.

 

Total 15/23 (65%) 5/5 5/ 10 5/8
(Table 4-3 — Scoring Hierarchy of MIT Measurement (Sparks & Deck, 1986)

43

Table 4-3. Scoring Hierarchy at MIT Measurement (Continued)

 

Level IV

Scoring Examples description Step 1 Step 2

Step 3 Step 4

 

First sentence: Succeeds in Steps 1, 2, and 3, requires a backup for
one 2 2
specific question in step 4, but answers all associative questions.

Second Sentence: Succeeds in Steps 1, 2, and 3, requires a backup
for Step 2 2
4, fails to answer any associative question.

Third Sentence: Succeeds in Steps 1 and 2, requires a backup to

initiate normal prosody in Step 3. Succeeds in Step 4 but no bonus

because of 2 2
failure on last associative question.

Fourth Sentence: Succeeds in Step 1, requires backups for Steps 2
and 3. 2 1
Succeeds with bonus in Step 4.

Fifth Sentence: Requires a backup to succeed in Step 1 and then
succeed . 1 2
in subsequent steps.

2 2

 

Total 37/45 (82%) 9/10 9/10

8/10 ”/15

(Table 4-3 — Scoring Hierarchy of MIT Measurement (Sparks & Deck, 1986)

44

CHAPTER IV

RESULTS

Findings and Analysis

Through the Functional Communication Measurement (FCM) and Melodic
Intonation Therapy Measurement (MITM), data on eight variables were collected. There
are two different measurements in the Functional Communication Measurement; First,
Spontaneous Speech Measurement; second, Naming Measurement. Five variables were
collected in the Spontaneous Speech Measurement (SPM) and two variables were
collected in in the Naming Measurement (NM), and one variable was collected in the
Melodic Intonation Therapy Measurement (MITM).

There were four problems for subject L’s case study: (1) to investigate whether
Therapeutic Singing (TS) combined with Melodic Intonation Therapy (MIT) helps adult
clients who are diagnosed with expressive aphasia as measured by the quality of their
fimctional communication skills; (2) to investigate whether MIT treatment itself helps
clients improve the quality of their functional communication skills; (3) to compare the
effectiveness of TS combined with MIT and MIT alone in increasing functional
communication skills; and (4) to determine whether TS enhances the benefits of MIT
treatment.

On the other hand, there was only one goal for the subject R’s case study; 1) to
investigate whether Therapeutic Singing (TS) itself helps clients improve the quality of

their functional communication skills.

45

In order to select an appropriate statistical test in these two case studies,
descriptive statistics was conducted after collecting the data on the eight variables. There
are two methods of evaluating the data: first, the analysis of the short-term effect and
second, the analysis of a long-term effect of the interventions. In order to prove the short-
terrn effect, paired t-tests were conducted to compare and contrast the differences
between pre and post-test variables. The subjects received the pre-test, interventions, and
then post-test in each session, so this t-test could indicate the short-term effect of the
intervention by comparing pre and post-test data.

Second, a long-term effect of intervention was evaluated by comparing the means
of eight variables. The long-term effect is shown by whether or not the means of second-
half sessions are higher than the means of first-half sessions. The first—half data
collections from the 1St to the 8th sessions were calculated as the means for the first-half,

and the means for the second-half data were calculated from the 9’h to the 16th sessions.

Spontaneous Speech Measurement

The results for the Spontanous Speech Measurement were determined by
collecting data on five different variables: 1) Total Length of Samples (See Figure 3-1
and 3-2); 2) Total Number of Utterances (See Figure 3-3 and 3-4); 3) Total Number of
Words (See Figure 3-5 and 3-6); 4) Average Length of Utterance (See Figure 3-7 and 3-
8); 5) Average Number of Words per Minute (See Figure 3-9 and 3-10). For more

information about raw data, See Appendix 14-1 and 14-2.

46

. Total Length of Samples

 

L's Spontaneous Speech Test
1. Total Length of Samples

 

H
N

 

H
O

 

A.

 

 

 

 

 

Total Length of Sample

QN-Fa\m

 

 

I T I I T T T r I

I I I I T 1

1 2 3 4 5 6 7 8 9 10111213141516

Sessions

"' 1-pre +1-post

 

 

(Figure 3-1. Total Length of Sample in Us Spontaneous Speech Test)

 

 

 

 

Paired Samples Statistics
Std. Error
Mean 11 Std. Deviation llean
Pair Pre_TesLl 6.7114 16 152915 .38229
1 Post'l‘estl 8.1625 16 181219 15305

 

 

 

 

 

(Table 5-1. L’s Paired t-test on Total Length of Samples)

Paired Samples Test

Figure 3-1 shows the
increasing tendency in post-
test data for subject L’s total
length of samples. There was
a long-term effect of the
intervention demonstrated by
comparing the means
between the first-half and the

second-half sessions.

 

Paired Differences

 

Mean

Std. Error

95% Confidence
Interval of the
Difference

 

Std. Deviation llean Lower

Upper

1 d1 Sig. (2-tailedl

 

Pair Pre_TesL1 -
l PosLTesLl

 

 

-l.45111

164604 .41151 -2.32822

 

 

 

 

-.57399

-3.526 15 .003

 

 

 

 

 

(Table 5-2. L’s Paired t-test on Total Length of Samples)

The mean for the second-half sessions (the 9’“ — the 16’“) on the pre-test (M=7.6)

was higher than the mean for the first-half sessions (the 1S” — the 8th) on the pre-test

(M=5.8). Also, the mean of the post-tests of the second-half sessions (M=9.4) was higher

47

than the mean of the post-tests of first-half sessions (M=6.9). It indicated that the total
length of samples increased on the second-half sessions. Therefore, it is shown that the
intervention of MIT and MIT combined with TS had a positive effect on the
improvement of total length of samples as a long-term effect.

In order to investigate whether there was the short-term effect on the sample
length, a paired-samples t-test was conducted to compare pre-test and post-test from the
first session to the sixteenth session. There was a statistically significant difference in the
scores for pre-test (M=6.71, SD=1.52) and post-test (M=8.16, SD=1.82); t(15)=-3.526, p
= 0.003 (p<0.05) (See Table 5-1 & 5-2). These results suggest that each intervention does
have an effect on the difference between pre and post-test. Therefore, the results indicate
that with both the interventions of MIT and MIT combined with TS, the total length of

samples increased.

 

Fi e 3-2 shows there was
R's Spontaneous Speech Test gm

 

 

 

 

 

 

1. Total Length of Samples the tendency of slightly

v, 12
d) ' ' -
2' 10 higher pornts on post tests
in: 8 than on the pre-tests in the
o

6
El: R’s total length of samples.
5 4
0
:3: 2 There was a slight difference
a 0 I I I I l I I I T I T 1 l 1

 

 

‘ ‘ on R’s long-term effect of
1 2 3 4 5 6 7 8 910111213141516

Sessions intervention by comparing

"A'- 1-pre +1-post

 

 

the mean of the first-half and

 

(Figure 3-2. Total Length of Sample in R’s Spontaneous Speech Test)
the second-half of sessions.

48

 

Paired Samples Statistics

 

 

 

Std. Error

Mean ll Std. Deviation Mean

Wesu 6.1885 18 139134 .34784
1 PosLTesLl 7.4896 16 1.58987 .3974?

 

 

 

 

 

 

(Table 5-3. R’s Paired t-test on Total Length of Samples)

Paired Samples Test

 

Paired Differences

 

95% Confidence
Interval of the

Std. Error Difference
, llean Std. Deviation llean Lover Upper t d! Sig. (2-tailed)
Pair Pre_TesL1 -

1 PosLTesLl -1.30104 2.03471 .50868 -2.38526 -.21682 -2.558 15 .022

 

 

 

 

 

 

 

 

 

 

 

 

(Table 5-4. R’s Paired t-test on Total Length of Samples)

The mean of second-half sessions on the pre-test (M=6.02) was slightly lower
than the mean of first-half sessions on the pre-test (M: 6.36). There were no
improvements on the pre-tests. On the other hand, the mean of the post-tests of the
second-half sessions (M=7.74) was slightly higher than the post-test mean of first-half
sessions (M=7.23). It indicates that the total length of samples increased during the
second-half sessions on only the post-tests. Therefore, it is shown that the intervention of
TS had a positive effect, though less than the L’s improvement of total length of samples,
as a long-term effect.

In order to investigate whether there was a short-term effect on the sample length,
a paired-sample t-test was conducted to compare pre-test and post-test from the first
session to sixteenth session. There was a statistically significant difference in the scores
for pre-test (M=6. l8, SD=1.39) and post-test (M=7.48, SD=1.58); t (15)=-2.558, p =

0.022 (p<0.05) (See Table 5-3 & 5-4). These results suggest that each intervention does

49

 

have an effect on the difference between pre and post-tests. Therefore, the total length of

samples increased through the intervention of TS alone.

2. Total Number of Utterances

 

L's Spontaneous Speech Test

 

8 2. Total Number of Utterances
:40

E35

0

3:30

225

220

315

S10

2 5

E o

13 1 2 3 4 s 6 7 8 9 10111213141516

Sessions
'4' Z-pre +2-post

 

 

 

(Figure 3-3. Total Number of Utterances in Us Spontaneous Speech Test)

 

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean— ll Std. Deviatifl Mean
Pair Pre_Test_2 18.1260 16 3.44238 .86060
1 PosLTesL2 23.5625 16 6.81145 1.70286

 

 

(Table 6-1. L’s Paired t-test on Total Number of Utterances)

Figure 3-3 shows the

increasing tendency in post-

test data in L’s total number

of utterances. The long-terrn

effect of the interventions

was shown through

comparing the means

between the first-half and the

second-half sessions.

 

 

 

 

 

 

 

Paired Samples Test
Paired [finances
95% Confidence
Interval ofthe
Std. Error Diffusin—
Mean Std. Deviation Mean Lower Upper t df Sig. (2-tailedl
Pair Pre.Test_2- _
1 PosLTesL? '5.43750 7.00446 1.75112 -9.16991 -l.70509 '3.105 15 .007

 

 

 

 

 

 

 

(Table 6-2. L’s Paired t-test on Total Number of Utterances)

The mean. for second-half sessions on the pre-test (M=l 7) was lower than the
mean for first-half sessions (M=19.25). There was no improvement on the pre-test. On
the other hand, the mean of the post-tests of second-half sessions (M=24.12) was higher
than the post-test mean of first-half sessions (M=23). It indicates that the total number of
utterances increased on the second-half sessions on only the post-test. Therefore, it is
shown that the intervention of MIT and MIT combined with TS had a positive effect on
the improvement of the total number of utterances on the post-tests as a long-term effect.

In order to investigate whether there was the short-term effect on number of
utterances, a paired-samples t-test was conducted to compare pre-tests and post-tests from
first session to sixteenth session. There was a statistically significant difference in the
scores for pre-test (M=l8. 12, SD=3 .44) and post-test CM=23.56, SD=6.81); t(15)=-3.105,
p = 0.007 (p<0.05) These results suggested that there was a short-term effect of the
interventions shown by the difference between pre and post-tests data. Therefore, the
results suggested that when MIT or MIT combined with TS were conducted, the total

number of utterances increased on the post-tests.

51

 

Total Number of Utterances

 

Sessions

R's Spontaneous Speech Test
2. Total Number of Utterances

 

1 2 3 4 5 6 7 8 910111213141516

'i- 2-pre + 2-post

 

 

(Figure 3-4. Total Number of Utterances in R’s Spontaneous Speech Test)

Paired Samples Statistics

 

Mean

Std. Deviation Mean

Std. Error

 

 

 

 

 

Pair Pre_TesL2 12W) 16 250—768 .65192
1 PosLTesLZ 14.0000 16 3.74166 .93541

 

 

 

(Table 6-3. R's Paired t-test on Total Number of Utterances)

Paired Samples Test

Figure 3—4 shows the
slightly increasing
tendency in post-test
data for R’s total
number of utterances.
There was a long-term
effect in the
interventions by
comparing the means
between the first-half
and the second-half

sessions.

 

Paired Differences

 

Mean

Std. Deviation

Std. Error

95% Confidence
Interval of the
Difference

 

Mean

Lower Upper

t df Sig. t2-tailedl

 

Pair Pre_TesL2 -
1 PosLTesLZ

 

 

-1.50000

 

3.65148

 

.9128? -3.44674 .44574

 

 

 

-1.643 15 .121

 

 

 

 

(Table 6—4. R‘s Paired t-test on Total Number of Utterances)

The mean for the second-half sessions (the 9m_ the 16*) on the pre-test (M=12.63)

was slightly higher than the mean for the first-half sessions (the 1" — the 8m) on the pre-

test (M=12.38). Also, the post-test mean for the second-half sessions (M=15.88) was

52

notably higher than the post-test mean for the first-half sessions (M=12.16). It indicated
that the total number of utterances increased during the second-half sessions both pre and
post-tests. Therefore, it is shown that the intervention of the TS had a positive effect on
the improvement of total number of utterances as a long-term efi‘ect.

In order to investigate whether there was the short-term effect on the number of
utterances, a paired-samples t-test was conducted to compare the pre-tests and the post-
tests fiom the first to the sixteenth sessions. There was no statistically significant
differences in the scores for pre-test (M=12.5, SD=2.6) and post-test (M=l4, SD=3.74);
t(15)= -l.643, p = 0.121 (p>0.05) (See Table 6-3 & 6-4). Although there was a slight
difference between the means (pre = 12.5 and post = 14), this result indicated that the
intervention did not result in a statistical difference between pre and post-test. Therefore,
the result showed that the total number of utterances did not significantly increase

through the intervention.

3. Total Number of Words

 

' Figure 3-7 shows an
L s Spontaneous Speech Test

3. Total Number of Words increasing tendency in
u, 800
E 700 L’s total number of
3 600
“5 500 words. The long-tenn
l-
0 400
.9
S 300 effect of the
E 200 intervention was shown
3 100
o
[- 0

 

through comparing the

1 2 3 4 5 6 7 8 910111213141516

Sessions means between the
'i- 3-pre +3-post

 

 

 

(Figure 3-5. Total Number of Words in L’s SP Test)

53

first-half and the second-half sessions.

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean ll Std. Deviation Mean
Pair Pre_TesL3 304.3750 16 15331139 38.32785
1 PosLTesL3 375.5000 16 106.62082 26.65521

 

 

 

(Table 7-1. L’s Paired t-test on Total Number of Words)

 

 

 

 

 

 

 

 

 

 

 

 

 

Paired Samples Test
Paired Differences
95% Confidence
Interval of the
Std. Error Difference
Mean Std. Deviation Mean Lower Upper 1 df Sig. (2-tailed)
Pair Pre_TesL3 -
1 PosLTesLS 71.12500 110.91310 27,72828-13022642 -12.02358 -2.565 15 .022

 

(Table 7-2. L’s Paired t-test on Total Number of Words)

The mean for the second half-sessions on the pre-test (M=354.6) was higher than

the mean for the first-half sessions on the pre-test (M=254). Also, the post-test mean for

the second-half sessions (M=413.7) was higher than the post-test mean for the first-half

sessions (M= 337.2). It is indicated that the total number of words increased through the

second-half sessions for both the pre and the post-tests. Therefore, it is shown that the

interventions of MIT and MIT combined with TS have a positive effect on the

improvement of total number of words, as a long-term effect.

In order to investigate whether there was a short-term effect on total number of

words, a paired-samples t-test was conducted to compare pre-test and post-test data from

the first session to the sixteenth session. There was a statistically significant difference in

the scores for pre-test (M=304, SD=153) and post-test (M=375, SD=106) : t(15)= -2.565,

p = 0.022 (p<0.05) (See Table 7-1 & 7-2). These results indicated that each intervention

54

 

does have an effect on the difference between the pre and the post-test scores Therefore,

the total number of words increased through the interventions of MIT and MIT combined

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

with TS.
R's Spontaneous Speech Test _
3. Total Number of Words F‘gure 3'8 Shows a
500 - - -
_§ 450 slight increasrng
O 400 ' a
E 350 tendency in R 5 total
9 300
S 250 number of words. The
'2 200
:I 150 long-term effect of the
_Z_ 100
g 50 intervention was shown
[_ 0
1 2 3 4 5 6 7 8 9 10111213141516 through comparing the
Sessions
_‘__ 3-pre I 3-post means between the
(Figure 3—6. Total Number of Words in R’s SP Test) first-half and the
Paired Samples Statistics second-half sessrons.
Std. Error
Mean ll Std, Deviation Mean
Pair Pre_TesL3 241.6250 16 76.98474 19.24618
1 PosLTesL3 280.6250 16 68.67981 17.16995
(Table 7-3. R’s Paired t-test on Total Number of Words)
Paired Samples Test
Paired Differences
95% Confidence
Interval of the
Std. Error Difference
Mean Std. Deviation Mean Lover Upper t df Sig. t2-tailed)
Pair Pre.TesL3 - L - . .
1 P MU est3 3900009 | 88.09843 22.02461 r85,94434 7.94434 1.741 15 .094

 

 

 

 

 

 

 

 

 

 

(Table 7-4. R’s Paired t-test on Total Number of Words)

The mean for the second-half sessions on the pre-test (M=255) was higher than
the mean for the first-half sessions (M=228). Also, the mean for the second-half sessions
(M=322.6) was higher than the mean for the first-half sessions on the post-tests (M=
238.6). It is indicated that the total number of words increased through the second-half
sessions in both the pre and the post-tests. Therefore, it is shown that the intervention of
TS had a positive affect on the improvement of total number of words as a long-term
effect.

In order to investigate whether there was the short-term effect on the total number
of words, a paired-samples t-test was conducted to compare pre-test and post-test data.
There was no significant difference in the scores for pre-test (M=24l .62, SD=76.98) and
post-test (M=280.62, SD=68.67); t (15)= -l .771, p = 0.097 (p>0.05) (See Table 7-3 & 7-
4). These results indicated that each intervention did not have a strong effect on the
difference between pre-test and post-test. Although there was no statistical difference
between pre and post-test, the post-test mean (M=280.62) was greater than the pre-test
mean (M=241.62). Therefore, the intervention TS did not have statistically significant

effect, but the total number of words increased on the post-tests as a short-term effect.

56

4. Average Length of Utterances

 

L's Spontaneous Speech Test
4. Average Length of Utterances
U3
0
U
1:
t!
In
0
1:
:9
H-
O
r:
O.)
—l
0
DD
6
5 1 2 3 4 5 6 7 8 910111213141516
Sessions
'i- 2-1-pre +2-1-post

 

 

 

 

(Figure 3-7. Average Length of Utterances in L’s Spontaneous Speech Test)

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean II Std. Deviation Mean
Pair Pre.Test_2_1 38.1375 16 20.73007 5.18252
1 96541755121 38.7750 16 11.99592 2.99898

 

 

 

(Table 8-1. L’s Paired t-test on Average Length of Utterances)

Paired Samples Test

Figure 3-5 shows there
was no increasing short-
terrn tendency in L’s
average length of
utterances as a result of
the interventions. The
long—term effect of
intervention was shown
through comparing the
means between the
first-half and the

second-half sessions.

 

Paired Differences

 

95% Confidence
Interval of the
Std. Error Difference

 

Mean Std. Deviation Mean Lower

Upper

t df Sig. (2-tailed)

 

Pair Pre_Test_2_1 -
1 PosLTesL2_1

 

 

 

 

 

-.63750 19.71662 4.92916 t11.14375 9.86875

 

 

-.129 15 .899

 

 

 

 

(Table 8-2. L’s Paired t-test on Average Length of Utterances)

The mean for the second-half sessions on the pre-test (M=42.6) was higher than

the mean for the first-half sessions (M=33.6). Also, the mean of the post-tests for second

half sessions (M=39.8) is higher than the mean of the first-half sessions (M= 37.7). It

57

indicates that the average length of utterances increased in both pre and post-tests through
the second-half sessions. Therefore, it is shown that the intervention of MIT and MIT
combined with TS gave a positive effect on the improvement of average length of
utterances as a long-term effect.

In order to investigate whether there was a short-term effect on the sample length,
a paired-samples t-test was conducted to compare the pre and the post-tests. There was no
statistically significant difference in the scores for pre—test (M=38.13, SD=20.73) and
post-test (M=38.77, SD=11.99) ; t(15)=-0.129, p = 0.899 (p>0.05)(See Table 8—1 & 8-2).
These results indicat

ed that although the pre and post means slightly increased, each intervention did
not have a statistically significant effect on the difference between the pre-tests and the
post-tests. Therefore, the intervention of MIT and MIT combined with TS did not affect
the improvement on the average length of the three longest utterances as a short-term

effect.

 

R's Spontaneous Speech Test
4. Average Lengh of Utterances

Figure 3-6 shows that
there is no increasing
tendency on the post-

test data in R’s average

 

length of utterances.

Average Length of Utterances

1 2 3 4 5 6 7 8 910111213141516

S . There was no long—term
CSSIOI'IS

"‘- 2-1-pre z'l'pOSt effect of the
(Figure 3-8. Average Length of Utterances in R’s SP Test)

 

 

 

58

intervention by comparing the means between the first-half

and the second-half sessions.

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean ll Std. Deviation Mean
Pair Pre_TesL2_1 41.6125 16 16.81459 3.95365
1 PosLTesL2_1 43.8875 16 13.02955 3.25739

 

 

 

(Table 8-3. R’s Paired t-test on Average Length of Utterances)

 

 

 

 

 

 

 

 

 

 

 

 

 

Paired Samples Test
Paired Differences
95% Confidence
Interval of the
Std. Error Difference
Mean Std. Deviation Mean Lovrer Upper t df Sig. (2-tailed)
Pair Pre_TesL2_1- b
1 PosLTesLZJ -2.27500 23.22144 5.80536 14.64883 10.09883 -.392 15 .701

 

(Table 8-4. R’s Paired t-test on Average Length of Utterances)

The mean for the second-half sessions on the pre-test (M=40.81) was lower than

the mean for the first-half sessions (M=42.41). Also, the mean for the second-half

sessions (M=43.81) was slightly lower than mean for the first-half sessions in the post-

tests (M= 43.96). There was no progression in both pre and post-tests. It is indicated that

the average length of utterances did not increased in both the pre and the post-tests

through the second-half sessions. Therefore, it is shown that the intervention of TS did

not have a positive affect on the improvement on the average length of three longest

utterances as a long-term effect.

In order to investigate whether there was the short-term effect on the sample

length, a paired-samples t-test was conducted to compare the pre and the post-tests. There

was no significant difference in the scores for pre-test (M=4l.61, SD=15.81) and post-

59

 

test (M=43.88, SD=13.02); t(15)=-0.392, p = 0.701 (p>0.05) (See Table 8-3 & 8-4).

These results indicated that each intervention did not effect the difference between the

pre-tests and post-tests. Although the mean for the post-tests (M=43.88) was slightly

greater than the pre-test mean (M=41 .61), there is no statistically significant difference

between the pre and the post-tests. Therefore, the intervention of the TS did not affect an

improvement on the average length of the three longest utterances as a short-term effect.

5. Average Number of Words per Minute

 

80
70
60
50
40
30
20
10

0

Average Number of Words
per Minute

 

L's Spontaneous Speech Test

5. Average Number of Words per Minute

 

1 2 3 4 5 6 7 8 910111213141516
Sessions
'1“ 3-1-pre +3-1-post

 

 

(Figure 3-9. Average Number of Words per Minute in L’s SP Test)

 

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean ll Std. Deviation Mean
Pair Pre_TesL3_1 44.4563 16 13.09488 3.27372
1 PosLTesL3_l 46.1562 . 16 8.23941 2.05985

 

 

(Table 9-1. L’s Paired t-test on Average Number of Words per Minute)

60

Figure 3—9 shows that
there was no significant
increasing tendency in
L’s average number of
words per minute. The
long-term effect of the
interventions was
shown through
comparing the means
between the first and

second-half session.

Paired Samples Test

 

 

 

 

 

 

 

 

 

 

 

 

Paired Differences
95% Confidence
Interval of the
Std. Error Difference
Mean Std. Deviation Mean Lower Upper t df Sig. (2-tailed)
1;“ 1138;515:5833: -1.70000 8.65602 2.16400 -6.31247 291247 -.786 15 .444

 

 

 

(Table 9-2. L’s Paired t-test on Average Number of Words per Minute)

The mean for the second-half sessions on the pre-test (M=45.2) was higher than the mean
for the first-half sessions (M=43.8). On the other hand, the mean for the second-half
sessions (M=43.7 9) was lower than that of the first-half sessions on the post-tests (M=
48.52). It is indicated that the average number of words per minute increased in the
second-half sessions on the only pre-tests. Therefore, it is shown that the interventions of
MIT and MIT combined with TS did not have a positive effect on both the pre and the
post-test on the improvement in the average number of words per minute as a long-term
effect.

In order to investigate whether there was a short-term effect on the average
number of words per minute, a paired-samples t-test was conducted to compare to pre-
tests and post-tests from the first to the sixteenth session. There was no statistically
significant difference in the scores for pre-test (M=44.45, SD=13.09) and post-test
(M=46.15, SD=8.23); t(15)= - 0.786, p = 0.444 (p>0.05) (See Table 9-1 & 9-2). Although
these results showed that the interventions do have no statistically significant effect on
the difference between pre and post-tests, there was a slight increase in means (pre=44.45,
post: 46.15). Therefore, the results indicated that MIT and MIT combined with TS did

not have a strongly positive effect on the average number of words per minute.

61

 

 

R's Spontanous Speech Test
5. Average Number of Words per Minute

60
50
40
30

20

Average Number of Words
per Minute

1 2 3 4 5 6 7 8 910111213141516

Sessions
'1‘ 3-1-pre +3—1-post

 

 

 

(Figure 3-10. Average Number of Words per Minute in R’s SP Test)

 

 

 

Paired Samples Statistics
Std. Error
Mein II Std. Deviation Mean
Pair Pre_TesL3_1 38.9375 16 7.48027 1.87007
1 PosLTesL3_l 38.0000 16 7.41791 1.85448

 

 

 

 

 

 

(Table 9—3. R’s Paired t-test on average number of words per minute)

Paired Samples Test

Figure 3—10 shows that
there was no significant
increase in R’s average
number of words per
minute. The long-term
effect of intervention
was shown through
comparing the means
between the first and

second-half sessions.

 

Paired Differences

 

Std. Error

95% Confidence
Interval of the
Diffaence

 

Mean Std, Deviation Mean Lover

Upper

t df Sig. (2-tailedl

 

 

Pair Pre_TesL3_1 -

1 Poerest.3_1 33750

7.54214 1.88553 -3.08142

 

 

 

 

 

4.95642

 

.497 15 .626

 

 

 

 

(Table 9-4. R’s Paired t-test on Average Number of Words per Minute)

The mean for the second-half sessions (M=42.01) was higher than the mean of

first-half sessions on the pre-test (M=35.86). Also, the posttest mean for the second-half

62

sessions (M=41.52) was higher than that for the first-half sessions (M= 34.47). It is
indicated that the average number of words per minute increased through the second—half
sessions in both pre and post-tests. Therefore, it is shown that the intervention of TS had
a positive effect to improving the average number of words per minute in both pre and
post-test as a long-term effect.

In order to investigate whether there was a short-term effect on the average
number of words per minute, a paired-samples t-test was conducted to compare pre—test
and post-test data from the first session to the sixteenth session. There was no statistically
significant difference in the scores for pre-test (M=38.93, SD=7.48) and post-test (M=3 8,
SD=7.41); t(15)= 0.497, p = 0.626 (p>0.05) (See Table 9-3 & 9-4). These results
indicated that the interventions generated no statistically significant difference between
pre and post-tests scores, as well as the difference of means in pre (M=38.93) and post-
test (M=3 8). Therefore, the TS did not affect the average number of words per minute as

a short-term effect.

63

Naming Measurement

6. Total Number of Items Named

 

L's Naming Test
6. Total Number of Items Named

 

14

 

12
10

 

 

 

 

 

 

 

r T T T T 7— V

Total Number of Items Named

Sessions

 

T T l' I

'15- 1)-pre +1)-post

8 A
6~Mfiy ’75} Axel—~—
4 -——;.;-7».‘—4—»m ‘:

x t l
2
0

1 2 3 4 S 6 7 8 910111213141516

 

 

(Figure 4-1. Total Number of Items Named in L’s Naming Test)

 

 

 

 

 

 

 

 

 

Figure 4-1 shows the
significant increasing
tendency in L’s total
number of items named.
The long-term effect of
the interventions was
shown through
comparing the means

between the first-half

 

 

 

 

Paired Samples Statistics and the second'half
Std. Error sessions.
Mean N Std. Deviation Mean
'—
Pair Pre_Test 4.5000 16 1.21106 .3027?
1 PosLTest 8.1875 16 204022 .51006
(Table 10—1. L’s Paired t-test on Total Number of Items named)
Paired Samples Test
Paired Differences
95%Conf1dence
Interval of the
Std. Error 01091911“
Mean Std. Deviation Mean Lower Upper t df Sig. (2-tailedl
Pail PrLTest'PosLTesl -3.68750 1.30224 .32556 -4.38142 -2.99358 -ll.327 15 .000

 

 

 

 

 

 

 

 

 

 

(Table 10-2. L’s Paired t-test on Total Number of Items named)

64

 

The mean for the second-half sessions on the pre-test (M=5.12) was higher than
the mean for the first-half sessions (M=3.88). Also, the mean for the second-half sessions
on post-tests (M=9.75) was higher than that of the first-half sessions (M= 6.62). It
indicates that the total number of items named increased through the second-half sessions
on both pre and post-tests. Therefore, it is shown that the interventions of MIT and MIT
combined with TS had a positive effect on the improvement of the total number of items
named as a long-term effect.

In order to investigate whether there was a short-term effect on the total number
of items named, a paired-samples t—test was conducted to compare pre-test and post-test
results through all sixteen sessions. There was a statistically significant difference in the
scores for pre-test (M#.50, SD=1.21) and post-test (M=8.18, SD=2.04); t (15): -11.327,
p = 0.000 (p<0.05) (See Table 10-1 & 10-2). These results suggest that each intervention
does have an effect on the difference between pre and post-test scores. Therefore, the
results suggest that when the interventions of MIT and MIT combined with TS were

conducted, the total number of Items named increased as a short-term effect.

 

Fi re 4-2 shows a sli tl
R's Naming Test gu gh y

6. Total Number of Items Named increasing tendency in R’s
total number of items named.
The long-term effect of
intervention was shown

through comparing the

 

means between the first half

1 2 3 4 5 6 7 8 910111213141516

Total Number of Items Named

Sessions and the second half sessions.
"A" 1)-pre +1J-post
(Figure 4-2. Total Number of Items Named in R’s Naming Test)

 

 

 

65

Paired Samples Statistics

 

 

 

 

 

 

 

Std. Error

Mean II Std. Deviation Mean

Pair Pre_Test.l 47.0625 16 7.46073 1.86518
1 PosLtesLl 64.0625 16 7.01872 1.75458

 

 

(Table 10-3. R’s Paired t-test on Total Number of Items named)

 

 

 

 

 

 

 

Paired Samples Test
Paired Differences
95% Confidence
Interval of the
Std. Error Difference
Mean Std. Deviation Mean Lower Upper t df sen-511a)
Pairl Pre_TesL1-Post.tesL. 4.00000 8.51665 2.12916 L11.53820 246180 -3.288 15 .005

 

 

 

 

 

 

 

 

(Table 10-4. R’s Paired t-test on Total Number of Items named)

The mean for the second-half sessions on the pre-test (M=48) was higher than the

mean for the first-half sessions on the pre-test (M=46). On the other hand, the mean of

the post-test for the second-half sessions (M=51.75) was lower than that of the first-half

sessions (M= 56.37). It is indicated that the total number of items named increased

through the second-half sessions only on the pre—tests. Therefore, it is shown that the

intervention of TS gave a positive effect on the improvement of the total number of items

named only on the pre-test as a long-term effect.

In order to investigate whether there was a short-term effect on the total number

of items named, a paired-samples t-test was conducted to compare pre-test and post-test

data for all sixteen sessions. There was a statistically significant difference in the scores

for pre-test (M=47, SD=7.46) and post-test (M=54, SD=7.01); t (15)= -3.288, p = 0.005

(p<0.05) (See Table 10-3 & 10-4). These results indicated that each intervention does

have an effect on the difference between pre and post-test scores. Therefore, the total

number of Items named increased after the intervention of TS was conducted.

66

7. Average Number of Prompts per Item Named

 

Average Number of Prompt
per Item Named

 

L's Naming Test

7. Average number of prompt per Item Named

 

 

 

2.5
A . .
2 "1 '1 I".
’4 .f, It ‘
1.5 I ‘l 7' 1‘: 5
8 \
' )1 ~ 1
' 8’ f

1

P
UI

O

 

 

 

1 2 3 4 S 6 7 8 910111213141516

Sessions
"i' 2)-pre +2)-post

Figure 4-3 shows the
reducing tendency in post-
test data for L’s average
number of prompts per item
named. The long-term effect
of the interventions was
shown through comparing

the means between the first-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(Figure 4-3. Average Number of Prompts per Item Named half and the second-half
in L’s Naming Test)
Paired Samples Statistics “35'0“"
Std. Error
Mean It Std. Deviation Mean
Pair Pre_TesL2 1.2906 16 .60006 .16002
1 PosLTesLZ .5375 15 .26552 .06638
(Table 1 1-1. L’s Paired t-test on Average Number of Prompts
per Item Named)
Paired Samples Test
Paired Differences
95% Confidence
Interval of the
Std. Error Difference
‘ Mean Std. Deviation Mean Lower Upper t df Sig. (2-tailed)
Ii“ 52:55:32. .65312 70225 .17556 .27892 1.02733 3720 15 .002

 

 

 

 

 

 

 

 

 

 

(Tablel l-2. L’s Paired t-test on Average Number of Prompts per Item)

The mean for the second-half sessions on the pre-test (M=1.03) was lower than

the mean of first—half sessions on the pre-test (M=1.55) The goal in this variable is to

decrease the number of prompts, so there was a slight improvement by reducing the mean

67

 

in the post-tests. On the other hand, the post-test mean for the second-half sessions
(M=0.68) is slightly higher than that of first-half sessions (M=0.6). It indicates that the
average number of prompts per item increased through the second-half sessions on the
post-test. Namely, the pre-test results were an improvement, while the post—test increase
was a slight regression. Therefore, it is shown that the intervention of MIT and MIT
combined with TS did not have a significantly positive effect in decreasing the average
number of prompts both on the pre and post-tests as a long-term effect.

In order to investigate whether there was the short—term effect on the average
number of prompts, a paired-samples t-test was conducted to compare pre-test and post-
test results for all sixteen sessions. There was a statistically significant difference in the
scores for pre-test (M=1.29, SD=0.60) and post—test (M=0.63, SD=0.26); t (15)= 3.72, p
= 0.002 (p<0.05) (See Table 11-1 & 11-2). These results suggest that each intervention
does have an effect on the difference between pre and post-test performance. Therefore,
the results indicate that the prompts by the researcher were reduced in the post-test as a

short-term effect.

 

. F‘ 4-4 h th t
R's Naming Test lgure S ows a

7. Average Number of Prompt per Item Named there was difference

 

,_, 0.16

2' 0.14 between pre and post-

E E 0.12

‘5 a 0.1 test data in R’s average

5 ‘z‘ 0.08

E E 0.06 number of prompts per

5 E 0.04 .

E” =- 0.02 I items named. The long-

: 0

< 1 2 3 4 s 6 7 8 910111213141516 tefineffectofthe
Sessions

interventions was

 

 

"A" 2)-pre +2)-post

(Figure 4-4. Average Number of Prompts per Item Named in R‘s Naming)

 

68

shown through comparing the means between the first half

and the second-half sessions.

 

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean ll Std. Deviation Mean
Pair Pre_TesL .0394 16 .04404 .01101
1 PosLTesLZ .0400 16 .02608 .00652

 

 

(Tablel 1-3. R‘s Paired t-tcst on Average Number of Prompts per Item Named)

 

 

 

 

 

 

 

 

 

 

 

 

Paired Samples Test
Paired Differences
95%Conf1dence
Interval of the
Std. Error Difference
Mean Std. Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair Pre_Test_2 -
1 PosLTesLZ '.00062 .03924 .00981 -.02153 .02028 -.064 15 .950

 

 

 

(Tablel 1-4. R’s Paired t-test on Average Number of Prompts per Item)

The mean for the second-half sessions on the pre-test (M=0.03 l) was lower than

the mean of first half sessions on the pre-test (M=0.047). Also, the mean for the second-

half sessions (M=0.037) was lower than that of the first-half sessions on the post-test

(M=0.042). There was improvement by reducing the number of the researcher’s prompts

on both the pre and post-test. It indicates that the average number of prompts per item

was reduced through the second-half sessions. Therefore, it is shown that the intervention

of the TS gave a positive effect on reducing the average number of prompts per item as a

long-term effect.

In order to investigate whether there was a short-term effect on the average

number of prompts per item, a paired-samples t-test was conducted to compare pre-test

69

and post-test data for all sixteen sessions. There was no statistically significant difference
in the scores for pre-test (M=0.039, SD=0.044) and post-test (M=0.04, SD=0.026) ; t
(15)= -0.064, p = 0.950 (p>0.05) (See Table 11-3 & 11-4) These results suggest that each
intervention has no statistically significant effect on the difference between pre and post-
test. The post-test mean (M=0.04) was slightly higher than the pre-test mean (M=0.039).
Therefore, the results indicate that the prompts by the researcher were not reduced by the
intervention, so there was no improvement on the average number of prompts as a short-

term effect.

Melodic Intonation T homer Measurement

 

 

8. MIT Scores of Percent
I sessions more than in the
L 5 Test
first-half sessions. The mean
100
*5 90 for the second-half sessions
E 80
a: 70 (M=75.1) was higher than
'45 60
'3 50 the mean for the first-half
g 40
U
W 30 sessions =58.5).
I: 20 (M
2 10 . . .
0 Therefore, it mdlcates that
1 2 3 4 5 6 7 9 1011 12 13 1415 15 there was an improvement in
Sessions

 

 

 

(Figure S-l. MIT Scores of Percent in L’s MIT Test) L S WT scores contmumg

Figure 5-1 shows that there was the increasing tendency through the second-half

in L’s MTT scores. To be exact, there was the tendency 53351095-

toward an average rate of increase in the second-half

70

CHAPTER V

DISCUSSION

Summon and Conclusion

The summary of results is shown in Table 12. There are two ways in which the
data for the eight variables are interpreted: 1) Long-term effect; 2) Short-term effect.
First, a long-term effect was measured by comparing the means of the first-half sessions
with the means of the second-half sessions on both pre and post-tests (Independent t-test).
If the second-half mean is higher than the mean of the first-half sessions, this result may
indicate that the interventions in the second-half period of sessions have been effective.
Table 12 shows a numerical increase, as ‘Yes’, and no increase, as ‘No’.

Secondly, the short-term effect was measured by comparing the scores of pre-tests
with those of post-tests (Paired t-test). Table 12 shows only whether or not there was a
statistically significant difference between the pre-test means and the post-test means. If
there was a significant difference (p values < 0.05), it is labeled as ‘Yes’, and if none, as

‘No’.

71

Table 12. Summagg at the data result in t-test

 

 

 

 

Variables L R
1. Long-term: No test of Significance
" Yes " indicates that Second-half means > 2 2
I I
F i’ st-half means Long- Term Short Long- Term Short
Term Term
2. Short-term: Test of Significance
Pre Post Pre Post
1. Total Length of Samples Yes Yes Yes No Yes Yes
2. Total Number of Utterances No Yes Yes Yes Yes No
3. Total Number of Words Yes Yes Yes Yes Yes No
4. Average Length of Utterances Yes Yes No No No No
5. Average Number of Words per Minute Yes No No Yes Yes No
6. Total Number of Items Named Yes Yes Yes Yes No Yes
7. Average Number of Prompts per Item
Yes No Yes Yes Yes No
Named
8. MIT Scores of Percent Yes - - -

 

 

(1. Long-term effect— Independent t-test, 2. Short-term effect — Paired samples t-test)

L ’s Results

The summary of results indicated that the music therapy interventions of Melodic

Intonation Therapy (MIT) and Therapeutic Singing (TS) had a positive effect on L’s

functional communication skill. In L’s long-term effects on pre-test measures, there was

improvement in six variables: l.the total length of samples; 3.the total number of words;

4.the average length of utterances; 5.the average number of words per minute; 6.the total

number of items named, and 7.the average number of prompts per item named, and no

improvement on only one variable, 2.the total number of utterances. On the other hand, in

the post-tests, there was improvement in five variables: l.the total length of samples;

72

2.the total number of utterances; 3.the total number of words; 4.the average length of
utterances; and 6.the total number of items named, and no improvement on the two
variables: 5.average number of words per minute, and 8.the average number of prompts
per item named. It was shown that the means of eleven among the fourteen variables
showed increases in both the pre-test and the post-test. Therefore, it might be indicated
that there was a long-term effect on L’s functional communication skills through the last
period of sessions. However, statistical significance was not achieved.

Also, in L’s short-term effect, there were statistically significant differences in
five variables: l.the total length of samples; 2.the total number of utterances; 3.the total
number of words; 4.the total number of items named; and 5.the average number of
prompts per item named, and no differences in two variables: l.the average length of
utterance; and 2.the average number of words per minute. It was shown that there were
statistically significant differences in most variables between pre and post-tests.
Therefore, it is indicated that MIT and TS interventions helped L improve his fianctional

communication skills in several areas as short-term effects.

Discussion of L ’s Four Goal ’s Areas

There were four goals targeted in L’s study. The first goal was to investigate
whether or not Therapeutic Singing (TS) combined with Melodic Intonation Therapy
(MIT) helps adult clients who are diagnosed with expressive aphasia as measured by the
quality of their functional communication skills. For investigating the first goal, the
paired t-test was conducted in the second-half sessions (9th - 16th sessions), which is the
second period of sessions conducted with MIT combined with TS, by comparing the

means of pre and post-tests.

73

Table 13-1 shows that the means for post-tests are distinctly higher than those of
pre-tests in all seven variables (see Appendix 15-1). It supports that there was
improvement on the post-tests in all seven variables after conducting MIT combined with
TS through the second-half sessions. On the other hand, table 13-2 shows that there were
statistically significant differences (p<0.05) in three variables: l.the total length of
sample, 2.the total number of utterances, and 6.the total number of items named, through
the second-half sessions, while there were no statistically significant differences (p>0.05)
in four variables: 3.the total number of words, 4.the average length of utterances, 5.the
average number of words per minute, and 7.the average number of prompts per item
named.

As a result, the increases in measured numerical quantities indicated that the
interventions of MIT combined with TS had a slightly supportive effect in the Functional
Communication Skills by increasing the means. The means for post-tests in all seven
variables were higher than the means for pre-tests, although the results did not show
statistically supported differences between the means of pre and post-tests. It is believed
that the limited number of sessions (N =8) may be the reason for a lack of statistically

supportive differences (See Table 13-1 in Appendix 15—1).

The second goal was to explore whether MIT treatment itself helps clients
improve the quality of their functional communication skills. For investigating the second
goal, the paired t-test was conducted in only the first-half sessions (13’ — 8th sessions), in
which only MIT intervention was conducted, by comparing the means for pre and post-

tests.

74

Table 14-1 shows that the means for post-tests are also distinctly higher than the
means of pre-tests in all seven variables. It supports that there was improvement on the
post-tests after conducting the MIT intervention during the first-half sessions (See table
14-1 in Appendix 15-2). However, the results in table 14-2 (see Appendix 15-2) shows
that there were statistically significant differences (p<0.05) in three variables: 3.the total
number of words, 6.the total number of items named, and 7.the average number of
prompts per item named during the first-half sessions, while, there were no statistically
significant differences (p>0.05) in four variables: l.the total length of samples, 2.the total
number of utterances, 4.the average length of utterances, and 6. Average number of
words per minute (See Table 14-2 in Appendix 15-2).

As a result, the means for post-tests improved more than those for pre-tests in all
seven variables although the results did not show statistically significant differences on
all seven variables (see Table 13-1). Therefore, MIT also had a positive effect on the
post-test results, but did not have a statistically supported difference in all seven

variables.

The third goal was to compare the effectiveness of TS combined with MIT and
MIT alone in increasing fiinctional communication skills. For investigating goal III, the
independent sample t-test was conducted on sessions by comparing the two means of a
variable: the numerical differences between the post-tests and the pre-tests in the first-
half and the second-half sessions. (Post-test Means Minus Pre-tests Means (lSt — 8'“) vs.
Post-test Means Minus Pre-test Means (9th —— 16th) (See table 15-1 in Appendix 15-3)

Table 15-1 shows that the means in groups 2 (Post-test means minus Pre-test

means in the second-half sessions) had more improvement than the means in groups 1

75

(Post-test means minus Pre-test means in the first-half sessions) on three variables: l.total
length of samples, 2.total number of utterances, and 6.total number of items named, on
the other hand, four variables showed no improvement: 3.total number of words,
4.average length of utterances, 5.average number of words per minute, and 7.average
number of prompts per item named.

Table 15-2, however, shows that there were no statistically significant differences
between the first-half (MIT) and the second-half (MIT+TS) sessions (p>0.05) except the
one variable: 6.the total number of items named. However, it is noted that the second-half
(MIT+TS) sessions had higher means than those of the first-half sessions on three
variables: l.the total length of samples, 2.the total number of utterances, and 6.the total
number of items names. To be exact, the negative scores of Mean Difference in Table 15-
2 implied that the means of the second-half (MIT+TS) sessions were higher than the
means of the first-half (MIT) sessions. On the other hand, the positive scores of Mean
Difference showed that the means of the first-half (MIT) sessions were higher than the
means of the second-half (MIT+TS) sessions on four variables: 3.total number of words,
4.average length of utterances, 5.average number of words per minute, and 7.the average
number of prompts per item named. As a result, it can be shown that MIT combined with
TS had a slightly positive effect on the improvement of functional communication skills

on those four variables. Therefore, no clear trend can be attributed to the interventions.

76

The fourth goal was to determine whether TS enhances MIT treatment (See
Figure 5-1 in Ch. 4). The results shows that MIT combined with TS interventions had a
positive effect on MIT scores, more so than MIT intervention alone. As the result, it is
considered that the TS intervention is beneficial in enhancing MIT treatment.

As the evidence of its effectiveness, TS can support the goals of other techniques,
not by taking their place as primary interventions, but as a separate technique in support
of the other. For example, the primary goal of this study is improving the fimctional
communication skills of the aphasic subjects, and MIT is one of the significant
techniques for this goal in enhancing speech function. Significantly, the supportive role
of TS could serve as an important reinforcement for improving the goals of MIT and this
study. TS is the practice of combining elements of speech production and also
incorporates emotional elements of speech and language. People who suffer from the
symptoms of a stroke also suffer from the emotional wounds such as depression, and this
emotional dysfunction may affect language and speech function negatively. A change in
therapeutic focus by combining TS with MIT could provide a more beneficial

intervention in the functional goals of this study, as well as the goals of MIT.

R ’s Results

R, who had high functional speech, received only the TS intervention, as has been
mentioned in the Methods chapter, because he already had the ability to repeat full
sentences, which represents the final stage of MIT. In R’s long-term effect, reflected in
. the pre-tests, there was improvement in five variables: 2.the total number of utterances;

3.the total number of words; 5.the average number of words per minute; 6.the total

77

number of items named; and 7.the average number of prompts per item named, and no
improvement in two variables: l.the total length of samples; and 4.the average length of
utterances. On the other hand, in the post-tests, there was improvement in five variables:
l.the total length of samples; 2.the total number of utterances; 3.the total number of
words; 5.the average number of words per minute; and 7.the average number of prompts
per item named, and no improvement in two variables: 4.the average length of utterances;
and 6.the total number of items named. It was shown that the means of most variables
were increased in both the pre-test and the post-test as a Iong-terrn effect, except in the
average length of utterances. Therefore, it could be considered that there were possible
long-term effects on R’s functional communication skills through the sixteen sessions
although R did not show the improvement in the length of longest utterances. These
results represent only numerical directions in improvement, as statistical significance was
not established.

In contrast, there was no statistically significant difference in the short-term
effect. Only the mean of two variables (1. the total length of samples and 6. total number
of items named) improved in the post-test. There was no short-term effect in five
variables (2.the total number of utterances, 3. the total number of words, 4. the average
length of utterances, 5. the average number of words per minute, and 7. the average
number of prompts per item named). The result shows that even though the mean of post-
tests is slightly higher than the mean of pre-tests, there was no statistically significant
difference between pre and post-tests. Therefore, it is considered that the TS interventions
may have helped R to improve functional communication skill as a long-term effect, but

there seems to be no statistically significant short-term effect.

78

Specifically, R’s result showed there was no improvement at all in the average
length of utterances both long-tenn and short-term. As discussed in chapter III, R had a
high functional communication level, and most functional ability may have been
recovered previously because the onset of his stroke was seven years ago. This fact may
indicate that the TS intervention may not have a significantly positive effect on the short-
term gains for a person who had been diagnosed with expressive aphasia several years
ago and had already recovered speaking functions. The goal for R in this study was to
gain information as to whether TS treatment itself helps clients improve the quality of
their functional communication skills. Although the TS gave positive results on some
variables in the functional communication measurement, R’s results did not show a
stronger effect than L’s results. This may be explained by the fact that R already had

recovered many fianctions during the seven years after onset of his stroke.

Anecdotal Improvement

L’s spouse stated at the middle of the research, “L is feeling like this is helping
him!”, and he had started to ride a bike for 4 miles away from his home and
communicated with the cashier of the gas station during the last period of sessions. This
was his positive feedback regarding gains assessed through the Spontaneous Speech
Measurement.

One day during the last period of sessions, R mentioned that his brother said that
R was talking more clearly than before by phone, and both his close friend and his wife

stated that his speech function was more stabilized than in the past.

79

Im lications or Intervention

Music is not a simple stimulus for activating brain function, but music is a
complex combination of many supportive ways to trigger response to improve restricted
functions in a human being. We cannot say that taking a dose of singing activities three
times a day will help clients to return to a normal state of speech function. However, we
can say that singing can trigger improvement in speech function by activating long-term
memory, supporting states of emotion, and triggering unconscious automatic words.
Jackson (1871) differentiated propositional speech from so-called automatic speech. He
stated that even when propositional speech became seriously damaged, automatic speech
could be preserved in aphasia. TS may be able to trigger unconscious memory for
producing automatic speech.

The findings of this study lent support to the idea that MIT and TS have positive
effects on improving functional communication skills. Therapeutic Singing (TS) was
added to serve a supportive role during the sessions in order to investigate whether TS
would help to improve fimctional communication skills in L’s MIT interventions, as well
as a main intervention in R’s sessions. The reason for investigating the combination of
both interventions was that TS and MIT have reciprocally important roles. MIT has the
benefit of focusing on the target phrases more than TS, and TS may be able to activate
long-term memory functions through the usage of familiar songs, as well as support a
state of pleasurable emotion to counteract depression. Singing also can help a person to
engage actively in pleasure and emotional satisfaction. Investigating whether there is

effectiveness in both MIT and TS can uniquely add to the body of knowledge in this area.

80

Since the 1970’s many researchers have been investigating whether or not singing
can help enhance speech intelligibility, as well as the effectiveness of MIT in speech
therapy. Studies until the 2000’s supported the effectiveness of MIT in speech production
(Baker, 2000; Belin, et al., 1996; Bonakdarpour, Eftekharzadeh, & Ashayeri, 2003;
Carroll, 1996; Galloway & Kraus, 1982; Laughlin, Naeser, & Gordon, 1979; Sparks &
Deck, 1986; Sparks, et al., 1974), as well as the effectiveness of Therapeutic Singing in
speech function (TS) (Cohen, 1992; Cohen & Ford, 1995; Cohen & Masse, 1993; Glover,
Kalinowski, Rastatter, & Stuart, 1996; Keith & Aronson, 1975).

Some recent, however, research studies showed that there was dissociation
between singing and speaking (Hebert, et al., 2003; Peretz & Hébert, 1995; Racette,
Bard, & Peretz, 2006). They stated that singing did not affect word production by
producing normal speech. Even though the subjects produced more word utterances while
singing, after repetition of the singing activity, speaking production with the same lyrics
did not improve. Similar to these studies, when the researcher asked subject L to speak
target phrases after MIT treatment with same materials, subject L could not speak most of
the target phrases even though he had produced the target contents within the musical
condition of MIT. As explanation, if a two-year-old toddler, for example, who has not
started speaking yet, sings Twinkle Twinkle Little Star along with his mom, the toddler
can then sing producing 60% of the lyrics. No matter how many words the toddler can
produce while singing, can the toddler speak the lyrics without music? Of course, the
answer is ‘No,’ because the toddler’s brain is not fully activated for speaking, even
though spoken language comprehension may have started already. Even so, can we

conclude there is dissociation between singing and speaking?

81

In contrast to Hebert’s study (2003), this study used, as pre and post-tests,
different measurement tools within a non-musical condition. In the most of NMT
intervention in Speech and Language Training, as well as Hebert’s research, word
production has been measured within a musical condition. This study, however, has used
measurement techniques designed after speech pathology tests of Functional
Communication Measurement within non-musical conditions. This differs from other
neurologic music therapy techniques. For a concrete example, MUSTIM is one of the
effective NMT techniques, and the variables are measured by how many words the client
produces during the MUSTIM intervention under musical conditions. For this reason, the
measurement tools chosen and used in this study can be valuable methods for future
researchers.

Also, the number of numerical variables gave this study integrity, even though it
developed into two case studies. The Paired t-test and Independent sample t-test was
conducted on fourteen to fifteen variables.

The rhythmic factor in speaking, emphasized by using castanets, also served as a
memory reinforcement in this study. Resent research showed that the intoned melody was
not the important factor in improving word production (Hebert, et al., 2003). The drum,
however, played an important role for all children in increasing the length and clarity of

response in one study (1996, Carroll).

The amount of time consumed was one of the limitations to having a large subject
pool in this study. In order to obtain the Spontaneous Speech Measurement data, it

required two and one-half hours for each session, including driving to the subject's

82

location, as well as three hours for coding data. Using five and one-half hours for each
session to obtain data on the fourteen to eighteen variables was prohibitive to having
more subjects. However, this study gains credibility from the amount of date collected
from the two subjects.

There was also a limitation in the recruitment of aphasic participants who are at
the same in level of speaking ability, because most of the clients who are diagnosed with
cerebrovascular accidents (CVA) have combined disorders with other diagnoses. For
example, L and R were diagnosed the same with expressive aphasia, but the researcher
recognized that L also had motor functioning problems in using his oral muscles. It was
discovered that L was also diagnosed with mild apraxia after engaging in several
sessions. For this reason, recruiting a homogeneous pool of subjects is a most difficult
step at the outset of the research. Although five subjects were recruited who were
diagnosed with adult expressive aphasia, the study was limited to just two, due to the

wide variation in speech abilities and the time consuming nature of the intensive sessions.

Im lications or Future Research

There are several suggestions for future research. First, the more sessions and
longer period, such as a year, the better will be the results in data analysis. This study was
conducted with restricted time line as a masters thesis, so subjects participated in a two-
month time line. The speech and language pathologists recommended that they have
sessions with persons with expressive aphasia for over a year in order to evaluate

improvement.

83

A second recommendation is in the intensity of treatment. According to the
original MIT protocols, it suggested a six-week intensive plan. At a rehabilitation center
in Michigan, speech pathologists also meet stroke patients five times a week in an
intensive plan. Subjects L and R had interventions around three to four times a week
during the second period of sessions.

For the future researcher, it is also recommended that a better way to measure
would be to reduce the time consumed by examining only a pre and post-test at the first
and last sessions. This would generate valuable data if gathered in long-term treatment
over a year with an intensive plan of intervention. Pre and posttests might also be given
at the beginning and end of each week.

Also, length of time post stroke had an effect on results. R, whose stroke occurred
seven years ago, had slightly more improvement than L, whose onset of stroke was a year
ago. This can be a reference for future researchers in recruiting subjects.

This research design may be a good model for a treatment plan, which was
conducted in the order of pre-test, treatment, and posttest. Namely, each session was
conducted in the order of conversation about favorite themes, musical intervention, and
conversation on the same topic. This design may be a stable model for music therapy
intervention.

Lastly, even though the testing is time consuming, the seven different variables in
functional communication skills can be good evaluation materials: 1.Total length of
samples, 2.Total number of utterance, 3.Total number of words, 4.Average length of
utterance, 5.Average number of words per minute, 6.Total number of items named,

7.Average number of prompts per item named. Those variables were selected with the

84

 

help of an expert in speech therapy, so they can be credibly used in the music therapy

field.

It would be encouraging news if persons with severe expressive aphasia could find the
means to speak again. As to the question of whether singing has any use in the recovery
of speech, the researcher would answer that language can be embedded with unconscious
automatic words ready to be “released” for conscious uses, and music can trigger this
unconscious automatism by stimulating the production of verbal utterances. This study is
this researcher’s first step in learning the power of music in rehabilitation for persons
with neurological conditions such as expressive aphasia. Through this first step of these
two case studies, the researcher is convinced of the effectiveness of MIT and TS
techniques to improve communication through speech for persons with expressive

aphasia.

85

APPENDICES

86

 

APPENDIX 1

Table 1. Overview of Melodic Intonation Therapy and Scoring Criterion

    

in.

   

I. Hummmglmelody 2X (HT) ‘
2. Humming in unison (HT)

3. Fade (HT) '

l. Humming (HT)

. ‘ 2. Unison smgrngGJT) 0"}: .
' 3. Unison singing with fading (HT) 1.,

4. Immediate repetition .

5. Response to a probe question

1. Introduction of item (HT)

2 Umsonwrthfad’mgy . .

3. Delayed repetition (HT)
(Back up: unison with fitting)

(Back up: Delayed repetition)

 

4

 

1. DelayedrepetmoMI-IT) . , :1
' (Backup. Umsonwrthf' " ) ' ' " g
*J-Noscore'.‘ '

   

2. Introducing sprochgesang

3. Sprechgegang With mg”). . . :1

(Back up: Unison spreehg'esang)
4. Spoken repetition with delay .
(Back up: Sprechgesang with fading)
5. Response to probe question

(Back up: Delayed repetition)

(HT: Hand Tapping)

87

 

APPENDIX 2

Melodic Intonation Therapy Measurement Form5

 

Name of Client:

Session #.

 

 

Date

 

Pre—test /

Post-test

 

 
 

. 3. Unison singin y .
* 4. Immediate repetition

 

 

Q?
1. Hummmg melody 2X (HT)
2. Humming 1n unison (HT)
3 Fade (HT) 7 . .
1.Hummmg (HT) V
2. Unison Singing

5 Response to a probe question 1
1. Introduction of 1;th (HT)

2. Unison with fading (HT)
3. Delayed repetititm (HT)

(330k up: unison With fading) - .

No score

 
 
 

No score.

(Back up: Delayed repetition). , .

1. Delayed repetition (HT)
(Back up: Unison With flitting)
2. introducing .sprechgesang (HT)

3. Sprechgesang with fading (HT) » V
4 (Back up: Unison Sprechgesang)
4. Spoken repetition with delay '

No score

(Back up Sprechgesang with fading) . H

5. Response to probe question
(Back up: Delayed repetition)

Total scores :

 

 

APPENDIX 3

Melodic Intonation Therapy (M11) Hierarchy

 

 

LEVEL I — learn intoning and hand tapping, response to hand-signals

 

Stimulus (Therapist) (T)

Hums melody twice with (HT) (HT) Hand Tapping
(T) fades participation but continues (HT)

 

Response (Client) (C)

Unison humming with (HT).

 

Score and Progression

 

No score. Proceed to next step 11.

 

 

 

 

LEVEL 2 — Ability repeating intoned sentences immediately,

Response to a question

 

 

Stimulus (Therapist) (T)

1.Hums melody with (HT),
2.1ntones sentence with (HT),

 

 

 

 

 

 

 

 

 

 

 

Step lgnals (C) to join in unison intoninj of sentence
1 Response (Client) (C) Intone sentence together with (HT) - Unison
Score and Progression Acceptable- 1 point. Proceed to Step 2, same sentence
Unacceptable - Discontinue progress for sentence.
Stimulus (Therapist) (T) Same as Stet) ! . .
Step But fades part1c1pation except (HT)
2 Response (Client) (C) Intone sentence with (HT) gradually Solo
Score and Pro ression Acceptable- 1 point. Proceed to Step 3, same sentence
g Unacceptable - Discontinue progress for sentence.
1.Signals (C) to listen
Stimulus (Therapist) (T) 2.1ntones sentence with (HT)
3.Signals (C) to repeat,
Step Cueing for initiation of utterance if necessary
3 Response (Client) (C) Repeat intoned sentence with (HT) immediately
. Acceptable— 2 point without cue, 1 point with cue
Score and Progressmn Proceed to step 4, same sentence
Unacceptable - Discontinue progress for sentence.
. . 1.1ntones question, “What did you say?”
Stimulus (Therapist) (T) 2.Signal (C) to repeat
Step Cueing for initiation of response if necessary
4 Response (Client) (C) Repeat intoned sentence with (HT)

 

 

Score and Progression

 

Acceptable— 2 point without cue, 1 point with cue
Proceed to step 1, for next sentence
Unaccegable — Discontinue pflress for sentence.

 

 

89

APPENDIX 3

(continued)
Melodic Intonation Therapy (M11) Hierarchy

 

 

i LEVEL 3 - Ability to the delay of responses by fading Therapist’s participation

Response to specific questions

 

 

Step

1.1ntones sentence with (HT),

 

Stimulus (Therapist) (T) 2.Signals (C) to join in unison sentence
3. Fade participation except (HT)
Response (Client) (C) Intone sentence with (HT) gradually Solo

 

Score and Progression

Acceptable- 1 point. Proceed to Step 2, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

1.Signals (C) to listen

 

Stimulus (Therapist) (T) 2.1ntones sentence with (HT)

3.Signals (C) to repeat afier l or 2 second Delay

4. If (C) fails, Backup step 1, same sentence (B) Back up
Response (Client) (C) Repeat Intoned sentence with (HT) after Delay

 

Score and Progression

Acceptable- 2 point without (B), 1 Point with (B)
Process to Step 3, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

 

1.1ntone a related question (e. g. what kind of, How

 

Stimulus Thera ist many)
( p ) (T) 2.Signals (C) to answer
3. If (C) fails, Backup step 2, same sentence
Response (Client) (C) Appropriate Answer (Intoned or Spoken)

 

 

Score and Progression

 

Acceptable- 2 point without (B), 1 Point with (B)
Process to Step 1 for next sentence
Unacceptable - Discontinue progress for sentence.

 

90

 

APPENDIX 3

(continued)

 

 

LEVEL 4 — Ability of Normal Speech
-

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

1.Signal (C) to listen

2.1ntone sentence with (HT)

3.Present it twice in Sprechgesang

4.Signal (C) unison Sprechgesang of sentence (HT)
5. If (C) fails, Backup to present in Sprechgesang

6. Retrial No. 4

Sprechgesang of sentence with (HT)

Acceptable- 2 point with Sprechgesang.

- 1 point with Backup

- Proceed to Step 2, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

1.Signal (C) to listen
2.Present sentence in Sprechgesang with (HT)
3.Signals (C) to repeat after 2 or 3 second Delay
4. If (C) fails, Backup to Step 1 same sentence
Sprechgesang of sentence with (HT) after 2 or 3 second
Delay.
Acceptable- 2 point without Backup

- 1 point with Backup

- Proceed to Step 3, same sentence
Unacceptable - Discontinue progress for sentence.

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

1.Signal (C) to listen

2.Present sentence once in Sprechgesang

3.Present sentence twice in Normal Speech Prosody
4.Signals (C) to repeat after 2 or 3 second Delay

5.If (C) fails, Backup to Step 2 same sentence

Repeat sentence in Normal Speech Prosody after 2 or 3
second Delay, no (HT)
Acceptable- 2 point without Backup

- 1 point with Backup

- Proceed to Step 4, same sentence
Unacceptable - Discontinue progress for sentence.

 

 

Step

Stimulus (Therapist) (T)

Response (Client) (C)

Score and Progression

l.Ask question about substantive information on the same
sentence.

2. If fail, Backup step 3 same sentence and retrial

3. Ask question about associative information

4. If fail, No backup and retrial if fail one.

Answer appropriate responses

3 point — bonus point, one or more associative responses
2 point — without Backup, substantive content

1 point - with Backup.

Proceed to next sentence.

 

91

 

APPENDIX 4

Spontaneous Speech Measurement Form
On Functional Communication Measure (FCM)

 

Name of Client:

Sexism—nit.

 

 

Date

Pre-test / Post-test

 

 

Subject is asked to choose the question from the questionnaire sheets

Response limited within 5 min.

Score Summary

 

1. Total Length of Samples

 

2. Total Number of Utterances

 

3. Average Length of Utterances

 

 

4. Total Number of Words

 

 

 

Scaring Hierarchy

 

1. Total Length of Samples

Sum of Seconds, including both intelligible
and unintelligible words

 

2. Total Number of Utterances

Including paraphasic errors

 

3. Average Length of Utterances

Average number of words for three longest
intelligible phrases
e.g.) 1. I like a ball (4 words)

* .me too (2 words)

2. I want to go home (5 words)

3. I love peanut (3 words)

*. Lovely (1 word)
= 4+5+3l3 = 4 (average length of
utterances)

 

4. Total Number of Words

 

 

Counting intelligible & correct words
(Included Nouns, Verbs, Articles,

Adjectives, Pronouns, and Preposition)

 

92

 

 

 

APPENDIX 4

(continued)

Spontaneous Speech Measurement Form
On Functional Communication Measure (FCM)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Questionnaire
Question 1:
Number Of Sub'ect’s res onses Length Of
Utterance J p Utterances
Total: Total:

 

 

 

93

 

 

APPENDIX 5

Questionnaire sheet

Spontaneous Speech On Functional Communication Measure (FCM)

What would you like to talk about during conversations?

_
a

Funny stories about your children

Your adventures as a young child/growing up
Dating and getting married

Being in the military

Your worst jobs

Your most important job/ career

Moving or traveling

Hobbies or unique interests

P?°>‘.°‘S":“E“!°

Family history/ ancestry/ genealogy

u—d
O

. Local events

p—s
I—I

. Current events

. Sports

I-I|—I|
{AN

. Favorite TV program

i—s
A

. Politics/ the economy/ the government
. Weather

. Favorite meals/ restaurants

u—II—It—I
\IQLII

. My house/ home town/ things to fix
18. My stroke and/ or other medical issues
19. Religion

20. Favorite movies

(Revised from Aphasia Need Assessment, written by Kathryn L. Garrett & David R. Beukelman, (2005)

94

 

 

APPENDIX 6

Naming Measurement Form
On Functional Communication Measure (FCM)

 

flame of Client:

Session #.

 

Date

 

Pre-test / Post-test

 

 

Subject is asked to see the Picture Stimuli Sheets and to answer the names of pictures.
Response limited within 5 min.

 

The list of Name (Client’s responses)

Number of Prompts (by Therapist)

 

 

 

Total Number of Items Named:

Average Number of Prompts pet Item Named:

95

 

 

 

APPENDIX 7

Hierarchy at Word Retrieval Prompts the Client with Aphasia

Naming Measurement Form
On Functional Communication Measure (F CM)

,9

Ask person to “say

-Ask person to complete a sentence with phonemic cues “You sleep in a be

-Less phonemic cues “You sleep in a b__”

~Verbal lead in without phonemic cues “You sleep in a _”

State or demonstrate function “You sleep on it..motion sleep, it’s a _”
-State function and supply a carrier phrase

-Direct person to demonstrate function “Show me what you do with it.”
oDirect person to state the fimction “What do you do with it?

Request the name “What’s this?”

96

9,

 

APPENDIXS 8

Picture Stimuli Sheets

Naming Test On Functional Communication Measure (F C M

97

APPENDIX 8

 

 

 

 

 

 

 

 

 

 

 

Toothbrush

 

 

 

 

 

 

 

98

APPENDIX 8

(continued)

 

 

_/

Canoe

 

 

 

 

 

 

 

 

Stethoscope

 

 

 

 

Unicorn

 

Tripod

 

99

 

 

 

 

APPENDIX 8
(continued)

 

 

 

Paint Bed

 

 

 

 

   

Tree Pencil

 

 

 

 

   

Whistle Scissors

 

 

 

 

 

APPENDIX
(continued)

 

 

 

 

 

 

 

r/”i>>
/ ',,‘//
(7.75:. ‘ x/
. \Kfitx/
\ J
Saw Helicopter
Broom

 

 

 

 

 

Hanger

 

 

Wheelchair

 

101

 

 

 

 

APPENDIX 8
(continued)

 

 

 

Camel Mask

 

 

 

 

   

Pretzel Racquet, Badminton

 

 

 

 

   

 

 

102

 

 

 

 

APPENDIX 8
(contmued)

 

 

Globe

 

 

 

 

Dart

Harmonica

 

 

Acorn

 

 

 

 

 

 

Wreath

 

 

 

 

Rhinoceros

 

 

103

APPENDIX 8

 

 

 

 

 

 

 

   

 

 

 

 

   

(continued)
Igloo ' Dominoes
Escalator Harp ' “
Knocker Bird, Pelican

 

 

 

104

 

 

 

 

(continued)

" a...

 

 

 

 

 

 

 

 

 

Accordion

 

 

 

Asparagus

 

 

 

 

  

 

 

T°ngs

 

 

105

APPENDIX 8
(continued)

 

 

 

   

 

Trellis Ruler

 

 

 

 

 

(( ))

Antenna Statue

 

 

 

 

 

 

in.

Crutch Suspenders

 

 

106

 

 

 

APPENDIX 8

(continued)

 

 

 

Calculator

 

Palm, hand

 

 

 

 

Microphone

 

 

Dice

 

 

 

 

Chopsticks

 

 

 

 

107

 

 

 

 

APPENDIX 8
(continued)

 

 

3»

Eyebrow, eye Binoculars

 

 

 

 

 

   

 

 

 

 

   

Flim Backpac

 

 

108

 

 

 

 

APPENDIX 8

 

 

 

 

 

 

 

 

 

 

 

 

   

(continued)
Wishbone Propeller
\e
Pliers Dustpan
Blimp Coffee pot

 

 

109

 

 

 

 

APPENDIX 8
(continued)

 

 

 

Tambourine

 

 

Spatula

 

 

 

 

Nose

 

Thumb

 

 

 

 

Kite

 

 

 

 

 

 

110

a?

 

APPENDIX 8

(continued)

 

 

 

Gun, Pistol

 

 

ill

Key

 

 

 

     

Scissors

 

 

Tweezers

 

 

 

 

 

 

 

 

 

Thermometer

 

Ill

 

 

 

 

APPENDIX 8
(continued)

 

 

   

Typewriter Banjo

 

 

 

 

 

 

Stool Candle

 

 

 

 

 

 

 

 

 

   

Ladder Flashlight

 

 

112

 

APPENDIX 9

 

 

 

MUSIC HELPS APHASIA

Research Project

 

The effectiveness of Melodic Intonation Therapy (MIT) and “Tera
peutic Singing (TS) has been established in past research since the
1970s in improving the speech production of persons with aphasia.

  
 
 
   

Seeking individuals who have expressive aphasia to participate in
research project. Participants would receive 16 sessions of music
therapy targeted at increasing functional communication. Each
session will include a brief pre- and post-test during session.

 

Where: MSU Music Therapy Clinic
Home visit Available

Tuned: 50 min. 16 sessions

Coat: Free

Redearcber: Emily YeaJu Rhee

Ifyou have questions, you are encour-
aged to contact me,
Mam 517 ~643-
0651at Michigan State Univ.
Music Therapy Department

"*1

 

 

 

 

 

 

 

 

113

 

 

 

APPENDIX 10
CONSENT FORM

The Eflectiveness of Melodic In_tonation Therapy and Therapeutic Singing on Functional
Communication Skills tor Adults with Expressive Aphasia

You are invited to be in a research study of music therapy about the effectiveness of
Melodic Intonation Therapy and Therapeutic Singing on Functional Communication Skills. You
were selected as a possible participant because of your diagnosis of expressive aphasia. We ask
that you read this form and ask any questions you may have before agreeing to be in the study.

This study is being conducted by YeaJu Rhee who is a graduate student in the music
therapy department at Michigan State University.

Background Intormation

The purpose of this study is to ascertain the effectiveness of Therapeutic Singing and
Melodic Intonation Therapy in improving the functional communication skills of adults with
expressive aphasia.

The problems to be investigated by this study are as follows: (1) to determine whether
Therapeutic Singing (TS) combined with Melodic Intonation Therapy (MIT) helps adult clients
who are diagnosed with expressive aphasia as measured by the quality of their functional
communication skills; (2) to determine whether MIT treatment itself helps clients improve the
quality of their fimctional communication skills; (3) to compare the effectiveness of TS combined
with MIT and MIT alone in increasing functional communication skills; and (4) to investigate
whether TS enhances the benefits of MIT treatment.

Procedures:
If you agree to be in this study, we would ask you to do the following things:

The procedures will consist of Melodic Intonation Therapy (MIT) in the first series of
sessions and Therapeutic Singing (TS) combined with MIT treatment in the second series of
sessions. For MIT treatment, you will be asked to sing and speak selected word phrases with a
therapist, and for TS interventions, you will be asked to choose the songs that would like to sing,
and participate in singing activity with the therapist. Previous music experience or training is not
required for your participation in this study.

Two measurement procedures will be used in a pre-test and post-test for each session: A
Functional Communication Measure (F CM) and Melodic Intonation Therapy Measure (MITM).
FCM will be composed of two different parts: Spontaneous Speech and Naming. In the
Spontaneous speech part, you will be asked to choose the questions you would like to talk about
and will have 5 minutes to talk in each pre and post-test. In the Naming part, picture stimuli
sheets will be given to identify by name. The MIT measure will be conducted while subjects are
participating in the treatment sessions. All sessions will be videotaped for the purpose of data
collection.

The intervention period will consist of between 16 and 20 individual or small group
sessions, three times or five times a week for 50 minutes, including pre and post-tests. The period
will be divided into two different terms of interventions of 9 to 11 sessions each: 1) MIT in first
series; 2) TS combined with MIT in second series.

For the purpose of gathering data, every session for every subject will be videotaped. In
order to be included in this study each subject must agree to be taped

114

 

APPENDIX 10
(Continued)

Risks and Benefits at being in the Studz

The study has no known risks at all. Potential benefits are those that arise from
participation of enjoyable musical experiences. There may be some gains in functional speech
resulting from the treatments though they are not guaranteed.

Confidentiality:

The records of this study will be kept private. In any sort of report we might publish, we
will not include any information that will make it possible to identify a subject. Research records
will be stored securely in a locked area, and only researchers will have access to the records. The
videotapes will be used for the purposes of research, and they will be erased after completion of
this research project.

Voluntary Nature at the Study:

Participation in this study is voluntary. Your decision whether or not to participate will
not affect your current or future relationship with any health care facilities or treatment providers.
If you decide to participate, you are free to not answer any question or withdraw at any time with
out affecting those relationships.

Contacts and Questions:

The researcher conducting this study is YeaJu Rhee. You may ask any questions you

have now. If you have questions later, you are encouraged to contact me, 517-643-0631,
Rheeyea@msu.edu at Michigan State University Music Therapy Department, or Associate
Professor Roger Smeltekop, 517- 355-6753, Smeltokop3@msu.edu

If you have any questions or concerns regarding this study and would like to talk to
someone other than the researcher(s), you are encouraged to contact the Michigan State
University’s Human Research Protection Program at 517-355-2180, Fax 517-432-4503, or e-mail
irb@msu.edu or
regular mail at 202 Olds Hall, MSU, East Lansing, MI 48824.

You will be given a copy of this information to keep for your records.

Statement of Consent:

I have reviewed the above information. I have had opportunity to ask questions and have
received answers. I consent to participate in the study.

Signature: Date:

 

Signature of parent or guardian: Date:

 

 

Signature of Investigator: Date:

 

 

115

ET-.. '9‘ "E a

 

APPENDIX 1 1

Spoken Language Comprehension
ASHA. (2003). FCMs. National outcomes Measurement System (NOMS):
Adult Speech-language Pathology User’s Guide.

LEVEL 1: The individual is alert, but unable to follow simple directions or respond to
yes/no questions, even with cues.

LEVEL 2: With consistent, maximal cues, the individual is able to follow simple
directions, respond to simple yes/no questions in context, and respond to simple words or
phrases related to personal needs.

LEVEL 3: The Individual usually responds accurately to simple yes/no questions. The
individual is able to follow simple direction out of context, although moderate cueing is
consistently needed. Accurate comprehension of more complex directions/messages is
infrequent.

LEVEL 4: The individual consistently responds accurately to simple yes/no questions
and occasionally follows simple directions without cues. Moderate contextual support is
needed to understand complex sentences/messages. The individual is able to understand
limited

LEVEL 5: The individual is able to understand communication in structured
conversations with both familiar and unfamiliar communication partners. The individual
occasionally requires minimal cueing to understand more complex sentences/messages.
The individual occasionally initiates the use of compensatory strategies when
encountering difficulty.

LEVEL 6: The individual is able to understand communication in most activities, but
some limitations in comprehension are still apparent in vocational, avocational, and
social activities. The individual rarely requires minimal cueing to understand complex

sentences. The individual usually uses compensatory strategies when encountering
difficulty.

LEVEL 7: The individual’s ability to independently participate in vocational,
avocational, and social activities are not limited by spoken language comprehension.
When difficulty with comprehension occurs, the individual consistently uses a
compensatory strategy.

116

 

APPENDIX 12

Spoken Language Expression
ASHA. (2003). FCMs. National outcomes Measurement System (NOMS):

Adult Speech-language Pathology User’s Guide.

LEVEL 1 : The individual attempts to speak, but verbalizations are not meaningful to
familiar or unfamiliar communication partners at any time.

LEVEL 2: The individual attempts to speak, although few attempts are accurate or
appropriate. The communication partner must assume responsibility for structuring the
communication exchange, and with consistent and maximal cueing, the individual can
only occasionally produce automatic and/or imitative words and phrases that are rarely
meaningful in context.

LEVEL 3: The communication partner must assume responsibility for structuring the
communication exchange, and with consistent and moderate cueing, the individual can
produce words and phrases that are appropriate and meaningful in context.

LEVEL 4: The individual is successfully able to initiate communication using spoken
language in simple, structured conversations in routine daily activities with familiar
communication partners. The individual usually requires moderate cueing, but is able to
demonstrate use of simple sentences (i.e., semantics, syntax, and morphology) and rarely
uses complex sentences/messages.

LEVEL 5: The individual is successfully able to initiate communication using spoken
language in structured conversations with both familiar and unfamiliar communication
partners. The individual occasionally requires minimal cueing to frame more complex
sentences in messages. The individual occasionally self-cues when encountering
difficulty.

LEVEL 6: The individual is successfully able to communicate in most activities, but
some limitations in spoken language are still apparent in vocational, avocational, and
social activities. The individual rarely requires minimal cueing to frame complex
sentences. The individual usually self-cues when encountering difficulty.

LEVEL 7: The individual’s ability to successfully and independently participate in

vocational, avocational, and social activities is not limited by spoken language skills.
Independent functioning may occasionally include use of self-cueing.

ll7

 

APPENDIX 13

Fluency
ASHA. (2003). FCMs. National outcomes Measurement System (N OMS):
Adult Speech-language Pathology User’s Guide.

LEVEL 1: Fluency is so disrupted that speech is often not firnctional for communication.
Attempts at speech communication are extremely labored in all situations, which renders
the speaker virtually unintelligible. Alternative means of speaking are used most of the
time. Listeners avoid spoken interaction with the individual.

LEVEL 2: Speech is functional most of the time, but labored in many day-to-day
situations due to extended disruptions of speech flow which sometimes render the
individual difficult to understand. Participation in vocational, avocational, and social
activities requiring speech is reduced overall. Listener discomfort is evident throughout
conversational interactions.

LEVEL 3: Speech is functional. Dysfluencies are evident in all situations, but are
particularly frequent in problem situations. Vocational, avocational, and social
participation requiring speech is occasionally reduced overall, and significantly reduced
within what the individual perceives as problem situations. Some listener discomfort is
evident throughout interactions.

LEVEL 4: Speech is functional for communication, but there is extreme situational
variation. The frequency and severity of disruptions of speech flow within problem
situation is distracting most but not all of the time. Vocational, avocational, and social
participation requiring speech is limited most of the time in problem situations. Listeners
are often aware of fluency difficulty.

LEVEL 5: Speech is firnctional for communication, and fluency can be maintained in
some situations. Self-monitoring is inconsistent. The frequency and severity of
disruptions of speech flow within problem situations is distracting some of the time.
Speech difficulties are noticeable when they occur, and sometimes limit vocational,
avocational, and social activities requiring speech in problem situations. Listeners are
occasionally aware of fluency difficulties relative to particular situations.

LEVEL 6: Speech is functional for communication, and fluency can be maintained most
of the time. Self-monitoring is consistent. Vocational, avocational, and social activities
requiring speech is not restricted most of the time. Listeners are infrequently aware of
fluency difficulties even in problem situations.

LEVEL 7: Disruptions in speech flow do not call attention to the speaker, and

participation in activities requiring speech is not limited. May include self-monitoring as
needed.

118

APPENDIX 14 & 15

Raw Data Tables
&
Statistics Data Tables
For L ’ Goal Areas

119

 

In

APPENDIX 14 -1

Sub 'ect L ’5 Raw Data 0 Bi ht Variables

Subject USS?!“

 

Session 1 2 3 4 13 14 15 16

l-pre 4 m. 4 m. 5 m. 6 m. 7 m. 6 m. 5 m. 6 m. 6 m. 6 m. 7 m. 6 m. 8 m. 6 m. 7 m. 10m.
27 s. 41 s. 36 s. 3 s. 47 s. 30 s. 19 s. 11 s. 4 s. 55 s. 50 s. 44 s. 17 s. 58 s. 17 s. 44 s.
l-post 5 m. 8 m. 8 m. 6 m. 7 m. 6 m. 7 m. 6 m. 7 m. 10m. 6 m. 9 m. 9 m. 10m. 10m. 10m.

315. 65. 105. 255. 17s. 25. 44s. 95. 185. 36s. 445. 54$. 35. 34s. 135. 505.

Z-pre 19 24 18 21 18 20 16 18 20 10 17 17 14 15 20 23
2-post 26 28 25 20 26 12 27 20 21 31 17 20 14 20 33 37

3-pre 115 208 225 191 435 288 258 313 314 141 258 208 389 348 418 761
3-post 242 411 385 307 405 293 359 296 399 303 208 355 379 531 577 558

2-1-pre 11 23.7 25 21.7 67 38 33.7 49 33 19.7 29.7 22 54.7 41 48 93
2-1-post 24.3 38 33.7 44 37 56 28 40.7 36 17.3 22 43 50.7 50 59.7 40

3-1-pre 25.8 44.4 48.9 31.6 55.9 44.3 48.5 50.6 51.8 20.4 32.9 30.9 47 50 57.4 70.9
3-1-post 43.9 50.7 47.1 47.8 55.6 48.6 46.4 48.1 54.7 28.6 30.9 35.9 41.9 50.3 56.5 51.5

 

1. Total Length of Samples, 2. Total Number of Utterances, 3. Total Number of Words,
2-1. Average Length of Utterances, 3-1. Average Number of Words per Minute

Subectflamtb

Session 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

 

 

1)—pre 3 4 3 4 4 3 6 4 4 4 6 S 4 5 6 7
1)-post 6 6 7 6 7

2)-pre 1 1.8 2.3 1.3 1.8 2 0.7 1.5 1.75 2 1 0.6 0.5 1.2 0.5 0.7
2)-post 0.5 0.7 0.7 1.2 0.6 0.2 0.6 0.3 0.7 0.7 0.5 0.6 1 0.6 1 0.3

 

 

1) Total Number of Items Named
2) Average Number of Prompt per ltem Named

Suebzct [Us MW

session 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

 

M1T(%) 53 48 58 67 65 58 63 56 65 62 80 82 75 81 70 86
Level 11 11 11 ll 11 11 ll 11 11 ll 11 11 11 111 111 111

 

120

 

APPENDIX 14 - 2
(Continued)

Subiect R ’s Raw Data at Seven Variables

Subject RSponIneousSpeeIhtestb

 

 

 

 

 

 

 

 

Session 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1'13"! 4 m. 6 m. S m. 6 m. 6 m. 6 m. S m. 8 m. 7 m. S m. 6 m. 4 m. 4 m. 4 m. 9 m. 4 m.
52 s. 25 s. 30 s. 33 s. 56 s. 29 s. 48 s. 20 s. 47 s. 54 s. 53 s. 33 s. 40 s. 49 s. 4 s. 28 s.
1-post 6 m. 5 m. 8 m. 9 m. 9 m. 6 m. 4 m. 6 m. 9 m. 7 m. 8 m 7 m. 8 m. 5 m. 6 m. 3 m.
515. Is. 42 s. 455. 355. 535. 205. 445. 505. 13s. ' 23s. 31 s. 13 s. 42$. 75.
Z-pre 9 18 16 8 11 12 13 12 16 12 13 11 10 12 13 14
Z-post 12 14 10 7 16 11 11 16 19 19 15 13 11 12 19 19
3-pre 128 244 270 208 197 217 238 323 341 189 304 226 152 194 432 203
3-post 240 183 283 233 318 194 224 234 414 237 351 275 366 245 323 370
2-1-pre 29 40.3 36 45 28 43 35.7 82.3 47.7 31 50.7 39 23.7 .36.7 71 26.7
2-1-post 41.7 21 58 63 61 32 42 33 52 27 45.7 40.7 65.7 39.7 38 41.7
3-1-pre 26.3 38.0 49.1 31.8 28.4 33.5 41.0 38.8 43.8 32.0 44.7 49.7 32.6 40.3 47.6 45.4
3-1-post 35.0 36.5 32.5 23.9 33.2 28.2 51.7 34.8 42.1 32.8 43.9 37.2 43.0 39.4 48.2 45.6
1. Total Length ofSamples, 2. Total Number of Utterances, 3. Total Number of Words,
2-1. Average Length of Utterances 3-1. Average Number of Words per Minute
ect 83 Test
Session 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1)-pre 43 46 57 54 35 53 43 38 53 44 48 SS 49 32 49 54
1)-post 46 53 63 63 46 59 52 69 52 43 49 SS 54 51 59 51
2)-pre 0 0.02 0 0.02 0.14 0.06 0.09 0.05 0 0.05 0.04 0 0.02 0.12 0.02 0
2)-post 0.09 0 0.03 0.02 0.07 0.03 0.06 0.04 0.04 0.05 0.06 0 0.04 0.06 0.05 0

 

 

1) Total Number of Items Named
2) Average Number of Prompt per Item Named

121

 

II
1

I
e:

L_.

 

 

Subject L ’s Paired t-test Data For Goal I

(Table 13-1 — The second-half period in L’s paired t-test)

APPENDIX 15-1

 

 

 

 

 

 

 

 

 

 

 

 

 

Paired Samples Statistics
Std. Error
Mean 11 Std. Deviation Ilean—
I.Total Length of Samples Pair TS-mffmj 7.6021 8 1.43475 .50726
1 rsrrrrrosu 9.4000 8 1.57765 .55768
2.Tota1 Number of Utterances Pair TSlll'I‘J’reJ 110000 8 4.07080 1.43925
New Number of Words “Istronst2 24.1250 8 8.35699 2.95464
Pair TSIII'I‘J’rej 354.6250 8 188.41591 66.61508
'I‘SIIl'U’osLS 413.7500 8 131.30200 46.42227
average Length of Pair 131111171621 42.6375 8 23.68127 8.37259
mm“ 4 TSMITJ’OSLZJ 39.8375 8 14.47490 5.11765
5.Average Number of words Pair TSIIII‘_Pre_3_1 45.1625 8 16.25677 5.74764
per Minute 5 ’I‘SIIITJ‘osLSJ 43.7875 8 10.98771 3.88474
6.TotalNumber 0mm, Pair Namingrre121911 5.1250 8 1.12599 .39810
Named 6 IIaminLPosLLZhalf 9.7500 8 1.48805 52610
I7.Average NiimbeirOf d Pair IlaminLPreJJhalf 1.0313 8 .57875 .20462
“"“P‘S P“ em 8““ 7 ”mustard: .6750 8 23755 06399
(Table 13-2 — The second-half period in L’s paired t-test)
Paired Samples Test
Pairedfierences
95" Confidence
Interval of the
Std. Error DiffHF-‘L
Mean Std. Deviation Mean Lower Upper t df Sig. (Z-tailed)
1:" gmmijgfl -1.79793 1.79794 63567 3.30104 -.29481 -2.828 7 .026
g” fiflj‘ofiz -7.12500 7.88194 27866911371447 -.63553 -2.557 7 .038
1;“ Efigfifa 59.12500 13611491 48.12389-17291991 64.66991 -1.229 7 .259
1:“ Eligiiifilf 2.80000 22.94641 7.90029 L16.88123 2148126 .354 7 .788
E.“ 2536:5511 1.37500 8.39043 2.96646 -5.63957 8.38957 .464 7 .657
:9 fiflgg—tfl’ffl’ -4.62500 1.06066 .37500 -5.51173 -3.73827 -12.333 7 .000
1;“ gflgfifi’fh‘ .35625 .65978 23327 -.19534 .90784 1.527 7 .171

 

 

 

 

 

 

 

 

 

 

(Pair 1) 1.Total Length of Samples, (Pair 2) 2.Total Number of Utterances, (Pair 3) 3.Total Number of
Words. (Pair 4) 4. Average Length of Utterances, (Pair 5) 5. Average Number of Words per Minute, (Pair
6) 6.Total Number of Items Named, (Pair 7) 7.Average Number of Prompts per Item Named.

122

 

 

(Table l4-1 — The first-half period in L’s paired t—test)

l.Total Length of Samples

2.Total Number
of Utterances

3.Total Number of Words
4.Average Length of
Utterances

5.Average Number of
Words per Minute

6.Total Number of Items
Named

7.Average Number of
Prompts per Item Named

 

APPENDIX 15-2

Subiect L ’s Paired t-test Data For Goal II

 

 

 

Paired Samples Statistics
Std. Error
Mean ll Std. Deviation Mean
Pair MlT_Pre_l 5.857 8 1.06716 .377?
1 1111138581 6.9250 8 1.02412 .36208
Pair 14101882 19.2500 8 2.43487 .86086
100198512 23.0000 8 5.37188 1.89926
Pair lilT_Pre_8 254.1250 8 95.36462 33.71648
3 MIT_POSL3 337.2500 8 61.44626 21.72453
Pair liTI‘_Pr6_2_l 33.6376 8 17.71367 6.26273
4 MILPOSLZJ 37.7125 8 9.81114 3.46876
Pair MIT.Pre_3_1 43.7500 8 10.10078 3.57116
5 MIT_POSL3_1 48.5250 8 3.45243 1.22062
Pair iiaririrraPraLiirair 3.8750 8 .99103 .35038
6 MamingJJOSLLIhatf 6.6250 8 1.06066 .37500
Pair NaminLPrellhalf 1.5500 8 .53184 .18803
7 iiarrriraPosLLnraii .6000 8 .30237 .10690

 

 

 

 

 

 

 

 

 

123

APPENDIX 15-2
(Continued)

Subiect L ’s Paired t-test Data F or Goal II

(Table 14-2 — The first half period in L’s paired t-test)

 

 

 

 

 

 

 

 

 

 

 

 

 

Paired Samples Test
Paired Differences
95% Confidence
Interval of the
Std. Error Difference
Mean Std Deviation Mean Lower Upper t Sig. (2-tailed)
‘Pairl MII‘_Pre.l-MII‘_PosL. -1.10429 1.51606 .53601 -2.37175 .1631? '2060 7 .078
P3112 Ml'I‘.Pre.2-111'I‘.PosL1-3.75000 6.04152 213600 8.80084 1.30084 -1.756 7 .123
P8113 1111‘.Pre_3-111'1‘_P05Ur83.12500 86.62800 30.62762-15554782 40.70218 -2.714 7 .030
Pair4 Ml'I‘_Pre.2_1-111T.

PosL2_l -4.07500 17.51283 6.19172 318.71609 10.56609 -.658 7 .531
PM 03552-314” -4.77600 8.27919 2.92714 -11.69668 2.14658 -l.631 7 .147
Pairfi llamingJ’reJJhalf-

P8117 NaminLPreJJhalf- .,

llaminLPosLZlhalf .95000 .64807 .22913 .40820 1.49180 4.146 4 .004

 

(Pair 1) l.Total Length of Samples, (Pair 2) 2.Total Number of Utterances, (Pair 3) 3.Total Number of
Words. (Pair 4) 4. Average Length of Utterances, (Pair 5) 5. Average Number of Words per Minute,
(Pair 6) 6.Total Number of Items Named, (Pair 7) 7.Average Number of Prompts per Item Named.

124

 

APPENDIX 15-3

Subiect L ’s Paired t-test Data F or Goal 111

(Table 15-1. Independent t-test for L’s Goal 119

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Group Statistics

Std. Error

Group 11 Mean Std. Deviation Mean
l.Total Length of Samples 131le 1 8 1.1043 1.51606 .53601
2 8 1.7979 1.79794 .63567
Z-Total Number of Did 1 8 3.7500 6.04152 2.13600
Uttc’ances 2 8 7.1250 7.88194 2.78669
310““ Number of Words 12151.3 1 8 83.1250 86.62800 30.62762
2 8 59.1250 136.11491 48.12389
4-Average Length of Diff.2_1 1 8 4.0760 17.51283 6.19172
”“6”““5 2 8 -2.8000 22.34541 7.90029
Words per Minute 2 8 -1.3750 8.39043 2.96646
6.Total Number of Items Naming_Diff_1 l 8 27500 '70711 25000
Named 2 8 4.6250 1.06066 .37500
7. Average Number of 11ammg_D1ff.2 l 8 -.9500 .64807 .22913
Prompts per Item Named 2 8 '.3563 .65973 .23327

 

 

 

Group 1 — Post-tests minus Pre-tests in the first-half sessions (lst— 8th)
Group 2 — Post-tests minus Pre-tests in the second-half sessions (9th -l6th)

125

 

 

APPENDIX 15-3
(Continued)

Subiect L ’s Paired t-test Data For Goal III

(Table 15-2. Independent t-test for L’s Goal III)

 

 

 

 

 

 

 

 

 

 

Independent Samples Test
Lflene'sTestfor
liqualibv of Variances t-test for 110116110 01 Means
95% Confidence
11118408101016
Mean Std. Error Difference
F Sig. t df Sig.(2-tailed) Difference Difference Lower Upper
m Equalvariances .
8551111841 .646 .435 -.834 14 .418 169364 .83149 '2.44701 1.05973
Equalvariances ..
not assuned -.834 13.612 .419 '.69364 .83149 '2.48149 1.09452
0111] Equalvariances
assumed 1.449 .249 1961 14 .353 '3.37500 3.51114 40.90565 4.15565
Equalvariances
_ not assailed -.961 13.115 .354 '3.37500 3.51114 '1095363 4.20363
M3 Eqma’m’ 1.815 .199 .421 14 .680 24.00000 57.04349 98.34812 146.34812
assumed
Equalvariances
L‘ not assuned .421 11.871 .681 24.00000 57.04349-100.43654 14843654
Difl_2_1 Equal variances .,
assailed .000 .998 .685 14 .505 6.87500 10.03753 -14.65337 28.40331
Equalvariances .,
_ not assumed .685 13.244 .505 6.87500 1003453 14.76928 28.51928
0111.31 Equal variances
mm .141 .713 1.476 14 .162 6.15000 4.16750 2.78839 15.0839
Equalvariances
not assumed 1.476 13.998 .162 6.15000 4.16750 '2.78854 15.08854
liammLDile Equal‘it;names 2.011 .178 '4160 14 .001 -1.87500 .45069 284164 -.90836
assune
Equalvariances ..
notassumed 4.160 12.196 .001 -1.87500 .4506!) -2.85523 -.89474
“MLDM Eqmmc“ .015 .904 -1.816 14 .091 -.69376 32698 -1.29505 .10756
assumed
Equalvariances . . . .
notassumed 'l.8l6 13.996 .091 -.59375 32698-129507 .10/57

 

 

 

 

 

 

 

 

 

 

 

 

Group 1 — Post-tests minus Pre-tests in the first—half sessions ( 1 358‘“)

Group 2 — Post-tests minus Pre-tests in the second-half sessions (9m-l6m)

(Difference-l) l.Total Length of Samples, (Difi-Z) 2.Total Number of Utterances, (Diff-3) 3.Total Number
of Words. (Diff-Z- l) 4.Average Length of Utterances, (Diff-3- l) 5.Average Number of Words per Minute,
(Naming-Diff-l) 6.Total Number of Items Named, (Naming-Diff-Z) 7.Average Number of Prompts per
Item Named

126

REFERENCES

127

REFERENCES

Albert, M., Sparks, R., & Helm, N. (1973). Melodic intonation therapy for aphasia.
Archives Neurology, 29(2), 130-131.

Baker (2000). Modifying the Melodic Intonation Therapy Program for Adults With
Severe Non-fluent Aphasia. Music Therapy Perspectives, 18(2), 110-1 14.

Baker, & Tamplin, J. (2006). Music therapy methods in neurorehabilitation: a clinician ’5
manual. London: J. Kingsley Publishers.

Belin, P., Van Eeckhout, P., Zilbovicius, M., Remy, P., Francois, C., Guillaume, S., et al.
(1996). Recovery from nonfluent aphasia afier melodic intonation therapy: a PET
study. Neurology, 47(6), 1504-1511.

Benson, D., Dobkin, B., Rothi, L., Helm-Estabrooks, N., & Kertesz, A. (1994).
Assessment: Melodic intonation therapy. Neurology, 44, 566-568.

Benton, A. (1977). The amusias. Music and the Brain. London: William Heinemann.

Bonakdarpour, B., Efiekharzadeh, A., & Ashayeri, H. (2003). Melodic intonation therapy
in Persian aphasic patients. Aphasiology, 1 7(1), 75-95.

Borchgrevink, H. (1982). Prosody and musical rhythm are controlled by the speech
hemisphere. Music, mind, and brain: The neuropsychology of music, 151-157.

Carroll, D. (1996). A study of the eflectiveness of an adaptation of melodic intonation
therapy in increasing the communicative speech of young children with down
syndrome. McGill University, Montreal.

Cohen, N. (1992). The effect of singing instruction on the speech production of
neurologically impaired persons. Journal of Music Therapy, 29(2), 87-102.

Cohen, N., & Ford, J. (1995). The effect of musical cues on the nonpurposive speech of
persons with aphasia. Journal of Music Therapy, 32, 46-46.

Cohen, N., & Masse, R. (1993). The application of singing and rhythmic instruction as a
therapeutic intervention for persons with neurogenic communication disorders.
Journal of Music Therapy, 30, 81-81.

Crocker, D. B. (1958). Using music in a speech therapy program. Music Therapy 195 7,
103-108.

Darrow, A., & Starker, G. (1986). The effect of vocal training on the intonation and rate

of hearin g impaired children's speech. Journal of Mu sic Therapy, 23(4), 194-
201.

128

 

Engineer, N., Percaccio, C., Pandya, P., Moucha, R., Rathbun, D., & Kilgard, M. (2004).
Environmental Enrichment Improves Response Strength, Threshold, Selectivity,

and Latency of Auditory Cortex Neurons. Journal of Neurophysiology, 92(1), 73-
82.

Galloway, H., & Kraus, T. (1982). Melodic intonation therapy with language delayed
apraxic children. Journal of MusicTherapy. 19(2), 102—113.

Garrett, K., & Beukelman, D. (1995). Changes in the interaction patterns of an individual
with severe aphasia given three types of partner support. Clinical Aphasiology,
23, 23 7-25 1.

Garrett, K., & Lasker, J. (2005). The multimodal communication screening test for
persons with aphasia (MCST-A): Retrieved.

German, D. (1992). Test of adolescent/adult word finding: TA WF: DLM Teaching
Resources/Taskmaster.

Gerstman, H. (1964). A Case Of Aphasia. J Speech Hear Disord, 29, 89-91.
Gleason, J ., & Goodglass, H. (1984). Some Neurological and Linguistic

Accompaniments of the Fluent and Nonfluent Aphasias. Topics in Language
Disorders, 4(3), 71-81.

Glover, H., Kalinowski, J ., Rastatter, M., & Stuart, A. ( 1996). Effect of instruction to sing
on stuttering frequency at normal and fast rates. Perceptual and motor skills,
83(2), 511-522.

Goodglass, H., & Kaplan, E. (1972). The Assessment of Aphasia and Related Disorders:
Lea & Febiger, US.

Hachinski, K., & Hachinski, V. (1994). Music and the brain. CMAJ: Canadian Medical
Association Journal, 151(3), 293.

Hebert, S., Racette, A., Gagnon, L., & Perets, I. (2003). Revisiting the dissociation
between singing and speaking in expressive aphasia. Brain, 126(8), 1838-1851.

Helfrich—Miller, K. (1994). A clinical perspective: melodic intonation therapy for
developmental apraxia. Clin Commun Disord, 4(3), 175-182.

Helm-Estabrooks, N. (1992). ADP: Aphasia Diagnostic Profiles: Riverside Pub. Co.

Helm-Estabrooks, N., & Albert, M. (1991). Manual of aphasia therapy: Pro-Ed.

129

 

Hobson, M. (2006). The Collaboration of Music Therapy and Speech-Language
Pathology in the Treatment of Neurogenic Communication Disorders: Part 1—
Diagnosis, Therapist Roles, and Rationale for Music. Music Therapy
Perspectives, 24(2), 58-65.

Jackson, J. (1871). Singing by speechless (aphasic) children. Lancet, 2, 430-431.

Jackson, S., Trehame, D., & Boucher, J. (1997). Rhythm and language in children with
moderate learning difficulties. International Journal of Language &
Communication Disorders, 32(1), 99-108.

Kaplan, E., Goodglass, H., & Weintraub, S. (2001). Boston Naming Test Record Booklet:
Philadelphia: Lippincott, Williams and Wilkins.

Keenan, J. (1987). First Aid for Aphasia: Home Exercises.

Keith, R. L., & Aronson, A. E. (1975). Singing as therapy for apraxia of speech and
aphasia: report of a case. Brain Lang, 2(4), 483-488.

King, B. (2007). Language and Speech: Distinguishing Between Aphasia, Apraxia, and
Dysarthria in Music Therapy Research and Practice. Music Therapy Perspectives,
25(1), 13-18.

Lathom, W., Edson, S., & Toombs, M. (1965). A coordinated speech therapy and music
therapy program. Journal of Music Therapy, 3, 118—120.

Laughlin, S., Naeser, M., & Gordon, W. (1979). Effects of three syllable durations using
the melodic intonation therapy technique. Journal of Speech, Language and
Hearing Research, 22(2), 31 1.

Leonard, H. (1985). The Best Songs Ever: Hal Leonard Corporation.

Leung, K. (1985). Improving the Use of Suprasegmentals with Severely Handicapped
Children through Music and Movement. BC Journal of Special Education, 9(2),
1 3 1-143.

Levitin, D. (2006). This is your brain on music: The science of a human obsession:
Dutton Books.

Lezak, M., Howieson, D., Loring, D., Fischer, J ., & Hannay, J. (1995).
Neuropsychological assessment: Oxford University Press New York.

Lucia, C. (1987). Toward developing a model of music therapy intervention in the
rehabilitation of head trauma patients. Music Therapy Perspectives, 4, 34-39.

130

Mackay, J ., & Mensah, G. (2004). The Atlas of Heart Disease and Stroke: World Health
Organization.

Naeser, M., & Helm-Estabrooks, N. (1985). CT scan lesion localization and response to
melodic intonation therapy. Cortex, 21, 203-223.

Ogden, J. (1982). A Study of the Value of Musical Activities in Articulation Therapy.
University of Kansas, Art and Music Education and Music Therapy.

Overy, K., Norton, A., & Ozdemir, E. (2005). Activation of left inferior frontal gyrus
afier melodic intonation therapy in a Broca’s aphasia patient. Program N. 595.7
2004 Abstract Viewer/Itinerary Planner. Society for Neuroscience. Washington,
DC.

Patel, A., Peretz, I., Tramo, M., & Labreque, R. (1998). Processing prosodic and musical
patterns: a neuropsychological investigation. Brain and Language, 61(1), 123-
144.

Patterson, P. B. A. (2004). Rise Up Singing. Bathlehem, PA: Sing Out Publications.

Peretz, I., & Hébert, S. (1995). Music and the Mind Machine. The psychophysiology and
psychopathology of the sense fo music: Music processing after brain damage: The
case of rhythm without melody. Berlin: Springer-Verlag.

Popovici, M. (1995). Melodic intonation therapy in the verbal decoding of aphasics. Rom
J Neurol Psychiatry, 33(1), 57-97.

Prior, M., Kinsella, G., & Giese, J. (1990). Assessment of musical processing in brain-
damaged patients: Implications for laterality of music. Journal of Clinical and

Experimental Neuropsychology, 12(2), 301-312.

Racette, A., Bard, C., & Peretz, I. (2006). Making non-fluent aphasics speak: sing along!
Brain, 129(10), 2571.

Sacks, O. (2006). The power of music. Brain, 129(10), 2528.

Sacks, O. (2008). Musicophilia: Tales of music and the brain: Vintage.

Schlang, G., Marchina, S., & Norton, A. (2008). From Singing To Speaking: Why
Singing may lead to recovery of expressive language function in patients with
Broca' aphasia Music Perception, 25(4), 315-323.

Sparks, & Deck, L. (1986). Melodic intonation therapy.

Sparks, Helm, N., & Albert, M. (1974). Aphasia rehabilitation resulting from melodic
intonation therapy. Cortex, 10(4), 303-316.

131

Sparks, & Holland, A. (1976). Method: Melodic Intonation Therapy for Aphasia. Journal
of Speech and Hearing Disorders, 41(3), 287.

Stone, A. (2009). America's All-Time Favorite Songs Amsco Publications.

Thaut, M. (2005). Rhythm, music, and the brain: scientific foundations and clinical
applications: Routledge.

Vanderark, S. B., J. S Stewart, S. (1986). Utilizing music in speech therapy with a
cerebral palsied girl. Contricutions to music education, 13, 1—6.

Vargha, & Gereb (1953). Aphasie T herapie. Jene, Germany: VEB Fisher.

Weisssman, D. F. a. D. (2007). The Great Family Songbook. New York: Black Dog &
Leventhal Publishers, Inc.

Yamadori, A., Osumi, Y., Masuhara, S., & Okubo, M. (1977). Preservation of singing in
Broca's aphasia. Journal of Neurology, Neurosurgery & Psychiatry, 40(3), 221-
224.

Zatorre, R., Evans, A., & Meyer, E. (1994). Neural mechanisms underlying melodic
perception and memory for pitch. Journal of Neuroscience, 14(4), 1908-1919.

132

 

"11111131111111fllfllflujililflifllflllfil“