A COMMUNICATION MANAGEMENT STRATEGY FOR INCREASING
THE ACADEMIC AND VOCATIONAL POTENTIAL
OF THE FUNCTIONALLY DISABLED, NONVERBAL POST-SECONDARY ADULT

by
Mary Beth Way

Submitted in Partial Fulfillment of the Requirements
for the Master of Arts Degree

h Communications
Chairman, The

-j

s Committee

Date
Date

Committee Member

Date

Committee Member

Date

Zo.

Committee

Date

ACKNOWLEDGMENTS

I would like to express my appreciation to the many people
who helped make this project possible:
To the Edinboro University Council for Faculty
Development and Research for the funding they
provided for the purchase of the software used in the
study,
To Mr. Jim Whiteman, for inviting me to participate
as a colleague in this research endeavor and for all
of his assistance with the equipment and
technological aspects of the project,
To the Office of Disabled Student Services for
allowing me the use of their facilities and the
opportunity to become involved with their programs
for the disabled student population,
To all of the members of my thesis committee whose
input helped shape my ideas and develop them into
this final product,
To Michelle Aulthouse, the subject of the study and
an amazing young woman, for her enthusiasm and
incredible patience throughout the most difficult
stages of the investigation,
and a very special thanks to Mr. John Boylan, Jr. for all of
the support and encouragement he has given me over the
course of my studies at Edinboro University,

He has

challenged me, inspired me, involved me in many exceptional
clinical experiences, and facilitated the achievement of
progressively higher academic and professional goals,

At

both the undergraduate and graduate levels, he has provided
me the guidance I needed and allowed me the opportunity to
interact as a fellow professional, as well as a student.
More than a teacher and colleague, he has been an ever
supportive mentor and friend.

TABLE OF CONTENTS

Page
ABSTRACT
INTRODUCTION

1

METHODS

14

SUBJECT

14

PARTICIPANTS

21

INDEPENDENT AND DEPENDENT VARIABLES

22

EXPERIMENTAL DESIGN

22

MATERIALS

23

PROCEDURES

26

RESULTS

30

DISCUSSION

33

SUMMARY, RECOMMENDATIONS, AND CONCLUSIONS

41

APPENDICES
A.

RECORD SHEET

B.

STIMULUS SENTENCES

C.

SAMPLE OF SIDETALK AND EZK/EZT CAPABILITIES .

• •

ABSTRACT

The current study was undertaken to evaluate the
expressive communicative effectiveness of a physically
handicapped student registered with the Office of Disabled
Student Services at Edinboro University of Pennsylvania in a
variety of communicative circumstances.

The primary purpose

of the investigation was to determine the comparative
effectiveness/efficiency of verbal communication as produced
naturally by the subject versus that produced via a personal
computer and voice synthesizer.

Two additional components

of the communicative exchange, the degree of familiarity of
the communication partner with the subject and the nature of
the utterance, were examined in relation to communicative
effectiveness/efficiency as measured in terms of mean
transmission times.

Results indicated that transmission

time was not reduced when using the augmentative
communication system utilized for this investigaton, but
was, in fact, significantly increased.

The degree of

familiarity of the communication partner and the nature of
the utterance had no significant effect upon time required
for message transmission,

Discussion of several extraneous

and confounding variables is included to clarify the
significance of these findings.

INTRODUCTION

Historically, the physically handicapped have frequently
been mistaken to be members of the mentally handicapped
population.

While there are many physically disabled

individuals who do, in fact, have severe cognitive deficits
in addition to their physical disabilities, there are many
others who are cognitively well intact.

Unfortunately,

however, for any number of reasons their abilities
frequently go unrecognized and they are denied the
opportunity to develop to their full potential and lead
productive, fulfilling lives.

This is particularly true of

the severely disabled nonverbal population.

Because they

lack the means to express themselves in such a way as to be
easily understood, they are all too often assumed to be
incapable of anything worth spending the time trying to
understand.

Consequently, over the years, such individuals

have frequently been institutionalized and there have been
many cases of inappropriate placement of severely physically
disabled nonverbal persons into facilities for the mentally
and developmentally disabled, where there is virtually no
hope of ever leading a "normal” life.
As in all things, our society has progressed in this
area and a new awareness of the potential of the physically
disabled nonverbal individual has begun to develop.

In a

2
study of the interactional patterns of two-year-old children
with cerebral palsyr Staudenbauer (1985) reported that these
children were found to interact with their caregivers in a
manner similar to that of able-bodied children, and that
although parents recognized differences between interactions
with their earlier-born, able-bodied children and
interactions with their disabled children, these differences
related more to the type of behavioral output (mode of
expression) rather than to the quantity of output.

These

findings are of great importance because the professional
community has often assumed that young, severely motorimpaired individuals do not engage effectively in early
communicative behaviors.

Staudenbauer1s

results suggest

that researchers who are usually unfamiliar with the
atypical motor responses and vocalizations of their subjects
may categorize the communication attempts of subjects as a
disordered rather than a different type of expression.
While this realization helps to facilitate the recognition
of specific communicative acts on the part of disabled
persons, we have a lot further to go before we can claim to
be truly knowledgeable about the communicative functioning
of the physically disabled nonverbal population.
At this point, having recognized the communicative
potential that exists with the physically disabled nonverbal
population, the next step, obviously, is to provide a means
by which the individual can express himself and be

3
understood by his audience,

As simple as the idea may
sound, this is by no means an easy task, Recently, within
the last ten to twenty years, tremendous strides have been

made in the area of augmentative and alternative
communication strategies.

Beukelman (1987), however,

reports that "as yet, AAC [augmentative and alternative]
approaches are viewed by some nonspeaking persons,
professionals, and family members as a means of last
resort."
This brings us to yet another highly debatable question
of professional ethics,

Just as the debate has raged on

through the years between those advocating a strictly auraloral approach for the habilitation of the deaf and those who
support manual communication as the answer, we now face a
similar question with regard to the management of the
physically disabled
nonverbal population.
Many suggest that it is our responsibility as
professionals to develop strategies that maximize the
individual's current abilities, but not to implement
"special" AAC approaches that are not typically used in the
general population,

While this may be a reasonable approach

for some patients, it is certainly not practical for all.
Etiology of the communicatively handicapping condition is a
significant issue when considering management options,
example, a patient who is physically disabled and

For

4
functionally nonverbal as a result of a stroke may recover
his speech and should be encouraged to use residual
abilities in an effort to stimulate and facilitate the
rehabilitation of his communicative system.

(The term

"functionally nonverbal" in the context of this text refers
to the inability to effectively use verbal communication as
the primary communicative mode.)

Another patient,

presenting the same physical characteristics attributable to
a progressive degenerative disease such as ALS, however,
will exhaust his ability to use one communicative mode after
another as he continues to weaken.

For this individual, the

use of the continually deteriorating speech is not a
reasonable expectation.
At the other extreme, there are others in our profession
who have developed a very narrow view of what our role in
intervention with the physically disabled and functionally
nonverbal population consists of.

Beukelman (1987), in his

appropriately titled forum article "When You Have a Hammer,
Everything Looks Like a Nail", discusses

the tendency of

unidimensional programs to focus on AAC solutions without
appropriate, simultaneous efforts to facilitate speech,
vocalizations, and/or gestures,

In some cases, effective

communication techniques may even be discouraged in an
effort to implement techniques that reflect the biases of
the intervention team.
One additional faction of the professional community

5
that must be mentioned at this time is that group who may be
willing to implement AAC strategies, but who lack sufficient
knowledge of this area to do so effectively,

The results of

a clinical and demographic database report on the
nonspeaking cerebral palsied (Lafontaine and DeRuyter, 1987)
suggest that many speech language pathologists are
unfamiliar with the skills required for augmentative
communication system usage and recognize the need for
referral.

The data gathered in that study support the

findings of the ASHA Ad Hoc Committee of Communication
Processes and Nonspeaking Persons (1982) final report that
indicated relatively little information and experience is
provided during the training period preceding professional
employment.
Obviously, what is needed now is a form of total
communication, as used with the deaf, that is suited to the
needs of the population of current concern.
compromise on both sides of the issue.

This requires

And it can be

expected that, as there are still staunch proponents of each
of the two unidimensional approaches to intervention with
the hearing impaired, there will be some inflexibility among
some members of the professional community,

However, the

tremendous need for compromise on this issue cannot be
overlooked.

Beukelman (1987) repeatedly emphasizes the fact

that we must achieve a proper balance between these two
views to meet the needs of the individual client.

He

6
further concludes that:

We cannot afford to have our judgments influenced
excessively by the tendency to view either the AAC or
the speech approach as the sole solution to serve
persons with severe communication disorders. Rather, we
should provide balanced intervention services that meet
our client's communication needs given their profile of
present and potential capabilities.
(Beukelman, 1987)
Vanderheiden and Yoder (1986), in the introductory chapter
of the ASHA reference text, Augmentative Communication:

An

Introduction, state that the emphasis in this area of
concern should not be placed on the development of an
augmentative communication system, but on the development of
an effective overall communication system, which includes a
combination of speech and standard and special augmentative
communication techniques as components of the system.
Beukelman (1987) makes a very obvious and valid point
that must be considered when attempting to develop an
effective overall communication system.

He reminds us that

both normal and disabled speakers use a variety of
communication techniques as part of their overall
communication system.

As normal speakers, we develop the

skill to efficiently manage the various techniques and the
ability to match the appropriate technique with a specific
communication need,

In the case of the disabled speaker,

the developmental process is much the same but requires
assistance from those in the individual's environment to
facilitate the evolution of various techniques practically

7
available to him.

Thus, an approach to intervention that

recognizes that the communication system of a nonspeaking
person can include both limited speech as well as aided and
unaided AAC techniques will get us started in the right
direction toward the development of an effective system
appropriate to the needs of our client.
Another debatable issue within the area of augmentative
communication is the use of voice output systems.

A number

of studies have been conducted for the purpose of
determining the effects of voice output systems on the
interactive communication patterns of the functionally
nonverbal individual.

While the studies are very similar in

format and even in their results, the interpretations range
from one extreme to the other.

In fact, the results of a

study of the interaction between augmentative communication
users and unimpaired speakers (Buzolich, 1983) have been
cited in the literature by two different authors whose
interpretations of the results differ greatly.

On the one

hand, Mathy-Laikko and Yoder (1986) discuss Buzolich*s
findings in terms of how mode of output affects
communicative effectiveness.

Their comments suggest that

unimpaired listeners are negatively affected by the use of
an electronic augmentative communication device using voice
output.

They indicated that listeners made positive

evaluations of the nonelectronic alphabet board because it
allowed them to

actively participate in the message

8
construction by permitting them to guess and predict the
message, resulting in a reduction in transmission time,

The

same results, as cited by Mirenda and Beukelman (1987), were
interpreted to demonstrate that nonspeaking adults with
speech output communication devices were more dominant and
more in control of conversational interactions than were
similarly disabled alphaboard users.

They reported fewer

communication breakdowns, a decrease in disruptive guessing
by natural speech listeners, and an increase in the number
of interruptions by augmented users were evident when speech
output devices were used.
Newell (1987) took a bipolar perspective of voice output
as a component of system augmentation.

He recognized the

following as advantages of synthetic speech:

the listener

can continue to look at the speaker, the speaker can talk to
a group of people with no difficulty, and there is no
problem with an unexpected visitor about whom you have just
been impolite "hearing" what you have just been saying (as
may occur with printed output systems).

In terms of

disadvantages, he noted the fact that listeners need to rely
on their memory to decode the message since speech output
alone does not have permanence,

The listeners cannot look

back to see what has been said previously, and this can pose
tremendous problems when the speed of communication is very
slow.

It is the rate of message transmission, he states, is

the overriding difference between communication via aids and

9
normal speech communication.

Even a person with very

dexterous hands can only type at approximately one-third of
normal speaking rate, and disability can reduce "speaking”
rate to excruciatingly slow speeds which can be intolerable
for all but the most patient listener.
Baker (1988), regarding the "panacea" of voice output,
discussed two other significant factors involved in
selecting an output mode:
cognitive demand.

critical number and physical and

He identified the critical number, a term

referring to the number of actuations required to produce an
utterance, as being between two and three keystrokes.

The

critical number is used as a sort of cut off point, beyond
which interactive communication is unlikely to occur via the
device.

This is directly related to the physical and

cognitive demands placed on the user.

Using a silent

system, listeners can predict the message topic given the
first two or three characters, thereby remaining within the
boundaries of the critical number.

With a voice output

system, however, an incomplete message is not so easily
decoded by the listener,

While these comments may appear to

reflect a negative attitude toward synthesized speech, Baker
had still more to say on the subject,

Silent systems have a

certain appeal at the surface level because they are easy,
flexible, and quicker than voice output systems,

However,

much of the appeal, from the listener's perspective, stems
from this allowance for prediction and guessing.

He

10
described the silent system as being "menu driven",

In

other words, after an initial character is selected, the
communication partner provides a menu of possible topics
form which the disabled individual can select, and so goes
the interaction between the two.

The listener provides the

language base of the interaction, as well as the menu items.
This type of interaction inculcates and strokes a
communicative passivity on the part of the user, which can
result in significant deficits in socialization skills and
in the development of independent thought and opinion,

The

individual is not given the opportunity to be an
"individual" in the true sense of the word, but is rather a
product of his environment and his limited interaction with
it.
So, what is the purpose of augmentative communication
strategies?

It is not to replace previously established

functional communication skills, but rather to supplement
residual expressive modes and facilitate communicative
independence and personal growth.

Not only must an

augmentative communication device suit the individual in
terms of his physical parameters, it must meet his personal
needs with regard to social-emotional expression and
independent functioning as a communicative being.
Since the area of augmentative communication is still a
relatively new consideration in the field of speech-language
pathology, there is still a great need for continuing

11
research regarding the development and use of augmentative
communication systems and strategies,

Unlike many other

categories of communication disorders, however, the
population of interest here is very limited and not easily
accessible for research purposes.

Within a given geographic

region, it would be unlikely to find a sufficient number of
subjects to form a sample group for investigation.
Furthermore, due to the heterogeneity of the population that
might benefit from the use of AAC strategies, group studies
are not a suitable approach to the subject matter.

Bauer

(1968) emphasized the importance of single-subject designs
for the investigation of communication disorders because
"they open complex human speech and interactional disorders
to exploration, discovery, and clinical accounting."
The current investigation involves a physically disabled
functionally nonverbal student at Edinboro University of
Pennsylvania.

Although only one of the large number of

disabled students at EUP, the subject of this study presents
problems that are, in many ways, representative of those
experienced by the majority of this population.

The

disabled/handicapped students admitted to Edinboro
University are attending college for the same reasons as
nonhandicapped students:

they are here as academically

capable students seeking a higher education with aspirations
of graduating form a degree program and beginning a career
in the field of their choice.

Unfortunately,they encounter

12
a seemingly endless string of obstacles and barriers that
hinder them as they matriculate through their educational
programs.

The most prominent and persistent problem this

population encounters if that of communication.
Communicative difficulties consistently interfere with
socialization, academic performance, vocational placement
and the transition to independent living in the community at
large.

Again, although the severely handicapped nonverbal

post-secondary adult may require a communication system
that, at the very least, enables him to express the most
basic of needs to his caregivers, there is no question that
the communication needs of a cognitively functional,
academically capable adult extend infinitely beyond this
very primitive level.

In an institution of higher learning,

it is expected that the student, whether he is disabled or
able-bodied, will demonstrate certain competencies and
achieve progressively higher academic goals.

Therefore, the

functionally disabled nonverbal adult requires a
communication system that allows for the most complex
language expression imaginable.
The present study addresses three research questions.
The first and foremost of these is concerned with the
comparative communicative effectiveness/efficiency (in terms
of utterance transmission time) of natural speech produced
by the subject versus synthesized speech presented through
the use of a personal computer, a keyboard enhancement

13
software package and a voice output device,

The second

question involves the effect of the degree of familiarity
with the subject's speech output, under both conditions of
communicative mode, on the time required for transmission.
The third concern under investigation here is again
concerned with message transmission time as it is effected
by message content (whether the utterance is behavioral or
affective in nature).

This is only a preliminary study of

the efficacy of augmentative communication strategies for
use with a functionally disabled nonverbal post-secondary
adult.

A much more comprehensive assessment of the

individual's personal communication needs is necessary
before intervention strategies of any kind are implemented.

14
METHODS

SUBJECT
This investigation utilizes a single-subject design
involving a student registered with the Office of Disabled
Student Services at Edinboro University of Pennsylvania.
The subject is a nineteen year-old (D.O.B. 10/20/68) female
from Pennsylvania,

She presents with a moderate involvement

of cerebral palsy.

Physical Attributes

The subject presents with a moderate involvement of
cerebral palsy.

She is ambulatory and gross motor control

is functional for most activities of daily living,

In the

area of personal care, she is completely independent with
the exception of her need for assistance with tying her
shoes and fastening her pants,

A personal care attendant,

employed by the Office of Disabled Student Services, is
therefore made available to her in the mornings.

Once

dressed for the day, the subject's only other requirement is
minimal assistance in the cafeteria.

Due to her physical

limitations , specifically the contracted position of the
musculature of her right upper extremity, she is unable to
carry her tray through the cafeteria line.

Consequently, a

15
meal aid is made available to help her.
She also needs
occasional assistance with cutting her food, again due to
the inability to use her right arm.
Although the subject's physical limitations are rather
mild at the gross motor level, her lack of fine motor
control has resulted in severe deficits in some areas,
particularly in communication.

Her ability to communicate

through graphic means is relatively well intact.

She

demonstrates functional writing skills with her left hand,
and therefore does not require the use of an academic aid to
take notes or write out tests for her.

She has also had

some training using the Word Perfect word processing program
and could potentially use the computer to print out homework
assignments and research papers.

However, her utilization

of that training has been very limited to this point and she
does not currently use the computer on a regular basis.
With regard to non-graphic communication modes, the
subject is rather limited in the means available to her for
effective message transmission.

She lacks the fine motor

control necessary to achieve the articulatory precision
required for normal speech production,

Her speech is very

labored and highly unintelligible to the untrained listener.
She does use finger spelling and some gestures from American
Sign Language, as well as original gestures, to communicate
expressively.

With gesture, as well, motor movements are

somewhat awkward and imprecise.

However, since the

16
necessary motoric sequencing features for gestural
communication are not as refined as those required for
speech, precision is not as significant a factor,

Facial

expression is also a component of message transmission.

Typical Communicative Methods Across a Variety of

Settings

The subject reports that within her home environment,
she uses different methods of communication for interaction
with each of her parents.

She and her mother use gestural

communication as their primary mode of interaction.

Her

father, however, encourages her to speak and does not use
sign.
The subject further reports that her interaction with
friends and familiar people is typically conducted through
the use of combined verbal and gestural cues,

She relies

heavily on gesturing and finger spelling to augment her
verbal communications.

Or perhaps it is rather a matter of

using verbal output as a means of enhancing the gestural
transmission of the message,

In some cases, with those

communication partners who are very adept at interpreting
her gestures, she uses sign almost exclusively.
Among people with whom she is unfamiliar, the subject
again reports using a combination of speech and gesture.
Her gestures, however, are different than those used with
friends.

Rather than using some of the more abstract

17
gestures of American Sign Language which might be easily
understood by those familiar with that system, the subject
uses very transparent and concrete gestural representations.
By the same token, she frequently spells out letters in the
air or on any available surface to provide a more easily
recognized letter representation than those used for finger
spelling.

Generally speaking, communication partners may

facilitate proper interpretation of her intended message
through careful listening, requests for multiple repetitions
and supplemental gestural cues as required, and some
knowledge of the conversational context.
In the academic environment, the subject's interactional
patterns are very different,

They are almost nonexistent,

as reported by the subject and confirmed by course
instructors.

She does not generally ask questions or

participate in class discussions.

She further demonstrates

very little communicative contact with her classmates in the
classroom and no interaction with them outside of that
setting.

However, the subject is academically independent,

in that she does not use academic aids and does not study in
group with other students.

Environmental Factors:
Communication Partners

Attitudes and Reactions of

In the home, the subject's mother is reportedly
accepting of her daughter's severe speech impairment and is

18
content to use gesture as an alternate means of
communicating with her.

The subject's father, however, has

indicated to her that he believes verbal communication could
be very functional for her if she tried harder and
concentrated more on her speech productions.
Within the university environment, the attitudes and
reactions toward the subject's communication vary
considerably.

The subject indicates that many of her

familiar communication partners are able to understand her
easily and expedite communicative exchange through the use
of certain strategies that have proven helpful through
repeated contact with the subject,

Interaction with the

many other people on campus is not as easily facilitated.
Due to the time element involved in communicating with the
disabled student, some individuals become very impatient and
frustrated.

Others, while politely waiting for the

completion of message transmission, struggle silently trying
to interpret the message and the intended impact of the
utterance is lost.

A still greater number of the members of

the university community never establish any contact at all
with the student for a variety of reasons,

In spite of the

size of the disabled population at Edinboro University of
Pennsylvania, many nonhandicapped individuals have simply
not been exposed to the disabled students on an interactive
level.

Many more are just not comfortable communicating

with this population because they are uncertain how to react

19
to an individual's physical disabilities,

Again, it is

primarily a matter of a lack of exposure to these disabled
persons.
In terms of current receptive/expressive language
functioning, six of the eight subtests of the Test of
MQj--escent Language were administered to the subject to
establish a base from which to judge more generalized
communication patterns.

Although specific scores were not

utilized for comparison to the norms presented with the
TOAL, some general conclusions can be drawn about the
subject's current language skills,

Receptive language

abilities

Concerns of the Individual

The subject's own attitude toward her communication
abilities, is without question, one of the primary concerns
when considering implementation of an augmentative
communication device.

The subject of this study was very

motivated to improve her communicative abilities,

She

indicated that she doesn't think her verbal communication
ability is ever going to progress to the point of being
functional for her in meeting her goals.

She clearly

disagrees with her father's opinion regarding more concerted
efforts to use and improve her speech.

While her verbal

skills have been sufficient to carry her through to this

20
point, she expresses tremendous frustration at times and all
to often refrains from conversational exchange to avoid the
frustration and the embarrassment that may sometimes result.
Her personal goals also demand careful consideration.
This particular student would like to enter the field of
special education and establish a career of teaching.
Obviously, her verbal communication skills, as they
currently exist, are inadequate for such a task and it is
doubtful that her natural expressive skills will ever
develop to the point where she would be able to provide an
appropriate speech model to a classroom full of children.
An option which has been only briefly discussed with the
subject is the possibility of using a system similar to the
one proposed here with a better quality output device within
the structured environment of a classroom.

Using a personal

computer, lessons could be programmed in advance for
classroom use and easily accessed when needed.

Again, this

has been mentioned only as a possibility for future
consideration.

Another issue that must be addressed before

such things can even be contemplated is the efficacy of her
vocational goals,

Although there seems to be no established

policy within the university system to allow or disallow, or
even to encourage or discourage, the pursuit of specific
vocational goals as related to the capabilities of the
physically disabled population, this matter needs to be
addressed.

To permit a student to struggle to achieve the

21
academic requirements within a given training program when
there is no possibility of vocational
success is cruel. By
the same token, however, our disabled students have proven
to be exceptional people in some instances and it would be
equally unfair to disallow them to attempt a goal that they
maYr

fact, be able to achieve,

This issue is a very

delicate one, but cannot go unrecognized.

PARTICIPANTS

Eight persons from each of three groups based on degree of
familiarity with the communication patterns of the subject
or similarly handicapped individuals were recruited to
participate in the investigation.

The first group included

eight persons identified by the subject as friends or as
people with whom she felt comfortable interacting
communicatively.

The second group of participants consisted

of eight personal care workers employed by the Office of
Disabled Student Services at Edinboro University who have
encountered similarly handicapped individuals but who have
had limited contact, if any, with the subject.

The third

and final group consisted of eight undergraduate students at
EUP who have not had any professional or paraprofessional
experience with the disabled population on campus. None of
the participants had any previous experience with synthetic
speech.

22
INDEPENDENT AND DEPENDENT VARIABLES

Three independent variables were included in the study:
communicative mode, degree of familiarity of the listener,
and content of the utterance,

Two means of message

transmission comprised the first variable:

natural

expressive communication as produced by the subject and
synthesized speech output from the Votrax Personal Speech
System.

The second variable recognized three degrees of

familiarity of the listeners with regard to the subject and
her current communicative functions:
generally familiar, and unfamiliar.

personally familiar,
Two varieties of

utterances, behavioral and affective in nature, comprised
the third variable.

EXPERIMENTAL DESIGN

This investigation uses a between-between-within (B-B-W)
three-factor mixed design.

The effects of listener degree

of familiarity and utterance content are measured using a
between design.

The communicative mode represents the

within-subjects (repeated) factor.

Listeners in all three

groups responded to ten utterances (5 behavioral in nature,
5 affective) presented by the subject first, under the
condition of naturally produced expressive communication and

23
then, through the use of a personal computer and voice
synthesizer. The mean time required for message
transmission using each combination of conditions was
recorded, time being the dependent variable under
examination.

This design allows for analysis of the main

effect for each of the independent variables, two-way
interactions between pairs of variables and the triple
interaction of all three variables combined.

MATERIALS

In terms of hardware utilized for this study, a portable
personal computer was selected as the central piece of
equipment from which the Votrax Personal Speech Synthesizer
could be activated.

These two devices were selected for

primarily three reasons:

they are portable, affordable, and

the combination allows for an infinite number of
applications.
Since one of the purposes of conducting the study is to
consider implementation of the augmentative communication
system for regular use by the student across a variety of
situations, it was only practical to evaluate the use of a
system that would meet the criteria for

transportability.

If such a system were to be used, the student would be able
to physically manage these devices physically with little
difficulty.

24
The second and third reasons for these selections were
closely related. Although other portable voice output
systems are available for less than the combined price of
the personal computer and the Votrax, and may be even more
easily transported, the potential for expansion of the
proposed system is immeasurable.

Because many devices are

designed strictly for the purpose of interpersonal
communication, they may feature only the voice or perhaps a
print option, and most probably cannot be interfaced with
other devices.

This leaves the user either without the

ability to access other kinds of equipment or with several
different devices to meet various needs.

With the personal

computer as the core device, the student also has all the
computer capabilities available as well.

For example, the

student could use the word processing, database,or
spreadsheet features, or run any number of software packages
on the system, opening up many more possibilities for
independent and creative communications.
The Votrax Personal Speech Synthesizer, aside from its
portability, was selected because of its relative
intelligibility.

Mirenda and Beukelman (1987) compared the

DECtalk, Votrax and Echo II speech synthesizers in terms of
intelligibility of the output,

They found the DECtalk to be

the most intelligible of the three.

This is primarily due

to the method used to produce the synthesized speech, which
is a combination of the techniques used in digitized speech

25
synthesis and those used in the "text to speech" method.
Unfortunately, however, the DECtalk system that was
available for possible use for this study was not portable.
Furthermore, the expense involved with such a piece of
equipment was too great to be considered an affordable
option for the student, should the system prove effective.
The Echo II, which was also available and which utilizes
the same

text to speech" method as the Votrax, was found to

produce less intelligible synthesized speech comparatively.
While the Votrax Personal Speech Synthesizer does not have
the same quality of voice output as the DECtalk, it is of a
better quality than the Echo II and has the advantage of
portability.
The keyboard enhancement software package used in the
study was the E Z Keys/E Z Talker package from Words+, Inc.
and was made available through funding provided by the
Council for Faculty Development and Research at EUP.
This software combines a number of powerful features
together into a single program,

It is capable of single

finger operation, abbreviation expansion/ and autospacing,
and it features the SideTalk speech output and keyboard
control.

In addition, since EZK/EZT is a memory resident

program, other programs can be run at the same time and can
utilize all of the features of the EZK/EZT system without
having to call up other files or functions.

26

PROCEDURES

The investigation was conducted over a three day period
during which the subject interacted on a one-to-one basis
with each of the twenty-four selected communication partners
under two different conditions,

Under each condition, she

was to transmit ten messages, which varied in the nature of
their content.

The time required for message transmission

was recorded for each of the ten utterances (Appendix A) .
The first interaction with each of the participants was
under the condition of naturally produced verbal
communication.

For this interactive period, the subject was

instructed to communicate ten sentences to each of her
communication partners using only the natural expressive
modes available to her:

primarily speech and gesture.

She

was not directed to use any particular strategies, but was
asked only to communicate as she normally would, using
whatever methods she would typically employ.
The second interaction was conducted in a similar
fashion.

The subject was instructed to limit her expressive

output to that which she was able to produce using the voice
synthesizer.

This particular procedure, however, was not

followed in the strictest sense, since the subject provided
reinforcement to her listeners via gestures to either
confirm or deny the accuracy of their responses.

The

27
production portion of this interactive exchange entailed
entering the sentence into the computer via the keyboard and
sending the message to the voice synthesizer for output by
pressing the ENTER key,

Again, the subject was not

instructed to use any particular strategies, but was
encouraged to optimize the functions of the keyboard
enhancement software package as she best knew how.

Appendix

C illustrates the basic display format of the E Z Keys/E Z
Talker SideTalk feature, including samples of word
prediction tables as they would appear on the screen as
characters were entered into the system,

In addition,

examples of the abbreviation expansion capabilities of the
package, as applied to some of the stimulus sentences used
in the study, are presented in Appendix B.
Transmission time, as referred to within the context of
this experiment, refers to the period of time between the
initiation of an utterance and the correct repetition of the
utterance by the listener,

Participants were instructed to

verbally express what they understood the message to be at
the conclusion of the subject’s presentation of the
sentence.

A stop watch was started at the initiation of the

utterance, or in other words, at the moment the subject
began to speak, gesture, or make her first keystroke on the
computer.

The timing of transmission completion was not

quite so easily determined,

For the purpose of this study,

successful transmission was judged to have occurred at the

28
moment the subject demonstrated acceptance of the listener's
response. This standard was followed throughout the
procedures even though the subject was somewhat inconsistent
with regard to her expectations for her communication
partners.
Throughout the experiment, counterbalancing techniques
were utilized to control for practice effects, which might
otherwise

prejudice the data toward an illegitimate

conclusion.

All participants first interacted with the

subject under the condition of natural speech followed by
the second trial under the condition of augmentation.

The

members of the three participant groups were seen on an
alternating schedule.

In other words, the subject

interacted first with a member of the familiar group,
followed by a member of the generally familiar group, and
finally with an unfamiliar listener,

Due to time

constraints and scheduling difficulties, this order could
not be strictly adhered to, but was varied sufficiently to
eliminate any bias in favor of one group versus another.
Stimulus sentence lists were also varied from one
participant to the next (Appendix B).

Participants who

interacted with the subject using Stimulus List A under the
natural speech condition were exposed to Stimulus List B
during the second trial using the computer and voice
synthesizer.

This was considered necessary in light of the

fact that participants were seen within a period of three

29
consecutive days and would very likely remember the
utterances used in the first interaction due to the limited
time interval between trials and the novelty of the
experience,

Stimulus sentence lists were also consistent in

terms of the number of syllables.
Due to the fact that one of the primary goals of the
investigation was to obtain as representative a sample of
the subject’s functional expressive communication skills as
possible, the stimulus sentences were given to the subject
on 3 x 5 index cards that she could look at and turn back
over during message transmission.

The reason for this was

that it was deemed necessary to take into account any
spelling or grammatical errors that might interfere with
effective transmission were she to use this system on a
regular basis.

30
RESULTS
A between-between-within three way analysis of variance
w&s psrformed, on the data, obtained.

Data consisted

specifically of the mean times for transmission of
behavioral and affective utterances under both the
nonaugmented and augmented conditions recorded for each of
the participants in the three groups.

Statistical analysis

of the data, as illustrated in Table 1, indicated that the
difference in transmission time between the two
communication modes examined was significant at the .01
level.

However, transmission time was not reduced when

using the augmentative communication system as had
originally been expected, but was, in fact, significantly
increased.

Degree of familiarity of the communication

partner and message content were not found to be significant
factors in the mean time required for message transmission.
Examination of the raw data revealed that transmission
times for individual utterances for the natural
communication mode ranged from approximately one second to
two minutes and thirty-two seconds, with most times
occurring within twenty seconds,

Under the augmented

condition, times ranged from ten seconds to ten minutes and
eight seconds,

The majority of the messages were

successfully communicated to the listeners via the speech

TABLE!
Summary Tablej__Between-Between-Within ANOVA

Source

df

Between-Subjects

47

44045.82

Familiarity

2

3084.31

1542.16

1.6858

Content

1

686.94

686.94

0.7509

Familarity X Content

2

1853.85

926.93

1.0133

Error:
Between-Subjects

42

38420.72

914.77

Within-Subjects

48

44826.48

Communication Mode

1

22730.42

22730.42

Familiarity X
Communication Mode

2

424.89

212.45

0.4483

Content X
Communication Mode

1

659.40

659.40

1.3914

Familiarity X
Content X
Communication Mode

2

1107.34

553.67

1.1683

Error:
Within-Subjects

42

19904.43

473.92

Total

95

88872.30

SS

MS

F

47.9631**

** -

p< .01

32
synthesizer in less than two minutes with most falling below
one minute. However r among the participants there was one
individual who had exceptional difficulty with the task and
consequently presented times in excess of seven and a half
minutes for three of the utterances.

(This same individual

also demonstrated the highest transmission time under the
condition of natural speech as well.)

Two additional

participants presented excessively high times on isolated
utterances, one over eight minutes and the other just over
ten.

Due to the extreme variability in transmission times

among the various participants, the error factor is
exceptionally high.

This must be noted in consideration of

an appropriate interpretation of these statistical findings.

33
DISCUSSION

In spite of the surface appearance, the results of this
study are far from conclusive.

Augmentative communication

strategies cannot be dismissed as a viable solution to the
communication problems of the functionally disabled
nonverbal population or even for this particular student on
the basis of the current findings,

Any number of extraneous

and confounding variables must be taken into consideration
when interpreting these results.

With regard to the

experimental design, it can be assumed with a fair degree of
certainty that the results would have been significantly
different had the procedures been a little different,

For

example, if the subject had been permitted to use only her
verbal skills rather than allowing gestural techniques as
well, the time required for message transmission would have
been dramatically increased, since the meanings of her
utterances were primarily established through the use of
"supplemental” gestures.

By the same token, if transmission

time under the condition of augmentation had been based on a
starting time at the initiation of the voice output rather
than manual input into the system, the total time required
for understanding would have been significantly reduced,
Review of videotaped portions of the study, demonstrating
the use of the augmentation, revealed that with the

34
exception of those few individuals who had unusual
difficulty with some isolated utterances, the majority of
the transmission time was accounted for by the time required
for the subject to enter the information into the system via
the computer keyboard.

Unfortunately, since the timing

procedures were not conducted in this manner, there is no
statistical evidence available from the current
investigation to substantiate this theory,

This is merely

an indication that further studies utilizing alternative
data collection and analysis procedures need to be
implemented to obtain a different perspective on the
relative effectiveness/efficiency of the communicative
exchange using a variety of communication techniques.
Aside from the procedural aspects of the investigation,
there were many extraneous variables that had not been
anticipated, but that did enter into play with regard to
transmission times.

Some of the patterns observed to be

characteristic of the subject's interaction contributed
significantly to the delay of information exchange.
The subject had a tendency, under both conditions of
communication mode, to present messages in a word-by-word
fashion, which removed the advantage provided by contextual
cues.

When using the speech synthesizer, she did present

the complete utterance initially, but quickly reverted to
this word-by-word method if the message was not immediately
understood.

For example, as she communicated the message "I

35
could fall asleep so easily," she almost always waited until
the word "fall" had been properly understood before
attempting the word "asleep".

It would seem that a listener

might be more apt to understand the concept of falling
asleep more easily than the two words presented in
isolation.
Among the strategies she did employ to transmit the
information under the condition of natural communication
were the following:

She used an abundance of gestural cues,

facial expressions, and frequent "writing" on the table
surface with her finger.
quite limited.

Her verbal output, however, was

Verbal strategies she used included multiple

repetitions, spelling out words orally, and reciting the
alphabet to the point of the target letter.

Oddly enough,

she never once went outside of the actual utterance to
clarify the meaning to the listener,

She demonstrated the

same kinds of behaviors when using the augmentative
communication system as well.
Another rather weak skill on the part of the subject was
her inability to pick up on the nonverbal cues from her
listeners when they were struggling to understand her.

She

frequently would adopt a particular strategy to convey the
message and was very slow to alter the means, even if it was
very apparent that her communication partners were still
experiencing difficulties,

For example, one of the

participants was quite plainly a poor finger speller, but

36
the subject persisted with this strategy throughout her
interaction with him, When using the voice synthesizer, she
P®*-formed multiple repetitions of the same utterance without
altering the communication.

For many reasons, it is

necessary to address her practical use of the equipment as a
separate issue.
The potential for improvement of transmission times
through strategy training and skill development is
tremendous.

It must be noted that the subject had had only

limited exposure to the augmentative devices prior to the
investigation.

While the features of the software package

had been demonstrated to herf she was unable to use them
optimally during the experiment due to lack of experience.
Transparency is a significant factor in communicative
effectiveness that must also be taken into account.

Since

the subject has always relied on her verbal and gestural
abilities to communicate, trials under the natural speech
condition required very little concentration,

When using

the computer, however, the subject was required to deal not
only about the message content, but also the novelty of the
new output mode and the sometimes perplexing and frequently
frustrating responses of her communication partners. Having
never been exposed to synthetic voice productions, several
of the participants had a very difficult time trying to
decode the presented message,

By the same token, the

subject, having never attempted interaction via an

37
augmentative device, was unprepared to respond to their
confusion.

In addition, since the subject was aware of what

the output device was 11 saying,” she did not seem to listen
very closely to the actual output.

Had she been paying

closer attention, she might have recognized the source of
some of the difficulties of her listeners.

For example, the

word "having" as produced by the speech synthesizer sounded
more like "has-ing", since the /v/ sounded like a /z/.
Although listeners should have been able to determine the
target word based on the grammatical context, it would have
been helpful to attempt other phonetically oriented
spellings.

These are but a few examples of the subject's

unsuccessful attempts to communicate her intended message,
but they attest to the fact that the subject's limited
experience with the devices had tremendous impact on her
effective use of the system.
Vanderheiden and Yoder (1986) address this same issue of
transparency from a somewhat different perspective.

They

identify skills as "those abilities that the individual
acquires through practice and continual use", whereas
strategies refer to "specific techniques that are discovered
by or taught to the individual."

Using these terms, the

effective operation of an aid and/or technique is primarily
a function of the individual's skills.

However, the

effectiveness in using augmentative components to actually
communicate is probably more a function of the strategies

38
that the individual has discovered or been taught (either
directly or through modeling).

In any case, the obvious

point is that the user of an augmentative device requires
extensive training and practice to effectively utilize an
augmentative system.
In all fairness to the subject, it must be noted that
the participants also demonstrated poor strategies for the
facilitation of effective communication.

Just as the

subject had failed to make optimal use of contextual cues,
so did the listeners.

This was, at times, attributable to

the excessively slow pace of the communicative process, as
in cases when so much time had elapsed between the
initiation of the first word and the understanding of the
final word that the listener had forgotten the beginning
portion of the sentence.

This was also, however, frequently

a matter of poor use of the context provided,

For example,

one individual had understood the message "open the window"
to the point of "open the w-i-n-", at which point she
attempted to guess the rest of the sentence, saying "open
the freezer, open the winter",

Another of the participants

had an exceptionally difficult time with spelled messages.
Having had the word "finals" spelled out to him several
times, his responses included "flights","food", and a number
of other incorrect, unrelated answers.
Since this study considered, in addition to
communicative mode, utterance content and degree of

FIGURE 1; MEAN TRANSMISSION TIMES
NATURAL COMMUNICATION

30 25 TIME
20 (SEC)

Behavioral

15 -

Affective

10 5 0
Personal

General

Unfamiliar

DEGREE OF FAMILIARITY

FIGURE 2: MEAN TRANSMISSION TIMES
AUGMENTED COMMUNICATION

80 ✓✓

70 ✓

60 ✓✓

TIME
50 (SEC)
40 -

Behavioral

•*

Affective

30 20 10 0
Personal

General

DEGREE OF FAMILIARITY

Unfamiliar

40
familiarity as related to communicative effectiveness, a
comment must be made about these two variables.

While the

statistical analysis failed to identify a significant
difference between the levels of either of these variables,
notable differences in transmission methods were observed
among the various familiarity groups and did influence
transmission times under the natural speech condition, as
illustrated in Figure 1.

Interaction with those

individuals whom the subject had identified as friends
typically consisted of gestures in adaptive sign language
and finger spelling, with minimal verbalizations.

The

communicative exchange involving members of the other two
familiarity groups consistently included fewer specific
gestural cues and an abundance of more recognizable
"writing" and spelling techniques.

Under the condition of

augmentation (Figure 2), transmission times generally fell
between forty and fifty seconds for all groups with the
exception of the unfamiliar group with affective utterances,
which was influenced by several extremely high transmission
times that distorted the mean.

Utterance content did not

significantly influence transmission time,

Although such a

relationship might be assumed, based on the graphic
illustration of results in Figures 1 and 2, the differences
seen can be attributed to the extreme variability among
participants and their transmission times.

41
SUMMARY, RECOMMENDATIONS, & CONCLUSIONS

The findings of this investigation do not provide
statistical support for the hypothesis that the use of an
augmentative communication device increases communicative
effectiveness/efficiency by reducing the amount of time
required for message transmission as determined by the
procedures used here.

However, careful attention must be

directed toward the numerous extraneous and confounding
variables that have been identified in the discussion of the
results.
These findings further suggest that there is no
significant difference in message transmission times based
on the degree of familiarity of the communication partner or
the nature of the utterance.

Although degree of familiarity

did not demonstrate significance in terms of transmission
times, distinct differences in interactional patterns of the
subject were observed and noted among the three groups.
Message content did not influence transmission time
significantly.
Interpretation of these results must be approached with
caution.

Extreme variability in mean transmission times,

taking into account the influence of a few extraordinarily
high scores, further demands a more careful look at the

42
details of the study when considering the implications of
the results reported here.
It is quite apparent that the procedures employed in
this investigation, combined with the inexperience of the
subject and participants, lent themselves to a certain
inevitable, however incorrect, result,

Consequently, any

conclusions drawn from this study are pure speculation at
best.

It would be advisable to conduct a similar

investigation using different procedures controlling for
many of the extraneous variables that influenced the outcome
of this study.

In addition, any number of other

augmentative communication systems could be considered.
With regard to any augmentative system based intervention
program for the subject considered here, further assessment
is still required.

Strategy training and more extensive

exposure to the proposed communication system (for both the
subject and her communication partners) would be highly
recommended.

Augmentative communication strategies should

not be dismissed as a possibility until an indepth
evaluation of the subject’s ability to use a system
effectively, given the appropriate training and practice,
has been completed.
One additional concern that extends far beyond the scope
of this study and the individuals involved herein is the
appropriate selection of attainable goals for the disabled
student.

The efficacy of the disabled student’s academic

43
and vocational goals is an issue that must be addressed at
the beginning of the student's academic program, but there
is no one within the university structure who has been given
the responsibility for this very necessary task,

Such an

issue is deserving of immediate attention.
In conclusion, it must be stated once again that the
goal of augmentative and alternative communication
strategies is not to replace an already functional
communication system but rather to expand commmunicative
abilities beyond those that currently exist and to develop
an effective overall communication system.

44
REFERENCES

American Speech-Language-Hearing Association (1981, August).
Position Statement on Nonspeech Communication.
ASHA, 23 (8), 577-581.
Beukelman, David. "When You Have a Hammer, Everything Looks
Like a Nail." AAC Augmentative and Alternative
Communication. June 1987. 94-96.
Blackstone, Sarah W. (Ed.). (1986). Augmentative
Communication: An Introduction. Rockville, MD: American
Speech-Language-Hearing Association.
Lafontaine, Linda M. and Frank DeRuyter. "The Nonspeaking
Cerebral Palsied: A Clinical and Demographic
Database Report." AAC Augmentative and Alternative
Communication. December 1987. 153-162.
Mirenda, Pat and David R. Beukelman. "A Comparison of Speech
Synthesis Intelligibility with Listeners from Three
Age Groups." AAC Augmentative and Alternative
Communication. September 1987. 120-128.
Newell, Alan F. "How Can We Develop Better Communication
Aids?" AAC Augmentative and Alternative
Communication. March 1987. 36-40.

APPENDIX A
NAME:

GROUP:

TOIAL I - VERBAL COMMUNICATION

TRIAL II - PERSONAL COMPUTER
AND VOICE SYNTHESIZER

Date
STIMULUS LIST:

Date
A

B

NATURE OF THE UTTERANCE:

BEHAVIORAL
TIME:

12 3

STIMULUS LIST:

2

3

BEHAVIORAL
TIME:

AFFECTIVE

1

2

4

3

4

5

6

5

6

7

8

7

8

9

10

9

10

TOTAL
TIME:

TOTAL
TIME:

AVERAGE
TIME:

AVERAGE
TIME:

(Behavioral) (Affective)
AVERAGE TIME FOR ALL ITEMS:

B

NATURE OF THE UTTERANCE:

AFFECTIVE

1

A

(Behavioral) (Affective)
AVERAGE TIME EOR ALL ITEMS:

I, THE UNDERSIGNED, HEREBY GRANT PERMISSION TO THE STAFF OF LEADER CLINIC AND
OFFICE OF DISABLED STUDENT SERVICES OF EDINBORO UNIVERSITY AND TO MARY BETH
WAY, THE PRINCIPLE RESEARCHER INVOLVED IN THIS PROJECT TO UTILIZE VIDEOTAPES
OF MY INTERACTION WITH THE SUBJECT OF THE SIUDY FOR EDUCATIONAL PURPOSES AND
RESEARCH AUGMENTATION.

APPENDIX B
STIMULUS SENTENCES
LIST A

LIST B

1.

Will you get my jacket.

1.

Close the window.

2.

That really bothers me.

2.

I need more sleep.

3.

Could you help me with this.

3.

Turn off the lights when you
leave.

4.

I could fall asleep so easily.

4.

I don't like having my picture
taken.

5.

Open the door.

5.

Put the bocks on my desk.

6.

My stomach is growling.

6.

I have a headache.

7.

Please put the book in my book bag.

7.

Would you bring me more milk.

8.

It's too hot in my room.

8.

I want to go home.

9.

Can I borrow a pencil?

9.

Would you please bring me
my folder.

10.

I'm worried about finals.

10.

I'm tired of this.

EXAMPLES OF ALTERNATE METHODS OF PRODUCING TARGET WORDS
A2.

Type:

tt (t3) rly (r3) bothers me (m2)

A4.

Type:

i cd fa3 asleep so (si, s*) easily

B2.

Type:

i nd mr sll

B3.

Type

tm (tul) 06 t2 19+s w7 (whn) u (yl, y*) le3

B7.

Type

wd u (yl, y*) brl me (m2) mr milk

B8.

Type

iwn to (tl, 2) go (g2, g*) h6

APPENDIX C
E Z KEYS/ E Z TALKER:

SIDETALK SCREEN FORMAT

1
2
3
4
5
6
.7
8
9
0

SideTalk - Press END to quit

t

to
the
that
this
take
time
think
there
they
tell

th

s

the
that
this
think
there
they
that's
them
than
thing
1

so
see
some
she
still

r

sure
something
say
should
same
like
long
let's
look
little
left
last
lunch
light
love

P

room
n
right
really
remember
red

not
need
no
now
night

radio
rather
run
real
read

new
nice
never
next
name

put
please
play
pleased
people
pants
pee
pack
pop
pretty