...‘

-.

w .. l; '_

é-‘$

.17.».

jﬁ

yd
'—

..,

’m.”

a;
4.
.

,. ea;

"7

1
.

4. ’3‘ "

gift-.331
A '4}

 

.‘S
I”???
‘ \

 

 

 

'-::-L-.r .

r ‘N. ..
”EPA"

l
!h"'..'4_’~ w
....r -

.v *
.Q

'1
‘ "3;”: ,1 «.mw
r ,mei'im 3:3" "'
amazes-:3
v £13.33:
- a»... ’

A .
3"
a"???

f.
:53

ft“

 

 

 

955.; ‘

 

~ is
5'; '3 £5734
LIBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

The Effect of Contingent Auditory Stimulation of Self-initiated
Movement on Three Quality of Life Measures in Young
Children with Severe Multiple Disabilities

presented by

Cathy H. McQuillan

has been accepted towards fulﬁllment
of the requirements for the

Ph.D. degree in Special Education

 

 

Major Professor’s Signature

£10.64,ch /é,L 900 3

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DatoDuopes-sz

THE EFFECT OF CONTINGENT AUDITORY STIlVﬂJLATION OF
SELF-INITIATED MOVEMENT ON THREE QUALITY OF LIFE
MEASURES IN YOUNG CHILDREN WITH SEVERE MULTIPLE DISABILITIES
By

Cathy H. McQuillan

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Counseling, Educational Psychology, and Special Education

2003

ABSTRACT
THE EFFECT OF CONTINGENT AUDITORY STIMULATION OF
SELF-INITATED MOVEMENT ON THREE QUALITY OF LIFE MEASURES IN
YOUNG CHILDREN WITH SEVERE MULTIPLE DISABILTIES
By
Cathy H. McQuillan

This study investigated the effect of providing musical stimulation, both
contingent to movement and noncontingent to movement, on 3 quality of life factors in 2
children, aged 4 and 6 years, who have severe multiple disabilities (SMD). The
contingent musical stimulation was provided by a movement sensor system called
Soundbeam. The Soundbeam system consists of a beam emitter, control box and
electronic keyboard. Movements that interrupt emitter beam produce sound on the
keyboard.

The assessments used to measure the effect of this intervention were: (a) quality
of behavior state as measured by the amount of time in states considered optimum for
learning; (b) amount of enjoyment cf the activity as measured by observation of indices
of happiness; and (c) amount of voluntary movement as measured by observation of these
movements.

The investigator used a single—subj ect multiple baseline design of A-BC-B-BC
across the 2 participants. Experimental sessions were conducted 3 times a week over a
period of 13 weeks in the participants’ homes. During the baseline phase (which
consisted of 3 sessions for the ﬁrst participant and 6 sessions for the second participant)

each participant had access to a preferred leisure activity while seated in her wheelchair.

During the ﬁrst intervention phase which consisted of 15 sessions, the Soundbeam
emitter was positioned near the participant in such a way that voluntary movements the
child was capable of making activated the instrument. During the third phase which
consisted of 6 sessions, the investigator removed the emitter and produced the musical
feedback with her movements, thus the participant heard the music but did not have
control over it. In the fourth phase, which consisted of 15 sessions for the ﬁrst participant
and 12 sessions for the second participant, control of the instrument was returned to the
child.

Data for the 3 quality of life factors were collected using observational protocols.
The data were graphed for visual analysis, effect sizes of the difference in means of the 4
phases were calculated, and ﬁeld observations of the investigator were reviewed.
Analysis of the data indicated that the introduction of the intervention had different
effects on the 2 participants. The ﬁrst participant exhibited a decrease in observed indices
of happiness after the ﬁrst introduction of Soundbeam and no changes in the other 2
measures across all 4 phases. The second participant exhibited a decrease in preferred
state intervals, an increase in indices of happiness, and an increase in voluntary
movement after the ﬁrst introduction of Soundbeam.

It was concluded that this investigator did not ﬁnd signiﬁcant evidence that
Soundbeam was effective in improving the quality of life factors of these participants.
Discussion focused on the usefulness of multiple quality of life assessments in detecting
behavioral changes in children with SMD, possible effects on family members when the
child is in a contingent situation, and the role of behavior state in the lives of children

with severe multiple disabilities.

DEDICATION

This dissertation is dedicated to the memory of four very special children;
Karen, Jonathan, Jacob, and Kirstie Ann. You live on in the hearts of all who loved you.
It was a privilege to have been your teacher.

iv

ACKNOWLEDGEMENTS

I am grateful to so many who have helped me with both my studies and the
dissertation process. I especially appreciate the cooperation of the families of the two
beautiful little girls, “Esther” and “Lucy,” who took part in this study. Their parents
welcomed me into their homes week after week and I thank them for participating.

My committee members could not have been more helpful and supportive. Dr.
David Stewart provided guidance and invaluable advice throughout my entire course of
study. Dr. Cynthia Okolo’s insightful comments always furthered my thinking and her
calm encouragement was greatly appreciated. Dr. Tim Stocks provided invaluable
advice on research methods during many hours of meetings. I am especially indebted to
my committee chair, Dr. Gail Dummer, for her years of assistance on this project. I
beneﬁted so much from her expertise in writing and research, endless patience, and
attention to detail. I will never be able thank her enough for her support and
encouragement.

I would like to thank Dr. Sally Roberts at the University of Kansas who provided
my training on the behavior state coding protocol developed by her research team. Her
advice and encouragement were greatly appreciated during my visit to Kansas.

Three staff members at the College of Education have been truly supportive of my
teaching and research activities during my time at Michigan State. I would like to thank
J aNice Benjamin, Karen Glickman, and Stephanie Opatrny-Smith for cheerfully helping
me take care of the details involved in negotiating the degree process. I’m sure I tried

their patience on more than one occasion.

I would like to express my appreciation to the administrators at the Jackson
County Intermediate School District for granting me a sabbatical leave that enabled me to
fulﬁll my residency requirement and complete early coursework. In addition, I would
like to thank Paula Alexander, Lou Alonso, Rick Banghart, Dr. Susan Bruce, Patti
Clarke, Dr. Patrick Dickson, Dr. Robert Graves, Dr. Tom Hwang, Sue Klassen, Sarah
Reid, Dr. Richard Rendall, Anne Rogers, Mikki Trudell, and the late Dr. Gerald Kratz for
their support or assistance during various phases of my program.

Frances Kidder ably served as my reliability co-rater, statistics consultant, and all
around support system. Frances is an angel on earth and I will always be grateful for her
help. I would also like to express my gratitude to Gena Cochran who produced the ﬁgure
drawings.

My family has been a source of comfort and support during this entire process and
I dearly love them all. My parents, Stephen and Ann Howell, have always encouraged me
and my siblings to pursue our dreams no matter what obstacles we encounter. My
husband, Richard, never failed to support my work through the roller coaster of emotions
that accompanies such a project. My sons, Stephen, Timothy, and Brian have grown into
ﬁne young men while Mom pursued two degrees at Michigan State. I am so proud of

them; they are truly the greatest accomplishment of my life.

vi

TABLE OF CONTENTS

Page
List of Tables .................................................................................................................... xi
List of Figures ................................................................................................................... xii
CHAPTER 1: Introduction ................................................................................................. 1
Overview of the Problem ........................................................................................ 1
Signiﬁcance of the Problem .................................................................................... 2
Purpose of the Study ............................................................................................... 5
Need for the Study .................................................................................................. 5
Research Questions ................................................................................................. 6
Operational Deﬁnitions ........................................................................................... 7
Severe Multiple Disabilities ........................................................................ 7
Behavior State ............................................................................................. 7
Preferred State ............................................................................................. 7
Behavior State Organization ....................................................................... 7
Happiness Indices ....................................................................................... 8
Voluntary Movement .................................................................................. 8
Overview of Methods ............................................................................................. 8
Limitations ............................................................................................................ 10
CHAPTER 2: Review of the Literature ............................................................................ 12
Characteristics of Children with Severe Multiple Disabilities ............................. 12
Child Development Theories ................................................................................ 14
Behaviorism .............................................................................................. 14
Constructivism .......................................................................................... 16
Systems Theory ......................................................................................... 20
History of Pedagogical Approaches ...................................................................... 24
Developmental Orientation ....................................................................... 25
Functional Orientation .............................................................................. 26
Quality of Life Orientation ....................................................................... 27
Quality of Life Indicators ...................................................................................... 29
Behavior State ........................................................................................... 29
Behavior State in Infants ............................................................... 3O
Behavior State in Individuals with SMD ...................................... 35
Happiness Indices ..................................................................................... 48
Movement ................................................................................................. 5 1
Contingent Stimulation as an Intervention ............................................... 52
Role in Development ..................................................................... 52
Effect on Behavior State ................................................................ 53
Effect on Movement ...................................................................... 54
Effect on Happiness Indices ........................................................... 57
Synthesis ............................................................................................................... 58

vii

CHAPTER 3: Methods ..................................................................................................... 61

Research Design .................................................................................................... 61
Single Subject Design ............................................................................... 61
Variables ................................................................................................... 63

Independent .................................................................................... 63
Dependent ...................................................................................... 64
Controlled ...................................................................................... 64

Design Rationale ................................................................................................... 65
Effect Size ................................................................................................. 67
Signiﬁcance ............................................................................................... 68
Power Analysis and Phase Length ............................................................ 68

Baseline (A) Phase ......................................................................... 69

First Intervention (BC) Phase ........................................................ 69

Removal of the Contingency (B) Phase ......................................... 70

Final Intervention (BC) Phase ....................................................... 71

Control of Threats to Validity ............................................................................... 71
History and Maturation ........................................................................... 71
Instrumentation Error .............................................................................. 72
Statistical Regression .............................................................................. 73
Treatment Integrity ................................................................................. 74

Order Effects ........................................................................................... 74

Participants ............................................................................................................ 75
Selection Criteria ...................................................................................... 75
Recruitment and Informed Consent .......................................................... 75
Description ................................................................................................ 76

Esther ............................................................................................ 76
Lucy .............................................................................................. 79

Instrumentation ..................................................................................................... 80

Behavior State ........................................................................................... 80
Description ................................................................................... 80
Validity ........................................................................................ 8O
Reliability ..................................................................................... 80
Measures Obtained ....................................................................... 8O

Happiness Indices ..................................................................................... 81
Description ................................................................................... 81
Validity ........................................................................................ 81
Reliability ..................................................................................... 82
Measures Obtained ....................................................................... 82

Voluntary Movement ................................................................................ 83
Description ................................................................................... 83
Validity ........................................................................................ 83
Reliability ..................................................................................... 83
Measures Obtained ....................................................................... 83

viii

Intervention ........................................................................................................... 83

Data Collection ..................................................................................................... 84
Data Collection Schedule .......................................................................... 84
Research Environment .............................................................................. 86
Data Collection Procedures ....................................................................... 86
Key Personnel and Qualiﬁcations ............................................................. 92

Data Analyses ....................................................................................................... 93
Research Question 1 ................................................................................. 93
Research Question 2 ................................................................................. 94
Research Question 3 ................................................................................. 94

CHAPTER 4: Results, Discussion, and Conclusions ....................................................... 96

Behavior State Proﬁles of Participants ................................................................. 96
Reliability of Observations ....................................................................... 97
Results for Esther ..................................................................................... 102
Results for Lucy ....................................................................................... 104
Discussion ................................................................................................ 106

Time in Preferred States ....................................................................................... 106
Reliability of Observations ...................................................................... 107
Results for Esther ..................................................................................... 107
Results for Lucy ....................................................................................... 111
Discussion ................................................................................................ 1 14

Happiness Indices ................................................................................................ 116
Reliability of Observations ...................................................................... 116
Results for Esther ................................................................................... 119
Results for Lucy ....................................................................................... 119
Discussion ................................................................................................ 124

Voluntary Movement ........................................................................................... 126
Reliability of Observations ...................................................................... 128
Results for Esther ..................................................................................... 128
Results for Lucy ....................................................................................... 132
Discussion ................................................................................................ 13 5

Conclusions .......................................................................................................... 137

CHAPTER 5: Implications, Limitations, Recommendations, and Summary ................. 140

Implications .......................................................................................................... 141
Behavior State ......................................................................................... 141
Happiness Indices .................................................................................... 141
Voluntary Movement ............................................................................... 142

Limitations ........................................................................................................... 143
Research Design ....................................................................................... 143
Instrumentation ........................................................................................ 145

Recommendations for Further Research .............................................................. 146

Summary .............................................................................................................. 148

ix

APPENDICES ................................................................................................................. l 50

Appendix A: Informed Consent ........................................................................... 151
UCRIHS Approval ................................................................................... 152
Consent Letter .......................................................................................... 153

Appendix B: Coding Protocols ............................................................................ 158
Behavior State .......................................................................................... 159
Happiness Indices .................................................................................... 161
Voluntary Movement ............................................................................... 162

Appendix C: Individual Protocols of Behavior State .......................................... 164
Esther ....................................................................................................... 165
Lucy ......................................................................................................... 166

Appendix D: Sample Session Log Sheets ............................................................ 167
Esther ....................................................................................................... 168
Lucy ......................................................................................................... 169

REFERENCES ................................................................................................................ 171

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Table 8:

Table 9:

Table 10:

Table 1 1:

Table 12:

Table 13:

Table 14:

Table 15:

Table 16:

LIST OF TABLES

Page
Behavior State Codes Used in Research
with Infants ............................................................................................... 31
Behavior State Codes Used in Research
with Individuals with Severe Multiple Disabilities .................................. 36
Current Behavior State Code .................................................................... 43
Descriptive Proﬁles of State Organization in Individuals with
Severe Multiple Disabilities ...................................................................... 46
Research Timeline .................................................................................... 62
Characteristics of Participants ................................................................... 77
Interobserver Agreement Percents for Individual Behavior States ........... 98
Number of Preferred State Intervals for Esther and Lucy ....................... 109
Descriptive Statistics for Esther—Preferred State ..................................... 111
Descriptive Statistics for Lucy-Preferred State ........................................ 113
Number of Happiness Intervals for Esther and Lucy ............................... 118
Descriptive Statistics for Esther-Happiness Indices ................................ 121
Descriptive Statistics for Lucy-Happiness Indices .................................. 123
Number of Movement Intervals for Esther and Lucy .............................. 129
Descriptive Statistics for Esther-Movement ............................................ 131
Descriptive Statistics for Lucy-Movement .............................................. 134

xi

LIST OF FIGURES

Page
Figure 1: Components of Soundbeam Apparatus .................................................... 85
Figure 2: Research Environment for Esther ............................................................ 87
Figure 3: Research Environment for Lucy .............................................................. 88
Figure 4: Typical Position of Esther with Soundbeam Emitter ............................. 90
Figure 5: Typical Position of Lucy with Soundbeam Emitter ................................. 91
Figure 6: Percentages of State for Esther in Baseline Phase ................................ 103
Figure 7: Percentages of State for Lucy in Baseline Phase ................................... 105
Figure 8: Mean Percentage of Preferred State Intervals for Both Participants ..... 108
Figure 9: Percentage of Preferred State Intervals for Esther in all Sessions ......... 110
Figure 10: Percentage of Preferred State Intervals for Lucy in all Sessions ........... 112
Figure 11: Mean Percentage of Happiness Intervals for Both Participants ............. 117
Figure 12: Percentage of Happiness Intervals for Esther in all Sessions ................ 120
Figure 13: Percentage of Happiness Intervals for Lucy in all Sessions .................. 122
Figure 14: Mean Percentage of Movement Intervals for Both Participants ............ 127
Figure 15: Percentage of Movement Intervals for Esther in all Sessions ................ 130
Figure 16: Percentage of Movement Intervals for Lucy in all Sessions .................. 133

xii

CHAPTER 1
Introduction
Overview of the Problem
The provision of educational services to children with multiple disabilities in the

United States has been mandated for over three decades. Many of these students ﬁrst
gained access to public education as a result of a Federal law, Public Law 94-142 passed
in the mid 1970’s (Siegel-Causey, Guy, & Guess, 1995). Public Law 94-142, titled the
Education for all Handicapped Children Act, mandated universal access to education for
all children, regardless of disability. Children receive support services and
accommodations according to their needs, which are categorized by disability.

Children who qualify for services in the category of multiple disability are a
heterogeneous group with a wide range of ability levels and support needs. Children
identiﬁed as having multiple disability currently comprise less than 2% of the total
population of students receiving special education services (United States Department of
Education, 1995).

One subset of students with multiple disabilities is that of children who have
severe multiple disabilities (hereafter SMD). Children who have SMD have been
described as those children whose cognitive abilities are measured at less than twelve
month levels, have various degrees of sensory loss, and are typically non-ambulatory and
non-verbal (Arthur, Hook, & Butterﬁeld, 1995). The education of children with SMD
presents a great challenge to educators and caretakers as few effective interventions have
been identiﬁed, it is difﬁcult to appropriately assess these students and there is

disagreement about appropriate educational goals for this population.

Signiﬁcance of the Problem

The intense needs of SMD students place great demands on the ﬁnancial,
physical, and emotional resources of caretakers and school systems. Providing
educational services to this population of students requires support from a wide range of
professionals (Dule, Komer, Williams, & Carter, 1999). Classroom teachers report
dissatisfaction with training and lack of time to adequately serve this population of
children (Izen & Brown, 1991). Large blocks of classroom time must be devoted to
maintaining the health and safety of the children (Campbell, 1987; McCubbin, 1987) in
addition to providing educational interventions.

Descriptions provided by researchers help convey the pervasive effects that SMD

have on the individual’s functioning. Morse (1992) describes a child with SlVﬂ):

Wally, aged 5, has been diagnosed with cortical blindness and a severe seizure

disorder. He also is physically impaired and lives most of his day in a wheel chair

. . . Wally generally gazes into space . . . he will brieﬂy ﬁxate on an appealing

stimulus at a distance of 8”-12” or near eye level. His left hand usually is

clenched and positioned on his tray near his chest. His right hand sometimes

engages in small random side-to-side swiping motions. Wally alerts to sudden
sounds by “startling,” during which his entire body extends and his eyes widen.

(p. 74)

Children with SMD are totally dependent on others to meet all of their daily living
needs. Parents have written poignant accounts of the almost overwhelming demands the
child with SMD places on the family (Featherstone, 1980; Kupfer, 1982; Morton, 1978).
For example, Featherstone (1980) expressed her reaction to an additional 15-minute care
routine suggested by a school nurse:

This new demand appalled me. I rehearsed angry, self-justifying speeches in my

head. Jody, I thought, is blind, cerebral-palsied, and retarded. We do his physical

therapy daily and work with him on sounds and communication. We feed him

each meal on our laps, bottle him, change him, bathe him, dry him, put him in a

body cast to sleep, launder his bed linens daily, and go through a variety of
routines designed to minimize his miseries and enhance his joys and development

2

. . . where is that ﬁfteen minutes going to come from? What am I supposed to give
up? . . .because there is no time in my life that isn’t spoken for, and for every
ﬁfteen-minute activity that is added, one has to be taken away. (p. 77-78)

Debate continues as to the most suitable and efﬁcacious ways to provide an
appropriate education for children with SMD. There is emerging consensus that curricula
used for children with mild and moderate disabilities are not appropriate for children with
more severe disabilities (Ayres, Meyer, Erevelles, & Park-Lee, 1994; Ferguson, 1985;
Sailor, Gee, Goetz, & Graham, 1988; Westlake & Kaiser, 1991). Children with SMD will
probably never attain the vocational or self- care skills that are the desired outcomes of
curricula for children with mild or single disabilities. An incomplete understanding of
the unique characteristics and development of this population may also be an obstacle to
the provision of interventions that would optimize their development.

Current research strands investigating characteristic ﬂuctuating levels of alertness
in children with SMD, new viewpoints on the appropriate outcomes of curricula, and the
application of systems theory to child development may lead to improved interventions
for this population. Direct care providers of children with SMD have noted that their
students often do not exhibit the prolonged periods of attentive and engaged behavior
necessary for observing and learning from their surroundings. These poor levels of
alertness may serve as obstacles to learning. Researchers have characterized these
ﬂuctuating levels of behavior as similar to ﬂuctuating states of awareness exhibited by
infants. This comparison has led to the body of research on the role these various levels
of awareness, called behavior states, plays in the lives of children with SMD.

In addition to the potential promise of new insights into the behavioral

characteristics of individuals with SMD, the reconceptualization of quality of life as a

more appropriate outcome goal, rather than vocational or self-care skill goals has driven

recent research on curricula. Deﬁnitions of quality of life, outcome measures, and
interventions designed to increase these goals have been investigated. Measures to assess
quality of life goals are being developed and investigated.

Theoretical frameworks have evolved in developing interventions for this
population. Guess (2000) identiﬁed the application of system theory to frame research
involving children with SMD as one of the most promising developments in the ﬁtture of
interventions.

Research that is framed within these themes may contribute to the discovery of
effective interventions that address some of the dissatisfactions reported by caregivers
and direct care providers. The dissatisfactions of lack of time to implement interventions
and lack of impact that interventions seem to have in the lives of children with SMD
continue to be expressed. As currently conceptualized, an effective intervention is one
that directly impacts the quality of life of the child and can be delivered without
disproportionate cost in terms of time by direct care providers. In addition, framing the
evaluation of possible interventions in terms of systems development adds to the
knowledge of development of children with SMD.

The potential of technology to improve the quality of life for individuals with
severe multiple disabilities has also received attention by researchers (Ellis, 1997; Guess,
2000; Gutowski, 1996; Kinsley & Langone, 1995). The use of technology to gain access
and control to the environment has been a continuing line of research that began several
decades ago when interventionists and others realized the possibility of these tools

(Behrmann & Lahm, 1982; Goldenberg, 1979).

Purpose of the Study

The purpose of this study was to investigate the effect of contingent auditory
stimulation provided by movement sensor technology on three measures of quality of life
in young children with SMD. The contingent stimulation was provided by a movement
sensor linked to a MIDI sound synthesizer. The movement sensor provided an immediate
auditory response to any movement initiated by the child. The effect of this immediate
auditory response on behavior state, happiness indices and amount of voluntary
movement was investigated.

Need for the Study

This study addressed several areas of investigation pertinent to improving
interventions for children with SMD. By framing investigation within a systems
framework the study addressed the need to: (a) identify possible factors that affect
behavior state organization in young children with SMD; (b) empirically evaluate a
device that has been marketed as an appropriate intervention for this population; and (c)
evaluate an intervention that does not require extensive training and time requirements by
direct care providers.

Information about the effects of use of movement sensor technology on quality of
life factors could be used to beneﬁt the population of over 112,000 students who receive
services in the category of multiple disability in the United States (National Information
Center for Children and Youth with Disabilities, 2002) and their direct care providers.

Behavior state is an important area of research for students with SMD. The
construct of behavioral state and its impact on the quality of lives of children with SMD
has been identiﬁed as an important area of research (Arthur et al., 1995; Guess, 2000;

Reid, Phillips, & Green, 1991; Richards & Stemberg, 1994). Studies such as this one that

5

attempt to identify important factors are needed to ﬁn‘ther the understanding of behavior
state.

Advances in computer technology have offered new opportunities to the
population of SMD children but also challenge the interventionist to select appropriate
and effective applications from an expanding list of possible systems. The commitment
of ﬁnancial resources to new technologies can be substantial. Empirical evaluations of
the use of new technologies with this population, such as this study, are essential to
justify the purchases and implementation of new technologies (Fields, 1999; Mount &
Cavet, 1995). Especially important is the investigation of technologies that may offer
young children the means to exert control over their environment as the experience of this
control may be foundational to later learning (Brown & Cohen, 1996; Guess, 2000).

The costs of an intervention in terms of time, as well as ﬁnances committed to
equipment should be considered, as lack of time has been identiﬁed by caregivers and
direct care providers as an obstacle to the implementation of interventions.

Research Questions

Three research questions will be investigated in this study.

1. Is there a difference in the behavioral state organization of a young child with
severe multiple disabilities across three conditions: (a) a typical play environment; (b) the
same play environment with the addition of auditory stimulation contingent to the child’s
movements; and (c) the same play environment with auditory stimulation provided
noncontingently?

2. Within intervals where the child is observed in an alert or interactive behavioral
state, is there a difference in the percentage of intervals where a happiness index is

displayed across the three experimental conditions?

6

3. Within intervals where the child is observed in an alert or interactive behavioral
state, is there a difference in percentage of intervals in which the participant displays a
voluntary movement?

Operational Deﬁnitions

Severe Multiple Disabilities

The functional deﬁnition of severe multiple disability used in this study is
individuals who: (a) have severe motor limitations; (b) have sensory loss or functional
sensory impairments; (c) are non-verbal; and ((1) receive interventions at a basic sensory
input level (Guess et al., 1990).
Behavior State

Behavior state refers to a construct of a level of awareness and ability to interact
with the social and physical environment ﬁrst identiﬁed in infants. The behavior state
scale used in this study includes eight distinct states; sleep, drowse, awake/inactive/alert,
interact, stereotypy, crying/agitated and aggression/self injury (S. Roberts, personal
communication, April 3, 2002).
Preferred State

Two of the behavior states described above, awake/inactive/alert and interact, are
designated as preferred states. Preferred states are behavior states that are presumed to be
the most conducive to learning.
Behavior State Organization

Behavior state organization refers to how state manifests itself in the lives of
individuals. The organization of state refers to the relative percentages of time that the
individual spends in each state, how often the individual changes state, and how clearly

the states are deﬁned in the individual. Five attributes of state have been reported; quality

7

(time in preferred state), stability, sequence patterns, cyclicity, and periodicity (Roberts,
1992). The attribute examined in this study was state quality or percentage of time in the
preferred states.
Happiness Indices

Indices of happiness have been proposed as a way to measure the construct of
personal happiness (a quality of life indicator) in individuals with severe multiple
disability. A happiness index is any facial expression or vocalization typically considered
to be an indicator of happiness among people without disabilities. Examples include
smiling, laughing, and yelling while smiling (Green & Reid, 1996).
Voluntary Movement

Voluntary movement is deﬁned in this study as movement of a body part that the
participant’s caretakers report that the participant is able to control.

Overview of Methods

The research questions in this study were investigated using a single subject
multiple baseline design with 2 participants. The participants were children with SMD,
aged 4 and 6 years, who live in southern Michigan. A total of 39 20-minute sessions were
conducted in the participants’ homes over a 13-week period. All sessions were
videotaped. The contingent auditory stimulation provided in the intervention phase was
generated by a movement sensor that translated any of the child’s movements into
musical phrases produced by an electronic keyboard.

In the baseline (A) phase of the experiment, the child was seated in her
wheelchair with access to a typically provided play or leisure activity. In the ﬁrst
intervention phase (BC), the sensor emitter was placed near the child and was activated

by the child’s movements, providing contingent auditory feedback. The emitter was

8

aimed at the area of the body over which the child had voluntary control. In the next
phase (B) the sensor emitter was placed in view of the child but was activated by the
investigator, providing auditory feedback that was not contingent on the child’s
movements. In the ﬁnal (BC) phase control of the auditory feedback was returned to the
child by placing the emitter near the child.

For Research Question 1 on the effect of the intervention on quality of behavior
state, an observational protocol was used to record the number of observed 5-Second
intervals in each of the eight state categories. The researcher coded all of the sessions as
they progressed. The percentage of intervals that the participant was observed in a
preferred state was compared across the phases.

For Research Questions 2 and 3, videotaped intervals when the participant was
identiﬁed as being in one of the preferred states were transferred from the master video
tape to a second videotape for analysis. The researcher viewed and coded these
videotapes in random order at the conclusion of the experiment. The data for Research
Question 2 on the effect of auditory contingent response on enjoyment of the activity
consisted of percentages of preferred state intervals during which the participant
exhibited a happiness index.

Data for Research Question 3 on the effect of auditory contingent response on
amount of voluntary movement consisted of the percentages of preferred state intervals
during which the participant exhibited a voluntary movement

Data points for all three questions were graphed for visual analysis. A descriptive
statistic of effect size, d, which is calculated by dividing the difference in means of
adjacent phases by their pooled standard deviation, was used as an adjunct to visual

analysis.

Limitations

All single subject designs have limited external generalizability. Statistical
generalization from a sample to a target population is not possible because of the non-
random selection of the participants and the non-independence of the scores. Decisions
about generalizability must be made rationally, rather than from statistical inference and
these decisions depend on the internal validity of the experiment.

Internal validity is more likely to occur when experimental conditions are
controlled enough so that the change of experimental conditions rather than other
simultaneously changing factors can be posited as the cause of any change in the
dependent variables being measured. Unfortunately, there were many factors such as
antecedent conditions and the participants’ health status that were not closely controlled
in this experiment.

One source of variability has to do with the characteristics of SMD. Children
with SMD have a very high incidence of illness and health problems. Although both
participants completed all of the sessions, there was no control for daily health condition
of the child which is a potential source of variability.

A second major source of possible variability resulted from difﬁculty in
standardizing the conditions under which the experimental sessions were conducted in
the participants’ homes. The sessions were conducted with varying levels of background
noise, interactions with siblings and parents and with slightly different placement of the
child’s wheel chair relative to the experiment. These conditions could not be completely
controlled in the natural settings of the home environment. Antecedent conditions such

as changes in schedules due to school vacations, snow days, doctor’s visits and medical

10

treatments are also part of the fabric of the lives of these children and could not be
completely standardized.

In addition to variability issues, the fact that only two children participated in the
study limited the opportunity for replications of any effect in this study. More participants
would have strengthened the internal validity of any conclusions.

Internal validity is also related to the construct validity of the measures used to
evaluate the outcomes. The three measures used in this study, behavior state, happiness
indices, and amount of movement are emerging assessments. Further study and
evaluation of the construct validity of these assessments may be necessary as they are

used by different researchers in a variety of settings.

11

CHAPTER 2
Review of the Literature

This chapter will provide a review of: (a) the characteristics of students with
severe multiple disabilities (SMD); (b) the theoretical perspectives that have been used to
inform interventions targeted to this population; (c) the curricula that have been
implemented for this population over the past few decades; and (d) the provision of
contingent stimulation as an intervention strategy.

Characteristics of Children with Severe Multiple Disabilities

With the passage of universal access to education laws in the mid-1970’s,
children in the United States who met the criteria delineated in one of 12 categories of
disability became entitled to special educational services. This legislation (PL 94-142)
ensured the right of education to virtually all children no matter how severe their
disabilities.

The category of multiple disability was created to include children who had more
than one disability. The federal government deﬁned children in this category as “those
students who have concomitant impairments, other than deaf-blindness, such as mental
retardation-blindness, mental retardation-orthopedic impairment, etc., the combination of
which causes such severe educational problems that they cannot be accommodated in
special education programs solely for one of the impairments” (U .S. Department of
Education, 1995).

This deﬁnition acknowledged the fact that the effects of more than one disability
are multiplicative; that is, the effects of one disability interact with the effects of the other
disability so that the multiple disability has a different and unique effect on development

than any of the disabilities in isolation (Brimer, 1990). Children who meet the criteria for

12

this category exhibit a wide range of abilities and educational needs. A continuing
frustration expressed by educators and stakeholders was the lack of functional deﬁnitions
for various subgroups of these students, particularly those who could be categorized as
having the most severe and profound disabilities. This particular subgroup of students has
provided special challenges to direct service providers over the past few decades
(Ferguson, 1985; Izen & Brown, 1991; Rainforth, 1982; Sailor et al., 1988; Stemberg,
1988).

There is no universal consensus on the educational deﬁnition for the subgroup of
students with SMD (Ferguson, 1985; Siegel-Causey, Guy, & Guess, 1995). Several
functional deﬁnitions have been proposed (Dewson & Whitely, 1987; McDonnell,
Hardman, McDonnell, & Kiefer-O’Donnell, 1995; Orelove & Sobsey, 1987; Reid et al.,
1991). Proposed deﬁning characteristics of students with SMD have included: (a)
cognitive abilities (when testable) with a mental age of less than one year; (b) sensory
impairments; (c) severe physical impairments with no ambulation; and (d) dependence on
others for all physical needs. Individuals with SMD usually are non-verbal and often
display only minimal responsiveness to visual, auditory and olfactory stimuli.

Historically, children and individuals in this subgroup of disability have made up
a signiﬁcant proportion of the populations of institutions where they received custodial
care. Changes in social mores in the 19605 and 19703 contributed to the community
normalization movements as well as efforts to de-institutionalize service to this and other
previously disenfranchised populations. Debate on whether students with SMD should
have access to educational services (Kauffman & Krouse, 1981) was resolved with court
action (Baer, 1981) that reafﬁrmed the intent of legislation was to provide universal

educational access regardless of the severity of the disability.

13

In recent years an increasing number of children with SMD receive educational
services in their home communities and schools. However, despite advocacy for
inclusion in regular education classrooms (Downing, 1996; Hanline & Fox, 1993;

J orgensen & Calculator, 1994), the majority of children with SMD continue to receive
education in segregated classrooms and facilities (Wisniewski & Alper, 1994).

Individuals with SMD need lifetime care and support in all areas of life. These
outcomes are quite different than the typically developing child or the child with mild to
moderate disabilities and thus have presented educators with the difﬁcult task of selecting
appropriate educational curricula. Curriculum, by deﬁnition, includes both instructional
methods and goals (Noonan & McCormick, 1993). Curricula generally address two
components of instruction, “how to teach” and “what to teach.” Two theoretical
perspectives on child development; behaviorism and constructivism, have inﬂuenced the
development and implementation of instructional methods, the component of “how to
teach” for this population.

Child Development Theories
Behaviorism

Early attempts to change the behavioral repertoires of individuals with SMD were
based on operant conditioning paradigms. Researchers demonstrated that individuals
with SMD could increase the frequency of certain behaviors in response to a contingent
stimulus (Haskett & Hollar, 1978). Bijou (1981) credited the use of behavioral
techniques with improving teachers’ attitudes about the learning potential of children
with a range of disabilities.

Increasing observable positive behaviors or extinguishing undesirable ones was

the focus of much of the behavioral research on persons with SMD through most of the

14

1970’s (Guess, 2000; Reid et al., 1991). Research in behavioral frameworks included
investigations on the relative efﬁcacy of different reinforcement schedules, evaluation of
strategies to teach generalization of skills, and identiﬁcation of effective reinforcers.

In a comprehensive overview of behavioral research that was conducted between
1949 and 1982, Reid et al. (1991) reviewed 39 studies that had the goal of teaching
adaptive skills or improving postural control to individuals with SMD. They deﬁned
adaptive skills as behaviors that allow an individual to take some degree of care of
himself or herself, express his or her wants or needs, participate to some extent in
remunerative work, function similarly to persons without disabilities in regard to use of
leisure time, or participate in community experiences.

They reviewed 13 studies conducted between 1949 and 1979 that demonstrated
discrete behaviors such as pressing a lever, raising an arm, vocalizing, or pulling a ring
could be increased when reinforcers such as edibles, vibration, or music were applied
contingent to exhibition of the target behavior by the individual with SMD. Six studies
conducted between 1975 and 1982 focused on improving the participant’s postural
control (holding the head up) by reinforcing this target behavior with music, cartoons, or
verbal praise.

Twenty studies conducted in the late 1970s and 19808 targeted a variety of
behaviors such as eating skills, social and communication behaviors, ﬁne motor skills,
and manipulation of leisure materials. Some behavior change was reported in all of the
studies, although not with all of the participants. Reid et al. (1991) reached two
conclusions in their review. They concluded the reviewed studies had not demonstrated
that persons with SMD could acquire any independent adaptive skills, as the studies did

not report the generalization of the targeted skills to different situations. The authors also

15

noted that the one-to-one client ratio used to conduct the training in the studies might be
too time consuming for implementation in actual practice.

Behaviorist interventions continue to be used with individuals with SMD
(Lancioni, O’Reilly, Van Dijk, & Klaase, 1998; Silliman-French, French, Sherrill, &
Gench; 1998) to increase or decrease speciﬁc behaviors. Instructional textbooks for pre-
service teachers (Alberto & Sharpton, 1988) provide information on task analysis,
reinforcement schedules, and reinforcer selection. Although behaviorist interventions
have been used effectively, not all students appear to beneﬁt from this approach. If the
targeted behavior does not change the individual’s ability to communicate or functionally
do anything, it can be argued that increasing that behavior does not appreciably improve
the individual’s life. It may also be difﬁcult to use a behaviorist intervention when an
appropriate reinforcer for the individual with SMD cannot be identiﬁed (Ivancic &
Bailey, 1996).

Constructivism

Constructivist perspectives are based on Piagetian and neo-Piagetian concepts that
learning and cognition are the result of children’s interactions with the physical and
social environment. These perspectives assume that children are intrinsically well
equipped for actively developing cognitive and social understandings through physical
interaction with their surroundings.

Current best practices in early childhood education of typically developing
children are heavily inﬂuenced by these constructivist, action based theories of learning
and development (Bredekarnp, 1993; Fox, Hanline, Vail, & Galant, 1994; Mallory,
1992). In constructivist frameworks, a signiﬁcant role of the early childhood teacher is to

present typically developing children with frequent opportunities to interact with objects

16

and people (Fox et al., 1994). Teachers are encouraged to provide children with a choice
of stimulating toys and materials that are accessible for child directed play. Instructional
targets are often based on child interests and initiations.

This view of the child as an active learner is seldom used to explain the behaviors
of children with SMD, primarily because of their limited abilities to engage in
interactions with the environment (Carta, Schwartz, Atwater, & McConnell, 1991). This
limited ability to engage in interactions results in a more directive role for the direct care
provider in prompting and mediating interactions with the environment. Rather than
basing instructional targets on the SMD child’s interests, the interventionist often selects
the instructional goal.

However, some interventions for children with SMD within this framework have
been investigated. One approach is to make features of the physical environment more
accessible to the child. The importance of a suitable and accessible physical environment
has been emphasized both in educational considerations for typically developing children
(Mallory, 1992) and children with disabilities (Hanline & Fox, 1993; Schneekloth, 1989).
Modiﬁcations to the physical environment for children with SMD have included
stabilizing objects and playthings very close to the children so that even small motor
movements will allow the children to engage in continued contact with these objects.

Specialized environments such as a “playpen” (Landesman-Dwyer & Sackett,
1979), a “Little Room” (Nielsen, 1991, 1992), and an “activity box” (Dunnett, 1999)
have been developed and investigated by researchers guided by this framework.

In the Landesman-Dwyer and Sackett (1979) study, the investigators were
interested in increasing the variety and complexity of behaviors exhibited by their

participants, aged 2-20 years who were residents in a state school and totally dependent

17

on others for care. The researchers constructed two square “playpens” that had open tops
and eight interchangeable panels for sides. The playpens contained stimulating objects
such as strings of multi-colored balls, ticking metronomes, pillows ﬁlled with scented
materials, mirrors and squeeze toys. Eight of the participants received an experimental
treatment that consisted of sessions in the playpen. The participants in the experimental
condition were positioned in the playpens for 30 minutes daily in an upright or elevated
position, either by themselves or with a peer over a period of four to ﬁve months.

Landesman-Dwyer and Sackett (1979) investigated the effect of the experimental
treatment on three dependent variables: (a) organization of sleep and active states; (b)
behaviors such as shifts in body position, touching objects, vocalizations, looking
behaviors, mouthing, facial expressions, and head movements observed in the
participants; and (c) test scores on the Bayley Scales of Infant Development.
Results were that the participants in the experimental condition increased their vertical
head movements, looking and touching behaviors and time where they demonstrated
activity increased. Bayley scores did not improve.

Nielsen (1991 , 1992) also constructed a specialized play area she called the
“Little Room” for participants in her investigation. The Little Room was constructed of
pipes that held three side panels constructed of various materials and Plexiglas top and
was placed over the participant who was lying on his or her back. Various objects were
attached to the top with elastics and were arranged to hang near the face and hands of the
participants who were placed in the Little Room on their backs. The goal of her
investigation was to see if exposure to this play area facilitated the development of the
concept of object permanence. Nine of her 20 congenitally blind participants (aged 8-16

months) had additional disabilities of mental retardation, and of those 9, 3 participants

18

had cerebral palsy. The participants were placed in the Little Room or a control
environment (consisting of just the frame of the room, without the panels) for eight 20-
minute sessions for four to ﬁve days.

Nielsen (1991) found that the total number of contacts with the toys and objects
made by many of the participants increased, as did the complexity of their movements.
She suggested that the receipt of the same feedback over and over again (as the objects
returned to their positions after being released) helped the participants learn to locate
speciﬁc objects. She also speculated that the absence of sounds interfering from the
outside environment while the child was enclosed by the sides and ceiling of the Little
Room facilitated the participant learning to connect his or her activity directly with the
sound it produced. Many direct care providers have constructed similar play areas that
have walls to cut off outside sounds and provide a variety of tactually interesting objects
near the child (Dunnet, 1999).

The constructivist emphasis on the child’s physical interaction with the
environment as the foundation of cognitive development is based on observations of
typically developing children. This template of development may not be applicable to the
development of individuals who have disabilities that prevent these physical interactions.
Some individuals with severe physical disabilities who produce little or no voluntary
movement since birth have been able to demonstrate normal cognitive and
communicative abilities when provided with augmentative communication (Beminger,
1988; Eagle, 1985). Thus, physical interaction with the environment may be only one of
several possible routes to cognitive competence.

Interventions based on constructivist approach have the advantage of not

requiring the time intensive one-to-one teacher-student ratios that are often needed to

19

implement behavioral training routines. An attractive feature for direct care providers is
that such an environment allows children with SMD to “experience ‘things”’ without
adult help or intervention (Dunnet, 1999).

Systems Theory

Systems (also called chaos, non-linear and dynamical system) theory is a
conceptual framework that has come to dominate the thinking in the biological and
physical sciences. Systems theory posits that all natural entities are composed of multiple
and diverse parts or sub-systems that combine in various ways to produce coherent and
patterned behavior. All entities are conceptualized as self-organizing themselves in
various ways and according to various timetables. The emerging patterns of organization
or development are not following a pre-determined and innate hard-wired plan, but rather
emerge in response to the context or subsystem timetables that exist within the entity.

Systems theory has been applied to such diverse phenomena as the development
of tornadoes, mineral crystal development, the evolution of social organizations and
development in human beings. The pattern of development has been described both
through mathematical modeling and the use of analogies. Systems theory has been
applied to the study of motor development in young children (Thelen, 2001),
development of communication skills (F ogel & Thelen, 1987), and cognitive
development (Lewis, 2000; Thelen & Smith, 1994; van Geert, 1991).

Some major tenets of systems theory are: (a) joint determination by multiple
causes; (b) context sensitivity and contingency— the signiﬁcance of any one cause is
contingent upon the state of the rest of the system; (c) development as construction; and
(d) distributed control- no one type of interaction controls development (Oyama, Grifﬁths

& Gray (2001).

20

One tenet of systems theory when applied to human development is the
assumption that cognitive, social/emotional, and physical development is the result of
various diverse parts of the system interacting and organizing in response to both internal
factors and environmental or external factors. According to systems theory, a biological
organism seeks a stable pattern of organization that is optimal for that place and time.
These preferred states of stability are called attractor states in systems vocabulary (Ennis,
1992; Fogel & Thelen, 1987). Attractor states can be described as shallow (easily
perturbed or reorganized) or deep (typical states of functioning if the system is
undisturbed). As different subsystem components are scaled differently (have different
time tables) a change in one component can move the system toward instability. This
instability may move the system out of a shallow attractor state into a new level of
performance or stability.

The subsystem components that appear to change the pattern of the system are
called control parameters. Different subsystems can serve as control parameters at
different times as the system is re-organizing (Fogel & Thelen, 1987). Control
parameters can be within the system (endogenous) such as current chemical balances in
the blood, or external (exogenous) to the system such as the availability of nutrition or
social support. With the myriad of existing endogenous and exogenous parameters, it
often difﬁcult to determine exactly which component is contributing to the development
of new patterns (van Geert, 1991).

It is during these periods of relative instability that qualitative changes in the
individual are presumed to take place. Some factors may act as catalysts to shift the
entire system into new modes of organization. These times of instability are an essential

characteristic of development (van Geert, 1991).

21

The framework of system theories may be an attractive lens through which to
view the development of children with SMD. Two systems concepts, “equiﬁnality” and
the “cascade effect,” seem particularly relevant. The concept of equiﬁnality refers to the
idea that there are multiple paths to competence. In motor development, for example, all
intact babies gain the ability to walk, but not all gain this skill on the same timetable, by
experiencing the same environments or child rearing practices, or by going through the
same intermediate stages (Thelen & Smith, 1994).

The concept of cascade effect, sometimes called the “butterﬂy effect” when used
as an example in weather systems, is the idea that small changes in initial conditions can
produce larger changes at a later time. The effect of that initial difference may cascade
through subsequent reorganizations of the system. As the knowledge about what these
differences are increases, and as the ability to provide early interventions that affect these
differences increases, the result may be an improved outcome for some children with
SMD.

Another attractive component of systems theory is that it emphasizes that the
organism cannot be studied apart from its environment. Thelen’s studies of crawling and
walking as forms of locomotion show that when different external control parameters
exist, different types of locomotion may emerge. Thelen gives the example of a typically
developing infant that she observed who never crawled but rather, but moved from place
to place by scooting around her home’s highly polished and slippery ﬂoors (Thelen,
1990)

The “solutions” that beneﬁt the individual with SMD may be different from those
that beneﬁt typically developing individuals. For example, observable emerging

communication competence in typically developing children may be babbling as a

22

precursor to pronouncing appropriate words. However, in a child with SMD an
observable emerging competence in communication might be head control in order to
activate a switch for an augmentative communication device. Using the typical
development template to target instructional goals in either behavioral or constructivist
frameworks may be counterproductive to what is the best solution for an individual with
SMD (McClennen, 1991; Mulderji, 2000.)

Some solutions that the individual with SMD exhibits may also appear counter-
productive to what would be considered optimal development. For example, many
individuals with SMD develop stereotypical behaviors that appear to interfere with other
activities. However, these behaviors may fulﬁll a need for the individual that is
unavailable through other means. In another example, patterns of learned helplessness
are often described in these individuals (Basil, 1992; Finkelstein & Ramey, 1977;
Robinson, 1986). Learned helplessness is described as an inability to complete an action
even if it is possible because there is an expectation of helplessness (Abramson, Seligrnan
& Teasdale, 1978) and is thought to result from early failures in attempting to engage
with the environment. Learned helplessness is often described as a secondary handicap
that prevents further attempts at engagement with the environment (Robinson, 1986), but
it could also be considered reasonable adaptation by the individual who indeed, will not
have much control.

As White (1980) observed when discussing goal setting for children with SMD,
form should not be confused with function. If the ultimate function is locomotion for
example, inappropriate concentration on walking (one form of locomotion) may be to the

detriment of ultimately attaining the function.

23

The usefulness of systems theory as a research framework for possible
interventions is still being explored. Oyama et al. (2001) write that developmental
systems theory is not a theory “in the speciﬁc model that produces predictions to be
tested against rival theories.” (p. 1-2) Rather, it can be considered a general theoretical
perspective on development.

Systems theory has been applied to studies of the characteristics and possible
interventions for students with SMD (Guess, 2000; Guess & Sailor, 1993; Guess &
Siegel-Causey, 1995). Studies conducted in this framework would ideally consist of
longitudinal studies of individuals with enough measurements to describe periods of
instability leading to growth and relative importance of possible control parameters
during these periods of instability leading to higher levels of organization (van Geert,
1991). Single subject studies in which various parameters are manipulated have also
been used to further knowledge within a systems framework. The challenge to
researchers of interventions, then, is both to identify possible important control
parameters during individual courses of development and manipulate them to determine
if more optimal courses of development result.

History of Pedagogical Approaches

There is continuing debate about suitable goals for SMD students because many
educational interventions have not appeared to make a signiﬁcant difference in their lives
(Snell, 1987). Students with SMD usually cannot be expected to acquire normalized types
of behavior such as work skills or independent self-care skills that are the focus of
curricula for students with no or mild/moderate disabilities. The question of the second
component of curricula, “what to teach” has been revisited many times over the past few

decades.

24

Developmental Orientation

Advocates of the developmental approach assumed that instruction should focus
on helping students attain typical developmental milestones (Cohen, Gross, & Haring,
1976). The goal of educational intervention from a developmental perspective is to
remediate a child’s underlying cognitive and physical deﬁcits. Instructional goals are
developed from comparing the student’s current level of performance to typical infant
and early childhood milestones, and then teaching to the next skill generally attained by
the typically developing child. Instructional materials are selected according to the
developmental or mental age of the children rather than their chronological ages.

From a remedial standpoint it “makes sense” to follow a normal developmental
template to select skills. Many self-care and functional skills are based on clear physical
abilities or prerequisite skills. For example, the ability to chew and swallow solid food
requires that the individual master prerequisite skills of tolerating the utensil placed in the
mouth, demonstrating an adequate gag reﬂex to protect against choking, and the ability to
make chewing movements.

However, individuals with SMD demonstrate so many deﬁcits in all areas of life
that remediation of all of them is not possible. The direct care provider may be
overwhelmed by the sheer number of needs. Children with SMD may not be able to
master one skill in the developmental sequence and could be trained endlessly on a single
skill with no measurable beneﬁt or mastery ever taking place. Another criticism of the
developmental approach for students with SMD is the tendency to provide these students
with objects that are not age appropriate. That is, some objects such as baby rattles might
correspond to the child’s developmental age but appear inappropriate for the child’s

chronological age.

25

Functional Orientation

Advocates of a functional orientation suggest that targeted skills should only be
those that are needed for the child to function as an adult as independently as possible
(Switzky, Rotatori, Miller, & Freagon, 1979). Brown, Neitupski and Harnre—Nietsupksi
(1976) suggested a “criterion of ultimate ﬁrnctioning” rather than typical developmental
progressions as the appropriate educational orientation. The interventionist using a
functional orientation selects skills that will be the most useful to the student in the
present or future environments (Holvoet, Guess, Mulligan, & Brown, 1980). The teacher
ﬁrst conducts a task analysis of functional skills (Browder, 1991) that will be needed in
future environments. Then, they directly teach each skill or skill component (Orelove &
Sobsey, 1987; Sailor & Guess, 1983; Wolery, Ault, & Doyle, 1992) and ﬁnally, link the
components together into completion of the entire task.

When students are unable to acquire some of the skill components, teachers are
still encouraged to teach the attainable components. Baumgart et al. (1982) called this
procedure “partial participation.” Partial participation was based on the premise that
completing some skill components would help include the student in the activities of
daily living. For example, a SMD child may not be able to remove her jacket, but her
partial participation goal might be to raise her arms with a verbal cue to allow her
caretaker to remove the coat (Ferguson & Baumgart, 1991).

As the main goal of functional orientation is to increase independence it may
have little beneﬁt for some SMD individuals who cannot functionally complete any tasks
related to self-care or independence. Interventions may have increased some skills, but

not enough to make a difference in the individual’s daily life.

26

Quality of Life Orientation

When many functional goals may not be attainable, some experts suggest that the
goals of intervention programs should target goals that are directly related to the
individual’s quality of life such as maintenance of existing behavioral repertoires,
prevention of neuromuscular degeneration, and the provision of more enjoyment (Reid et
al., 1991). There is emerging consensus that these “quality of life” goals are both more
appropriate and attainable than developmental or functional goals for the SMD
population. However, goals related to quality of life may prove to be more difﬁcult to
quantify than goals related to developmental progression or functional skills, as quality of
life is a more subjective concept.

An individual’s experience of quality of life depends on two aspects. The ﬁrst
aspect is the enviromnent of the person and can be objectively measured. Questions such
as “Is there food for the person to eat?” or “Does the person live in a low crime area?”
can be answered objectively. The second aspect of quality of life is the individual’s own
subjective perception of his or her circumstances. Questions such as “Am I satisﬁed with
my relationships” or “Do I like my work” can only be answered by that individual and
may not be directly related to the objective measures. Most experts agree that both
objective descriptors and subjective evaluations should be included in an assessment of a
quality of life (Felce & Perry, 1995).

Researchers who focus on the quality of life of individuals with SMD assume that
factors important to typically developing individuals are also important in the lives of
individuals with SMD. Researchers have investigated ways to increase the student’s
tolerance of therapeutic and medical procedures (Green & Reid, 1999b) to increase

physical comfort and provide freedom from pain. Researchers have investigated ways to

27

teach the individual’s choice making skills (Newton, Ard, & Horner, 1993; Reid,
Everson, & Green, 1999) and ways to determine the SMD individual’s preferences
(Houghton, Bronicki, & Guess, 1987; Thompson & Martin, 1994). Researchers have also
investigated ways to train appropriate communication partners for students (Brown,
Gothelf, Guess, & Lehr, 1998; Downing, 1996; Kaiser & Goetz, 1993) in order to
provide more opportunities for socialization.

Several researchers have focused on the aspect of choice as a particularly
important aspect of quality of life (Brotherson, Cook, Cunconan-Lahr, & Wehmeyer,
1995). Everson and Reid (1997) noted that individuals with SMD often have very few
opportunities to make choices on a day to day basis. Often, the direct care provider
makes the choice for the individual with severe multiple disabilities (Bannerman,
Sheldon, Sherman, & Harchik, 1990; Bedini, 1993).

Two strategies that have been used to identify preferences have been observing
the individual’s reactions to the activity or presenting two or more stimuli at the same
time and prompting the participant to choose (Newton, Ard, & Homer, 1993). Everson
and Reid (1997) point out that it is possible to ascertain choices made by individuals with
SMD by observing the time that the individuals attends to certain activities or their
behavioral indices that reﬂect enjoyment such as smiling.

An example study that informed the realization that opportunities for choice was
limited was conducted by Houghton et a1. (1987). The researchers observed 37 students
aged 85-21 years in classroom settings. They observed whether staff responded to
choices indicated by the children through body movements or vocalizations. For
example if a child spit out a spoonful of one type of food, a staff response that

commented on the child’s dislike of the food and willingness to offer another would be

28

scored a response to choice. Likewise, if the child began crying while positioned in one
position, and the staff repositioned the child or offered a toy with the comment that the
child didn’t like the position would be scored a response. They recorded several hundred
student initiations and reported that the staff responded to less than 20% of these
initiations. The researchers suggested that this low of rate of response resulted from the
emphasis on behavioral teaching paradigms. The emphasis on stimulus control
techniques may have distracted the caretakers from observing and responding to student-
initiated behavior.

Quality of Life Indicators
Behavior State

Teachers and researchers have noted that some students with SMD often appear to
be non-engaged with their surroundings. At various times individuals with SMD appear
drowsy, sleep at inappropriate times or engage in stereotypical behaviors which appear to
interfere with attending to external input. Teachers often continue to implement their
educational routines without regard to whether the child is attending to the stimulus
presented.

Several researchers suggested that these varying levels of alertness observed in
individuals with SMD might be comparable to the construct of behavior state described
in the literature on typically developing infants (Rainforth, 1982; Simeonsson,
Huntington, & Parse, 1980). The construct of behavior state in the infant literature has
been well established. One of the basic tenets of infant state theory is that the child’s
ability to interact with and learn from her surroundings is optimal in certain alert states.
Thus, it was suggested that the possible poor state control exhibited by individuals with

SMD might interfere with their development and quality of life.

29

Behavior state in infants. The construct of behavior state in infants was ﬁrst
suggested by Wolff (1959). He noted that infants displayed periods of varying alertness
and receptivity to their surroundings that ranged from deep sleep to alert activity. From
his observations of typically developing infants, he identiﬁed the six behavior patterns
(later called states) that he posited encompassed the range of the infants’ receptivity to
their surroundings.

Wolf’s original taxonomy of states was expanded and reﬁned by several groups of
researchers according to their own preferences and scientiﬁc backgrounds (Brazelton,
1973; Brown, Leavitt, & Graham, 1976; Komer, 1972; Prectl, 1974; Thoman, 1990).
States can range from deep sleep to alert /active and various proposed taxonomies
describe varying numbers (typically around six) of discrete states between these two
extremes. Examples of behavior scales that have been used in infant research are
reproduced in Table 1.

There is general consensus on the general theoretical description of behavior state
as a set a non- learned set of behaviors, which reﬂects the young infant’s ability to react
to and interact with environment and social variables.

Korner (1972) describes behavior states as qualitatively different conditions with
distinct types of internal organization. The same external stimulus can provoke a different
response in the young infant depending on the infant’s current state.

Behavior state has been described as a distinct mode of neural activity and
reactivity reﬂecting the condition of the nervous system (Prechtl, 1974), a gating
mechanism (Thoman, Komer, & Kraemer, 1976) and the means through which the
organism controls levels of stimulation (Brazelton, 1978). No consensus on the exact

deﬁnitions and appropriate measurements for state has been reached. State has been

30

Table 1

Behavior State Codes Used in Research with Infants

 

Campos & Brackbill (1973)

 

Quiet sleep

Active sleep

Drowsiness

Quiet awake

Active awake

Infant’s body appears relaxed muscularly. Eyes are
usually closed.

Infant exhibits diffuse movements of relatively frequent
occurrence. Some twitches, grimaces, smiles, sucks,
occasional movements of the eyeballs are present.

Infant exhibits ﬂuttering of eyelids. Eyes, when visible
have a glassy appearance. Frequent relaxation followed
by sudden jerks.

Infant displays little gross motor activity. The eyes are open
and appear bright and shiny.

Infant displays considerable gross motor activity.
Vocalizations are of the cranky, fussy variety.

 

 

Crying awake The criteria is the same as active awake but accompanied
by crying. Tears may or may not be present.
Komer, Brown, Reade, F emback, & Thom (1988)
Quiet sleep Infant’s eyes are closed and respirations are regular.

Active sleep

Drowsiness

Infant’s eyelids are mostly closed. Respirations are
faster and more irregular than quiet sleep. Motor
activity varies from slow limb movements to
generalized body writhing. Grimaces and other

facial expressions are frequent. Rapid eye movements
are present.

Infant is relatively inactive. Eyes are open but
have a dull or glazed appearance.

31

Table I, continued

Alert inactivity

Waking activity

Crying

Unclassiﬁable

Infant’s face is relaxed, does not grimace, and the
eyes are open and have a bright and shiny
appearance. Body movements, if any, are slow and
not vigorous.

Infant engages in vigorous or diffuse motor activity,
frequently involving the whole body.

Infant exhibits crying vocalizations associate with
vigorous, diffuse motor activity. The eyes may be
open or closed.

Infant exhibits behaviors that do not clearly meet
the criteria of any of the other states.

 

Colombo, Moss, & Horowitz (1989)

 

Deep sleep-A

Deep sleep-B

Active sleep

Drowsiness

Alert inactivity

Waking activity

Infant exhibits no movement except mouthing or
startles. Deep, regular and slow abdominal
respirations.

Infant exhibits no movement except mouthing or
startles. Respirations are more irregular and faster
than Deep sleep -A but still deep and

abdominal.

Infant exhibits facial movement during sleep. Respirations
are costal, shallow, irregular and fast. Rapid
eye movements may be present.

Infant’s eyes may be open but appear dazed. Slow
facial and bodily movement may be present.

Infant’s eyes are open and bright. Infant is quiet and
inactive.

Infant displays generalized motor activity. Facial grimacing

or grunting with brief vocalization or cry outbursts is
present.

32

Table I, continued

 

Sahni, Schulze, Stefansi, Myers, & Fifer (1995)

 

Code 1 Infant’s eyes are closed. There is a predominately
ﬂaccid appearance. Body movements are limited to
startles.

Code 2 The infant makes small body movements. Motor activity

includes slow intermittent writhing movements,
frowns, smiles, chewing and sucking movements.

Code 3 Infant displays rapid eye movements. Body movements are
similar to Codes 1 or 2. Eyes may occasionally open or
close or remain brieﬂy half-open.

Code 4 Infant displays wakeful behavior.
Code 5 Infant is crying.
Code 6 Infant is feeding.

 

33

measured by physiological measures such as heart rate and respiration (Prechtl, 1974)
and by behavioral observations of characteristics such as activity level, amount of
vocalization and eye movement (Brazelton, 1973; Komer, 1972).

The continuing disagreement on taxonomy, exact deﬁnitions, and appropriate
measurement procedures may limit comparisons between various studies (Thoman, 1990)
although there is general consensus that infants’ interactions with their surroundings
appear to be dominated by behavior state until three months of age. Estimates of time in
alert states of infants younger than 3 months range from 11% to 21% (Columbo &
Horowitz, 1987), although more time in alert state is correlated with the amount of time
spent with the mother (Thoman, Davis, & Deneberg, 1987).

Alert state is a control variable in the majority of reports on perceptual and
cognitive abilities in newborns and young infants (Ashton, 1973). It is difﬁcult to elicit
consistent responses from infant participants unless state is controlled. Researchers
agree that the infant may be the most receptive to input from the environment while in
one of the alert states (Thoman, 1990), and that achieving this state may be necessary for
optimal learning to occur (Brazelton, 1978).

In addition to being a control variable, behavior state has been a dependent
variable in a large body of infant research (Becker & Thoman, 1983; Boismier, 1976).
Researchers have investigated both soothing strategies to facilitate sleep states and
activating strategies to facilitate alert states. The relative inﬂuences of internal and
external factors in the manifestation of state is a source of ongoing debate (Becker &
Thoman, 1983).

State organization has also been investigated as a way to differentiate between

typically developing and compromised infant populations. For example, Prechtl,

34

Theorell, and Blair (1973) found that the percentage distributions and durations of
various states in typically developing infants were different than atypical infants who had
Down syndrome, jaundice, or cerebral functional disturbance. These state organization
differences were described based on one six-hour polygraphic examination period
conducted on the infants between the third and eleventh days of life.

Behavior state in individuals with SMD. Several researchers (Landesman—Dwyer
& Sackett, 1978; Rainforth, 1982) questioned the possible inﬂuence of behavior state in
the lives of individuals with SMD. Special educators and other direct service providers
had noted that these students often exhibit levels of non-alertness or stereotypy that may
interfere with their ability to beneﬁt from educational interventions (Sailor et al., 198 8) or
that interfered with assessment (Richards & Stemberg, 1989). Rainforth (1982)
questioned whether these periods of non-alertness and stereotypy could be comparable to
the behavior states described in the infant literature.

The possible inﬂuence of behavior state in the SMD population has been
investigated by researchers over the past two decades. Researchers have attempted to: (a)
develop observational protocols to describe and quantify behavior state; (b) describe state
organization in individuals with SMD; and (c) observe and investigate of possible
relationships between state and internal/extemal characteristics of the individual and
his/her environment.

At least four codes describing behavior state have been developed and used by
researchers with populations of SMD individuals. These taxonomies are presented in
Table 2. The four research groups that developed these taxonomies relied on pilot

observations of observable behavioral indices to create their categories.

35

Table 2

Behavior State Codes Used in Research with Individuals with Severe Multiple

Disabilities

 

Landesman-Dwyer & Sackett (1979)

 

Sleep

Low level of activity

Moderate level of activity

High level of activity

Characterized by closed eyes, slower rate of breathing.
Lack of responsiveness to the environment.

No clear responses to the environment. Minimal peripheral
movements of the body. Eyes are open.

Represents the individual’s moderate range of activity.
Clear movements of some body parts.

Eyes are open and moving. Some vocalizations.
Clear response to, or interaction with the environment.

 

Simeonsson, Huntington, Short, & Ware (1988)

 

Deep sleep

Intermediate sleep

Active sleep

Drowsy

Quiet awake

Eyes closed, regular respiration, no movements.

Eyes closed; few minor facial, body and/or mouth
movements; respiration is periodic, alternating periods
of shallow and deep breathing.

Eyes closed, irregular respiration, some gross motor
activity (stirring, writhing, grimacing), mouth or other
facial movements.

Eyes open and closed intermittently, ﬂuttering eyelids,
eyes have glassy appearance, frequent relaxation followed

by sudden jerks.

Relatively inactive, eyes open and appear bright and shiny
respiration regular.

36

Table 2, continued

Active awake

Fussy awake

Mild agitation

Uncontrollable agitation

Eyes open, difﬁrse motor activity of limbs and whole body,
vocalizations of a content nature

Eyes open, vocalizations of a fussy, cranky variety.

Eyes open, diffuse motor activity, moderate crying, tears
may or may not be present

Screaming, eyes open or closed, tears may or may not be
present

 

Richards & Stemberg (1992, 1993)

 

Seizure
Asleep

No orienting

Agitated

Arousal/ unclear

Arousal/ orienting

A seizure occurs within the interval.
Eyes are closed. May be some movement and vocalization.

Eyes are open but no eye movement. Motor movements
are jerky or slight and non-purposeful.

Awake. Vocalizing in a manner indicating discomfort or
anger.

Eyes are open but not clearly focused on environment. May
be voluntary movement but no purposeful interaction with
any object or person.

Eyes opened and focused. May be gross or ﬁne motor
movements toward interaction with environment.
May be changes in facial expression and/or
vocalizations in response to environmental events.

 

Guess, Roberts et al. (1993), Guess, Roberts, Siegel-Causey, & Rues (1995)

 

Asleep-inactive

Eyes are closed. Respiration is slow and regular. Body tone
is relaxed.

37

Table 2, continued

Asleep-active

Drowsy

Daze

Awake inactive-alert

Awake active-alert

Awake active-stereotypy

Crying-agitated

Seizures

Eyes are closed. Respiration is uneven, may exhibit
sporadic movements. May exhibit rapid eye movements,
occasional facial expressions and or vocalizations.

Eyes are either open and eyelids appear “heavy,” or eyes
are opening and closing repeatedly. Vocalizations may
occur.

No orientation to visual, auditory or tactile stimuli
predominates. Few or no vocalizations.

Eyes are open. Some active visual or auditory
orientation. Non-orienting body movements may occur.
Vocalizations may occur.

Engages/interacts by making contact with a person or
objects. Vocalizations in direct response to verbalized
questions.

Exhibits stereotyped behaviors (idiosyncratic, repetitive
rhythmic movements of body or body parts).

May exhibit intense vocalizing, crying or screaming.
Overall increased tension in body tone with
accompanying agitation behaviors.

Exhibits seizure behavior. The deﬁnition of seizure
behavior is individual for each person and determined by
direct care providers.

 

38

Landesman-Dwyer and Sackett (1979) conducted an early investigation that used
behavior state as a dependent variable. They developed a four-category code of “levels of
activity” which were sleep, low, medium, and high. They used this scale to describe the
sleep- wake cycles of 16 participants for the duration of their study designed to test an
intervention. The participants, aged 2-20 years old, were described as nonambulatory,
profoundly mentally retarded, and unresponsive to external stimulation. Landesman-
Dwyer and Sackett (1979) reported 94% agreement between two coders using percent
agreement formulas.

The authors used the scale to compare the percentages of time that their two
groups of participants spent in various states after 8 of the participants were exposed to
an intervention designed to facilitate a higher level of alertness. From observations made
over a 24-hour period they found that participants spent an average of 12.4 hours in the
sleep state, 9.0 hours in low level activity, 3.1 hours in moderate level activity, and 2.0
hours in high level activity.

Richards and Stemberg (1992) developed a six-category scale of states from pilot
observations of 7 participants they described as profoundly retarded, unable to perform
self care skills independently, non-verbal, and nonambulatory. The six states they
described were asleep, no orienting, agitated, awake/unclear, awake/orienting and
seizure. Their protocol required 15-second observations for ﬁve-minute observation
periods conducted both in the morning and afternoon in educational settings. They
reported average interrater reliabilities of 87%, which were calculated by dividing the
total number of agreements by the total number of intervals. Richards and Stemberg
(1992) used this code to ﬁnd possible correlations between endogenous and exogenous

factors and existing behavior state in individuals with SMD.

39

Richards and Stemberg (1992) reported that participants spent approximately
13% of the time asleep, 14% of the time in the no orienting state, 5% of the time in the
agitated state, 38% of the time in the awake/unclear state and 33% of the time in the
awake/orienting state. No seizure activity was observed.

A nine-category taxonomy of behavior state is described in the Carolina Record
of Individual Behavior (CRIB), an instrument for the assessment of developmental
functioning of infants and children with disabilities (Simeonsson, Huntington, Short, &
Ware, 1982). The researchers reported that they included state observations in the
assessment protocol in order to document state inﬂuence on performance and because
poor behavior state organization might be an early predictor of poor developmental
outcomes. The evaluator using the CRIB documents the child’s behavior state at the
beginning and throughout systematic observations of the child’s activities.

Guess and his colleagues have extensively documented the development of their
behavior state code which has undergone several revisions. For two initial studies on
observing behavior as reported in Guess, et al. (1988), the researchers adapted a behavior
state scale from Brazelton (1973) that described six behavior states. The states were
asleep/inactive, asleep/active, drowsy, awake/inactive/alert, awake/active/alert and
crying/agitated. A pilot study was conducted with 11 students with SMD ranging from
one to 19 years of age. All of the participants were identiﬁed as profoundly retarded with
varying degrees of physical and sensory impairment. A momentary time-sampling
procedure, that is, at lO-second intervals the observed behavior state was noted. Data
were collected for 20 lO-minute sessions conducted over 5 to 6 weeks.

A second pilot study involved 10 students aged 3-15 years. The same

observational and recording procedures were used in this study with one exception. A 10-

40

second time interval procedure was used, that is the most prevalent state during the 10-
second interval was recorded. Two coders rated 25% of the sessions in both studies and
mean interrater reliability was reported at 90 and 93%, for the ﬁrst and second studies,
respectively. The researcher reported mean percentages of time in behavior states
(averaged between the two studies) of approximately 14% of time in the two sleep states,
11% of the time in the drowsy state, 45% of the time in the awake/inactive/alert state,
20% of the time in awake/active/alert and 7% of the time in crying/agitated.

In later studies (Guess et al., 1990; Guess, Roberts etal., 1993; Guess, Siegel-
Causey etal., 1993) the researchers amended their scale to include two new categories;
stereotypy and daze. These new states were added to the original code to better describe
behavior states that the researchers observed in this population.

Guess, Roberts, et al. (1993) observed 66 participants for one 5-hour period. The
participants ranged in age from 20 months to 21years and were described as: (a) having
severe motoric limitations; (b) were not responsive to the environment; (c) were
nonverbal; and (d) received intervention implemented a basic, sensory level. In this
study, a similar overall percentage agreement (93%) between raters was reported.

The behavior coding scale was subsequently expanded to 13 categories that
included two subcategories of awake/alert/inactive and two-subcategories of stereotypy
(Guess, Roberts & Rues, 2002). The most recent modiﬁcation of the scale describes
eight state categories that will be used in future research (S. Roberts, personal
communication, April 3, 2002). This scale is reproduced in Table 3.

The meaningfulness of the construct of state and the ability to reliably measure
and report it has been questioned by Mudford, Hogg and Roberts (1997). They reported

that they were unable to replicate the interobserver reliability ratings that had been

41

reported in the Guess et al. research. Mudford and his colleagues reported that obtained
within-state reliability scores when they attempted to replicate use of the 13-category
scale with 3 participants who had profound mental retardation. In a second report,
Mudford, Hogg, & Roberts (1999) reported low agreements when ﬁve observers used the
thirteen category scale to score videotapes of 2 participants, with a 15 minute section of
tape for developing the consensual analysis. Mudford et al. (1999) recommended
training to criterion of 80% agreement on the occurrence of each component (state) of the
scale as a more stringent criterion for future researchers.

Guess and his colleagues responded that training procedures need to be carefully
implemented and that coding should not be done from videotapes (Guess, Roberts,
Behrens, & Rues, 1998; Guess, Roberts, & Rues, 2000). The Guess research group also
maintained that their current eight-category protocol (rather than the l3-category one
used by Mudford and his colleagues) would yield more acceptable reliabilities. Guess and
his colleagues have concluded that behavior state could be reliably measured when
observers received appropriate training and coding is done where the coder can closely
observe the participant in the live setting. They have developed extensive training
procedures (Guess et al., 1990) as well as a multi-media project reported by Bashinksi
(1996) to train observers who are both researchers and direct care providers (Ault, Guy,
Guess, Bashinksi, & Roberts, 1995; Guy, Guess, & Ault, 1993).

The four codes developed to describe behavior state in individuals with SMD
have been used for various research agendas. Research has been conducted to describe
state organization in individuals with SMD, evaluate the effects of an intervention on
behavior state organization and identify possible factors that contribute to state

organization. State organization refers to several attributes of state including the overall

42

Table 3

Current Behavior State Code

 

S. Roberts, personal communication, April 3, 2002

 

Sleep

Drowse

Daze

Awake/inactive/alert

Interact

Stereotypy

Person’s eyes are closed. Respiration can be slow, even or
uneven. Motor activity may occur including characteristics
such as mouthing, twitching, tossing and turning. Person
may exhibit rapid eye movements (REM). Other behaviors
may include occasional facial expressions (smiling,
grimacing, frowning) and/or vocalizations (sighing,

grunting, gurgling).

Person’s eyes are either open and eyelids appear “heavy,”
or eyes are opening/closing repeatedly. Vocalization may
occur.

Non-orientation to visual, auditory, or tactile stimuli
predominates. If person’s vision is intact, eyes are open and
appear glassy, dull and immobile. Motor movements (that
are not orienting) may occur such as brief limb/body
movements, startles. Respiration is regular.

Person’s eyes are open and responsive to visual or auditory
stimuli. Visual activity may include tracking,
orienting/focusing on stimuli, turning head, eyes looking
towards stimuli. Occasionally eyes may be closed, but
displays other signs of orientation such as smiling, learning
toward, head turning. Non-orienting movements may
occur (brief limb/body movements, startles). Demonstrates
regular respiration. Vocalizations may occur.

Person engages/interacts/manipulates by making contact
with a person or objects. Body movements to avoid stimuli
or interaction (pulling away from someone). If person’s
vision is intact, eyes are open, bright, and focused. Visual,
auditory, or tactile interactive patterns are exhibited with
distinct ﬁne and/or gross motor movements.

Person exhibits behaviors that are stereotypical

(idiosyncratic, repetitive rhythmic movements of body or
body parts). Movements may include repetitive touching or

43

Table 3, continued

Crying/ agitated

Aggression/self injury

movements such as head weaving, rocking, mouthing hand
or objects, teeth grinding, arm and ﬁnger ﬂapping.

Person may exhibit intense vocalizing, crying or screaming.
Grimacing and/or frowning may occur with or without
intense vocalizations or crying. Respiration may be
irregular and eyes may be open or closed. Overall
increased tension in body tone with accompanying
agitation behaviors (intense vocalizing, facial expressions,

crying).

Person exhibits aggressive behavior toward self or others,

which may include hitting, biting, pinching, etc. Behavior

may be accompanied by intense vocalization/screaming or
overall increased tension in body tone.

 

percentage of time spent in various states and the frequency at which an individual
changes state. An important ﬁnding was that individuals with SMD generally exhibit
organization patterns that are not optimal for learning; patterns that include a relatively
large proportion of time in non-alert states and frequent changes between states. Guess
et al. (1990) averaged the percent occurrences for each state across 50 students with SMD
who were observed in educational settings. As a group, they spent 42% of the time in
states that are not optimal for learning. Myers (1990) found that four preschoolers with
SMD spent an average of 28% of the time in non-alert states over the course of three 55-
minute observation sessions in inclusive educational settings. For comparison, she also
observed four classmates matched for age and gender with the children with SMD. The
typically developing children averaged about 10% of the time in non-alert states.

Another ﬁnding was that groups of individuals could be grouped by proﬁles that
represented similar state patterns (Guess et al., 1990, Guess et al., 1991; Guess, Roberts
et al., 1993; Guess, Roberts, et al., 1995). Table 4 lists these descriptive proﬁles. Two
aspects of these proﬁles were signiﬁcant. One aspect was that the total population
percentages were similar across studies. In other words, a cross section of participants in
each of four studies yielded similar proportions of participants in each proﬁle.

A second aspect was that these proportions were similar for all age groups. In
other words, these less than optimal proﬁles do not appear to improve with age as similar
percentages were found across three age groups. In one study, (Guess, Roberts, et al.,
1995) the proﬁles of 66 students were described. When the student proﬁles were
compared by three age groups; 20 months-7 years, 7.5-12 years, and 12.5-21 years, there

was no difference in the percentage occurrences of the proﬁles. In addition, a

45

Table 4

Descriptive Proﬁles of State Organization in Individuals with Severe Multiple
Disabilities

 

Proﬁle Descriptions

 

1 Individuals who spend 75% or more time in
preferred states

2 Individuals who spend at least 75% of time
in preferred states but less than 20% of that
time interacting with the environment

3 Individuals who spend less than 75% of time
in preferred states, often exhibit stereotypy or

crying/agitation

4 Individuals who spend less than 75% of time
in preferred states, often exhibit sleep, drowse or
daze

5 Individuals who do not ﬁt into the above
proﬁles but display a large number of states

46

series of observations of eight children with SMD from birth to age 4 years, Guess,
Roberts, and Rues (2001) found that the typical non-optimal behavior state patterns or
proﬁles stabilized for these children around the age of 3 years.

The ability to reliably assign students to these general proﬁles was also obtained
in a study of ﬁve individuals conducted by Richards and Stemberg (1993). They used a
using a different state classiﬁcation scheme of ﬁve states but found that they could
reliably assign students to these proﬁles.

In addition to large proportions of time in non-alert states, individuals with SMD
were observed to change state quite often which may have implications for their ability to
beneﬁt from any orienting responses that they do display in alert states. The 25
participants observed in classroom settings in a study by Guess, Roberts et al. (1993)
shifted states an average of 1.28 times per minute over a ﬁve-hour observation period.

Poor state organization has been described as a characteristic of individuals with
SMD, an indicator of poor quality of life, and as an obstacle to learning. If individuals
cannot maintain an appropriate level of alertness, they have diminished opportunities to
maintain orienting behaviors and therefore beneﬁt from environmental input.

Behavior state researchers have sought to identify factors that are associated with
state organization in individuals with SMD. The construct of state has been considered
within systems framework (Guess & Siegel-Causey, 1995). The task has been to identify
both internal and external factors interact to result in behavior states. Although it is the
interaction of both of these factors that result in the manifestation of certain states, certain
parameters serve as control parameters in that they have the most amount of inﬂuence on
the system at a given point in time. For example, it has been suggested that some

combination of chemical or physical anomalies associated with the physical syndromes

47

of Lesch-Nyhan, Cornelia de Lange, and Rett syndromes results in the high incidence of
stereotypy observed in individuals with these syndromes.

Other internal factors may be more inﬂuenced by environmental interactions. For
example, children with SMD often have feeding difﬁculties resulting from poor oral
motor control or gastrointestinal disorders. Feeding difﬁculty can result in low calorie
intakes and poor nutritional status (Ault, Guy, Rues, Noto, & Guess, 1994). Children with
seizure activity require medications that have side affects of drowsiness. Poor nutritional
status and some medications are associated with excessive drowse and sleep states in
some children with SMD (Ault et al., 1994).

Several studies have been conducted to look for correlations between possible
external factors and state in individuals with SMD (Guess et al., 1990; Guess et al., 1995;
Guess, Roberts et al., 1993; Richards & Stemberg, 1997). Environmental variables that
have been investigated for signiﬁcant correlations with state include body position,
presence of social contact, ambient temperature, surrounding activity levels, time of day,
and presence of materials (Guess et al., 1995; Richards & Stemberg, 1992). Of these,
only two variables, position of the student and the presence of social contact, have been
consistently and signiﬁcantly correlated with alert states.

Happiness Indices

One important component of several measures of quality of life is the concept of
personal happiness (Felce & Perry, 1995). Personal happiness is a subjective construct
often quantiﬁed by self-report when measured in populations of typically developing
individuals. As happiness is related to, but is not totally dependent on, objective
measurement of one’s life circumstances, subjective scales are considered the most

appropriate way to quantify this concept. Subjective scales are more challenging to

48

construct for individuals who are non-verbal and non-mobile. The emerging importance
of considering all aspects of quality of life for individuals with SMD has focused
attention on the development of appropriate measurements of happiness for this
population.

Research has centered on behavioral manifestations of emotions called indices of
happiness. One main goal of this research has been to construct and validate an
observational protocol that then could be used to measure outcomes of interventions
designed to increase happiness.

Green and Reid (1996) reported an early attempt to quantify happiness in
individuals with SMD. Their six participants were 18-41 years of age. The researchers
selected observable responses generally associated with the expression of happiness.
They deﬁned happiness indices as “any facial expression or vocalization typically
considered to be an indicator of happiness among people without disabilities including
smiling, laughing and yelling” (Green & Reid, 1996, p. 69).

In this initial study they used a 10-second partial interval recording procedure.
The speciﬁcs of observer training were not described other than directions to the
observers that recording an index should only be done if they were certain they observed
the index. Forty two percent of the sessions were rated by two raters and overall
agreement was 98%. An agreement was scored if any index of happiness was noted
during the interval.

In addition to demonstrating acceptable interobserver reliability of observations,
the researchers sought to validate the relationship between observed happiness indices
and actual personal happiness in two ways. First, participant videotapes were rated by

two groups of practitioners, one unfamiliar with the participants, the second, the

49

participants’ direct care providers, on how happy the practitioners perceived the
participant to be. These ratings were compared to the formal count of happiness indices
and were found to be highly positively correlated.

The second method of validation was the comparison of the number of happiness
indices exhibited by each participant during previously identiﬁed preferred activities of
the participant versus those exhibited by the participant during his or her least preferred
activities. Increased levels of happiness indices were noted during sessions when the
participant was presented with preferred activities.

A similar method was used by Favell, Realon, and Sutton (1996). They
videotaped 7 participants with SMD during recreational activities and had ﬁve college
students (no explanation of their training procedures other than it lasted less than 30
minutes was reported) observe the video tape and record occurrence or non-occurrence of
the happiness index in a 10-second interval. The indices count was then correlated with
judgments of happiness reported by two direct care providers who worked with the
clients. The validity of facial expressions, especially smiling as an indicator of happiness,
was supported by the high correlations between smiling and independent staff ratings of
happiness.

The researchers in both studies concluded that measuring happiness indices might
provide one of the few methods available for evaluating programming for people who
have difﬁculty demonstrating functional effects of reinforcement. These observational
procedures were used in subsequent studies to evaluate interventions to increase
happiness indices (Green, Gardner, & Reid, 1997; Green & Reid, 1999a; Ivancic, Barrett,

Simonow, & Kimberly, 1997).

50

Movement

The type and level of movement exhibited by individuals with SMD has been the
focus of study for several decades. However, the ability to make a voluntary movement
has typically not been considered a quality of life indicator, rather it has been studied as
the only observable indicator of learning and cognitive development in this non-verbal
population.

An increase in the frequency of trained movements was presented as evidence
that learning could occur in individuals with SMD. The availability of switches, devices
that could be activated in a variety of ways to turn a battery operated toy or device on and
off, made it feasible to automatically provide contingent responses to the individual with
severe multiple disabilities (Pronger, 1995). Many single subject studies were conducted
using a behavioral paradigm to demonstrate increased movements in response contingent
conditions.

Although studies showing increased switch use have been criticized for not
adding to quality life, it can be argued that the ability to make a voluntary movement and
the cognitive awareness of the effect of that movement can only become more important
to the quality of life of individuals with SMD in the future. The increasing availability of
computer technology to this population already allows a person with any voluntary
movement to activate an array of environmental control, communication and devices.
Thus, it can be argued that the ability and willingness to make a voluntary movement will
add to the individual’s quality of life by offering an avenue of control and choice when
typical means of communication or interaction with the environment are not possible.

Finding ways to increase non- stereotypical movements that result in meaningful

51

interactions with the environment provides the individual with more opportunity to
experiment with control and feedback.
Contingent Stimulation as an Intervention

Role in Development

One intervention that has received extensive attention is the provision of
contingent stimulation (also called contingent response) to children. Contingent
stimulation has been deﬁned as a stimulus that has an immediate and predictable
relationship to an individual’s behavior (Sullivan & Lewis, 1993). Contingent stimulation
has been characterized as a reinforcer in behaviorist frameworks (Haskett & Hollar,
1978), a necessary component in the development of cause and effect in the early
sensorimotor period in constructivist frameworks (Glickman, Deitz, Anson, & Stewart,
1995), and as an external control parameter in systems vocabulary.

Contingent stimulation can be social such as a mother’s response to a child’s
vocalization or nonsocial such as a toy that squeaks when a child squeezes it. Non-social
contingencies have been identiﬁed as an important part of the range of experiences of a
typically developing child (Brinker & Lewis, 1982a).

It has been proposed that a contingently responsive environment has positive
implication for a foundation of motivational factors as well (Lewis, Sullivan, & Brooks-
Gunn, 1985). Each time that a child experiences stimulation that is contingent on his/her
efforts to interact with an object, the child’s expectation of control over the environment
increases. Brinker and Lewis (1982b) consider that the primary developmental task
facing an infant is learning how to engage and interact with the environment.

Contingent stimulation is largely missing from the lives of individuals with SMD

(Langley, 1990; Lancioni et al., 1998). The nature of the combination of physical and

52

sensory disabilities prevents these children from being able to affect their environment in
a consistent and meaningful way. Investigation of ways to introduce contingent
stimulation into the lives of children with SMD has been the focus of many studies.
Investigations on the effect of non- social contingent stimulation on three quality of life
indicators are described next.

Eﬂect on Behavior State

No studies could be located that addressed the effect of non- social contingency
responses on behavioral state organization in children with SMD. However, the
availability of contingent non-social stimulation has been demonstrated to affect state
organization in typically developing infants. For example, Lewis et al. (1985) examined
the level of alertness that typically developing infants displayed at 10, 16, and 24 weeks
of age when engaged in a simple contingency experience. Matched (age and gender) pairs
of children were randomly assigned to a contingent or noncontingent sessions where they
faced an apparatus which provided a visual and auditory stimulus. The stimulus
consisted of the presentation of 3 seconds of a color slide of a smiling face accompanied
by a children’s tune.

Children who were in the contingent session could activate the stimulus with a
pull of a ribbon connected to their wrist. Children who were in the noncontingent session
were exposed to a similar amount of the stimulus but could not control the display. The
sessions were terminated when the infant displayed poor states such as fussiness or
drowsiness. The contingent participants averaged 14.4 minutes in an alert state, the non-

contingent control participants averaged 8.4 minutes in alert state.

53

Effect on Movement

The majority of studies of learning in young children with SMD have focused on
increasing speciﬁc motor responses in response to contingent “reinforcement” of that
movement (Kinsley & Langone, 1995). Early on, educators realized that technology
could be used to afford access to cause and effect explorations of the environment
(Behrmann & Lahm, 1982; Goldenberg, 1979). An interaction that was not previously
possible because of physical or sensory limitations could be provided by technology
through computer switches.

The contingent responses provided by switch technology have included music
(Glenn & Cunningham, 1984), visual displays and music (O’Brien, Glenn, &
Cunningham, 1994), and the delivery of edibles (Gutowski, 1996). For example, Dunst,
Cushing, and Vance (1985) reported an ABA single subject study with 6 participants who
were profoundly retarded male infants with multiple disabilities. After six sessions of
operant conditioning, 4 of the 6 participants demonstrated higher average head turns in
the contingent response phase of the study as compared to the baseline (no contingent
response) and return to baseline phases.

In another representative study (Realon, F avell, & Dayvault, 1988), the authors
investigated whether their participants with SMD would increase the activation rate of
switches during an intervention phase when the switches were connected to battery
operated leisure items (such as a vibrating doll or racing car set). In both the baseline and
withdrawal phases the switches were disconnected. The authors reported that 5 of their
10 participants with severe multiple disabilities, aged 18-42 years, increased rates of
switch activation by 40 to 50% during the intervention phase. Switch activation rates by

the participants decreased during the return to baseline.

54

Saunders et al. (2001) provided 8 female participants with SMD, aged 28-51, with
access to switch controlled leisure items. No prompting or instruction was needed for 7
of the 8 participants to activate the switch more of the time in the feedback condition as
compared to the non-feedback condition. In this study, participants were given access to
the switches for up to two-hour sessions in contrast to many studies which reported 15 to
30 minute sessions.

Similar procedures were followed in other (Glenn & Cunningham, 1984;
Leatherby, Gast, Wolery, & Collins, 1992) studies. Participants were ﬁrst observed to
discern which of their voluntary body movements might be most likely to activate a
switch. The use of the switch was demonstrated for the participants. Then, a switch was
presented without the contingent stimulation and number of activations counted. The
switch was then connected to the contingent stimulation to determine if the participants’
activations increased in this condition. A withdrawal of the contingent stimulation
followed, with a resulting decrease in activations. An interesting factor is that most
studies reported increased switch activations in the contingent condition by some, but not
all participants. Possible reasons for the non-learners were lack of reinforcing value of
the stimulation, satiation with the stimulation, lack of control of other factors such as
state and inability to control the speciﬁc movement required.

In an effort to increase the availability of contingent response for young children
with disabilities, the Contingency Intervention Curriculum was developed in the 1980’s
(Brinker & Lewis, 1982a, 1982b; Sullivan, 1990). In addition to toys operated by
switches, interventionists using this curriculum also provide infants with access to

personal computers. A switch activated by the child generates new visual and auditory

55

displays on the computer monitor. Several studies replicated the result increase in rate of
responding in contingent response conditions when the computer was used.

Research on using switches and toys with young children with SMD has been
criticized. Kinsley and Langone (1995) write that these studies pay little attention to the
generalization of switch activation skills. A second criticism was that the majority of
these studies did not include any measures of self-initiated use of motor skills. There is
also a lack of empirical support for the role that this intervention plays in long-term
improvement in the course of cognitive development in infants exposed to the
contingency intervention program as compared to those who were not (Howell, 1992;
Watson, Hayes, & Vietze, 1982).

Advances in technology have required less skill on the part of the user to activate
a response from the environment. One of these advances is movement sensor technology
that converts any movement to a response. One of several such technologies is called
Soundbeam (Jacobs, 1997). These types of technologies which provide consistent
consequences for emitted behaviors so that, at an early age children may be more likely
to learn about cause and effect and may ultimately demonstrate a broader array of
purposeﬁrl communicative and motor response than is currently possible, (Guess, 2000).
Ellis (1995, 1996, 1997) has investigated Soundbeam as a potential intervention for
children with SMD. His participants are described as having profound and multiple
disabilities.

An initial study (Ellis, 1995) was conducted with 2 participants who were
exposed to Soundbeam for a weekly 20-30 minute session for an ll-week study. He
videotaped the sessions and examined them looking for separate activities such as arm or

head movement, laughter, or speciﬁc directional movement. Ellis described the

56

progression that one of the participants made. He found an increase in the variety of
movements, increase in observable positive affect, and a progression from gross to ﬁne
movements.

In a subsequent report, Ellis (1997) described the progression of one 8 year old
participant over the course of two years. The child received a weekly session with
Soundbeam during the school year. Ellis observed a change in the participant’s
movements from apparent random and agitated movements to more controlled
movements which were smoother and less frequent.

Eﬂect on Happiness Indices

An interesting consequence reported in many studies that focused on increasing
movement in response to contingent stimulation was the observation that the participants
displayed more positive affect in the contingent condition. Sullivan, Laverick, and Lewis
(1995) reported a case study on the results of providing musical toys that could be
activated by a hand switch to one participant with Rett Syndrome. In addition to the
increase in switch activations that were noted they also reported that the participant’s
positive emotional response after six months of exposure to the contingent program also
increased. They reported that the participant displayed clear anticipatory excitement
when she was being positioned for switch activities.

In addition, negative emotional reactions have been reported when the
contingency situation has been withdrawn as Watson et a1. (1982) reported in a case
study of an infant with severe multiple disabilities. The infant had demonstrated
increased activation of switches connected to various stimulating toys. When the
contingent situation was inadvertently discontinued. The researchers reported a change

from “exuberant smiling at the ﬁsh (the stimulation) to averted gaze.” p. 200

57

Lancioni, O’Reilly, Singh, Oliva, and Groeneweg (2002) conducted a study to
compare the effect of a contingent stimulation program with a noncontingent stimulation
program on happiness indices. Three participants with SMD, aged 13-46 years, were
given access to switches during an intervention phase. The switches were connected to a
control program that activated selected events such as ﬂickering lights, massaging
brushes, musical instruments for 8-10 seconds. These events had been previously
determined to be preferred by each participant. During the stimulation phase, the
reinforcing events were presented by a research assistant for one to two minutes at a time.
Preceding the intervention phase, each participant was taught to use the two available
switches.

The authors (Lancioni et al., 2002) found that the number of indices of happiness
for 2 of the 3 participants were comparable across the intervention and stimulation
phases. The participants received less stimulation in the intervention (switch activation)
due to rate of activation and shorter availability of the stimulation provided by the switch
as opposed to the stimulation presented by the research assistant. The authors concluded
that even with less total stimulation, the contingency component was enough to provide
just as much enjoyment for two of the participants as the more labor intensive non-
contingent stimulation condition.

Synthesis

The unique needs of children with SMD have lead to a reconceptualization of
appropriate educational interventions and assessments for this population. Children with
SMD will probably not be able to obtain self-care, functional, or vocational goals.
Therefore, the exclusive use of either developmental or functional curricula to frame

educational interventions has proved unsatisfactory and a source of frustration for direct

58

care providers. These children often show little change when assessments based on
developmental or functional curricula are used to evaluate progress.

There is growing consensus that a quality of life framework may be a more
reasonable and attainable outcome for this population. Some suggested quality of life
outcomes are ability to exercise choice and some control of one’s circumstances, freedom
from physical discomfort, and access to enjoyable activities. Interventions focused on
improving these quality of life factors may provide a better outcome for students with
SMD and more satisfaction for the interventionist.

Speciﬁc assessments suggested for quality of life factors have been evaluations of
the behavior state organization of the individual, evaluations of the individual’s
enjoyment of a provided activity, and evaluation of the amount of voluntary movement
that an individual displays. Behavior state organization can be considered a foundation
on which further quality of life indicators are built. If the individual is unable to regulate
and maintain preferred states, the individual will be less able to beneﬁt from
environmental input and social interactions. Evaluation of the individual’s enjoyment of
an activity by observing the individual’s behavioral indications of happiness has been
suggested as one of the few ways to evaluate the effectiveness of activities designed for
the individual’s enjoyment. Encouragement of voluntary movement is an important
quality of life factor because the ability to make a controlled movement is one method by
which a SMD child can convey choice and control his or her environment.

Interventions to improve these quality of life factors may include direct one to one
teaching, interactions with others through social interactions, or direct application of

stimulation. Interventions can also be provided in a non-social way by arranging the

59

environment in ways that may help the SMD individual remain alert, have an enjoyable
interaction, or be encouraged to use movements.

Although many interventions are delivered effectively through one to one
teaching situations, it is also important to investigate interventions that improve quality of
life through independent play. Practitioners have indicated that lack of time is the largest
barrier to providing quality interventions. If the children can be provided with effective
uses of their own time, there is less “down time” for the child when caretakers are
occupied elsewhere.

One way of arranging an independent play situation for children with SMD is to
use technology. Technology provides children who have very limited voluntary
movement repertoires to still interact with the environment. Dozens of switches have
been developed that allow the child to activate communication devices, toys and
environmental devices with very small movements. A movement sensor can be
considered an improvement over a switch because any movement can activate the desired
feedback. One movement sensor, Soundbeam provides musical feedback to any
movement.

Obtaining these technologies, including Soundbeam, represents a signiﬁcant
commitment of ﬁnancial resources. The current study is an empirical evaluation of the

effect that Soundbeam has on three quality of life factors.

60

CHAPTER 3
Methods

The purpose of this investigation was to determine the effect of contingent
auditory stimulation of the children’s self initiated movement on three quality of life
measures in young children with severe multiple disabilities (SMD). Three research
questions were asked.

1. Is there a difference in the behavior state organization of a young child with
severe multiple disabilities across three conditions: (a) a typical play environment; (b) the
same play environment with the addition of auditory stimulation contingent to the child’s
movements; and (c) the same play environment with auditory stimulation provided
noncontingently?

2. Within intervals where the child is observed in an alert or interactive behavior
state, is there a difference in the percentage of intervals where a happiness index is
displayed across the three experimental conditions?

3. Within intervals where the child is observed in an alert or interactive state, is
there a difference in percentage of intervals in which the participant displays a voluntary
movement?

Research Design
Single Subject Design

A multiple baseline A—BC-B-BC design (Barlow & Hersen, 1984) with 2
participants was used. During all phases of the experiment the participants were observed
in independent play sessions three times a week. The baseline condition was presented
for three sessions for the ﬁrst participant and six sessions for the second participant. The

research timeline is presented in Table 5.

61

Table 5

Research T imeline

 

 

Phase
Week No. of sessions Participant 1 Participant 2
(cumulative)

l 3 A A
2 6 BC A
3 9 BC BC
4 12 BC BC
5 15 BC BC
6 18 BC BC
7 21 B BC
8 24 B B

9 27 BC B
10 30 BC BC
1 l 33 BC BC
12 36 BC BC
13 39 BC BC

 

Notes.

A= Baseline

BC= Auditory stimulation contingent to movement
B= Noncontingent auditory stimulation

62

The investigator visited each participant three times a week to observe a play
session. The ﬁrst participant’s play sessions were held at 3:00 pm on Monday, Tuesday,
and Wednesday. The second participant’s play sessions were held at 4: 15 pm on the
same three days. The participant was positioned in an upright position in her wheelchair
for each session. In the baseline (A) phase of the experiment, the child was seated in her
wheelchair with access to a typically provided play or leisure activity. In the ﬁrst
intervention phase (BC), the Soundbeam emitter was placed near the child and was
activated by the child’s movements, providing contingent auditory feedback. In the next
phase (B), the Soundbeam emitter was placed in view of the child but was activated by
the investigator rather than the child, providing auditory feedback that was not contingent
on the child’s movements. The investigator told the child that it was the investigator’s
turn to play with the Soundbeam. In the ﬁnal (BC) phase control of the auditory
feedback was returned to the child by placing the Soundbeam emitter near the child.
Variables

Independent. The independent variable was the presence or absence of auditory
stimulation contingent on the participant’s movements in the ﬁrst intervention phase and
the presence of noncontingent auditory stimulation in the second intervention phase. The
nature of this intervention consisted of two components, a novel type of auditory
stimulation and contingent control of the stimulation. In the BC phases the musical
output was directly related to the participant’s movements, that is, she was actually
“playing” the keyboard through movement rather than activating an on/off switch of a

musical recording.

63

During the B phase, the musical output was produced by the investigator rather
than the participant. That is, in each B session the participant heard similar musical
output than the previous phase but did not have control over it.

Dependent. Three dependent variables were measured for each play session. The
ﬁrst variable was the percentage of 5-second intervals that the participant was observed
in a preferred state during the session. This percentage was calculated by dividing the
number of observed preferred state intervals by the total number (120) of observation
intervals during each session.

The second dependent variable was the percentage of preferred state intervals in
which a happiness index was observed. Only intervals previously identiﬁed as preferred
state intervals were coded for the presence of a happiness index. Intervals in which a
happiness index was observed were called “happiness intervals.” The percentage of
happiness intervals for each session was calculated by dividing the number of happiness
intervals by the total number of preferred state intervals for the session.

The third dependent variable was the percentage of preferred state intervals in
which a voluntary movement was observed. Only intervals previously identiﬁed as
preferred state intervals were coded for the presence of a voluntary movement. Intervals
in which a voluntary movement was observed were called “movement intervals.” The
percentage of movement intervals for each session was calculated by dividing the number
of movement intervals by the total number of preferred state intervals for each session.

Controlled. Controlled variables were: (a) the participant’s position; (b) the
amount of social interaction between the experimenter and the participant; (c) the time of
day of the intervention; and (d) the type of leisure activity offered across all four

experimental phases. The children were positioned in typical upright positions in their

64

wheelchairs. One participant was always positioned directly upright, the second
participant’s wheelchair backrest was generally tipped back 20 degrees from directly
upright as she was unable to support her head in the straight upright position.

Brief social interactions were conducted before and after the sessions, but the
researcher avoided interactions during the session except as necessary to ensure the safety
and comfort of the participants. All sessions were conducted after the participant’s return
home from school.

Ambient lighting during the sessions was dependent on outside weather as
sessions for both participants were conducted in sun porch type rooms in the residence
and were illuminated by light from the surrounding windows. The lighting ranged from
very bright to bright. The investigator did not note any wide ﬂuctuations in room
temperatures. Background noise was controlled for the ﬁrst participant by closing a door
between the sun porch and the family living room. Some street noises or the voices of
her siblings could occasionally be heard. Background noise for the second participant
was more variable as her sessions were conducted in a room open to the dining room and
kitchen although her mother did not allow any siblings to play in the same room.

Design Rationale
Single subject designs have a long history of use in investigations where the
effect of interventions or treatments on individuals is of interest (Barlow & Hersen, 1984;
Kazdin, 1982; Tawney & Gast, 1984; Tripodi, 1994). The researcher implements a single
subject design by obtaining repeated measurements of dependent variables on one or
more participants under at least two conditions; baseline and treatment. Although these

designs are often associated with behavioral frameworks (Hersen, 1982; Ittenbach &

65

Lawhead, 1996) they are also appropriate to examine periods of development or change
in systems frameworks (Oyama et al., 2001; Thelen, 2001).

Researchers using systems frameworks to study human development are
challenged to closely observe individual trajectories of behavior while attempting to
identify and experimentally manipulate parameters that might affect changes in the
developmental course of the participant (Thelen, 2001). Single subject design, with its
repeated measurements of the individuals, provides useful information on how the human
system reacts to the control parameters of interest.

A basic characteristic of single subject designs, repeated measurements of the
same individual, results in non-independent data. Data are usually presented in graphical
form for analysis (Kratochwill & Williams, 1988; Matyas & Greenwood, 1990; Morley
& Adams, 1991; Ottenbacher, 1990). However, there is growing consensus that statistical
analyses can serve as a useful adjunct to visual analysis (Matyas & Greenwood, 1990;
Reboussin & Morgan, 1996). The appropriateness of using various inferential various
statistical procedures to analyze single subject data is a source of continuing debate
among researchers (Bass, 1987; Holcombe, Wolery, & Gast, 1994; Park, Marascuilo, &
Gaylord-Ross, 1990; Parsonson & Baer, 1992).

To conduct an effective single subject experiment, the researcher must design an
experiment that takes into account both the characteristics of the participants and the
requirements of the proposed analysis. In the current study, the investigator considered
the factors of effect size, signiﬁcance, statistical power, and N (the number of

observations in each phase length in single subject design) in the design.

66

 

Effect Size

Cohen (1988) describes effect size as the proportion of variability in the
dependent variable related to changes in the independent variable. Effect size (BS) or
treatment effectiveness describes the actual magnitude of the difference between mean
measures of groups rather than the p-value of a particular statistical test (Cornell &
Mulrow, 1999). Komrey and Foster-Johnson (1996) suggest that effect size statistics are
appropriate for use in single subject designs, as calculation of this descriptive statistic
does not rely on the independence of data points. The magnitude of effect size is not
affected by sample size (N).

Effect sizes related to the difference in means, can be calculated in different ways.
The effect size d was used in this study. The d effect size is calculated by dividing the
difference in means of two adjacent phases by the pooled standard deviation of the
phases. Comparisons were made between the A and BC phase, the BC and B phase, and
the B and ﬁnal BC phase for each of the three dependent variables. The effect size
numeral represents differences express in standard deviation units. That is, an effect size
of 1.0 represents one standard deviation of difference between means (Komrey & Foster-
Johnson,l996).

The use of this measure has not yet become commonplace in single subject
research. Cohen’s (1988) often cited conventions for psychological investigations
describes a d of .20 as small, .50 as moderate, and .80 as large. Researchers who have
conducted meta—analyses of treatment effects in mental health and education have found
that most studies obtain effect sizes between .20 and .80 (Lipsey, 1990).

A meta- analysis that reports effect sizes of interventions with individuals with

SMD has not been conducted. Therefore, there is no context in which to compare the

67

magnitudes of effect sizes obtained in the current study. Decisions on possible changes
between phases was based only on the relative size of the change for each participant
relative to changes between the other phases. It has been suggested that effect sizes
calculated by comparing the effect of interventions on the same individual are higher than
those obtained in between- group designs where there is independence of observations
(McCartney & Rosenthal, 2000). Therefore, even an effect size of .80 (“large” according
to Cohen, 1988) might not represent a large change in the current study.
Signiﬁcance

Signiﬁcance, or the possibility that any change in mean levels of the dependent
variables between phases was found by chance (Type I error), was addressed in this
study by avoiding response guided experimentation decisions on phase length. Response
guided experimentation as described by Edgington (1983) refers to a procedure whereby
a research plots each data point as it is obtained and then makes a decision whether to
change or continue the current phase of the experiment. These multiple looks at the data
can be considered analogous to repeated testing (Allison, Franklin, & Heshka, 1992).
There is the possibility that repeatedly examining the data in order to determine phase
length increases the chance that one will stop on a high or low “bounce” leading to a
greater change in level at the phase shift. Allison, Franklin, and Heshka (1992) argue that
visual inspection combined with response guided experimentation could easily lead to
unacceptably high Type I error rates. Therefore, phase lengths were predetermined in the
current study before the experiment began.
Power Analysis and Phase Length

A power analysis involves ensuring that enough data are collected and that the

level of power is set appropriately so that if an effect exists, there is a reasonable chance

68

of detecting it, thus avoiding a Type 11 error (Cohen, 1988). In single subject research,
the researcher must consider power when setting the phase lengths of the experiment to
increase the chance that if an effect is present, it will be detected. The current study
consisted of three conditions A (baseline), BC (both components of intervention), and B
(one component of intervention).

Baseline (A) phase. A desirable baseline exhibits no trend and little variability.
At the same time, a researcher should not unduly prolong the introduction of the
intervention to the participants, especially the last participant who receives the
intervention in a multiple baseline design (Kratochwill & Williams, 1988). The baseline
phase should also be long enough so that any suspected temporal cyclicity of behavior is
ruled out (Allison, 1992).

In the current study, the investigator decided a priori to set the baseline for the
ﬁrst participant at three sessions over the course of one week. The baseline for the second
participant was set at six sessions over the course of two weeks. In this way, possible
weekly cyclicity was evaluated, and a compromise between the often-conﬂicting
desirabilities of a stable yet not overly long baseline was made.

First intervention (BC) phase. The investigator planned the ﬁrst BC phase as a
learning phase for the participants. The assumption underlying this intervention was that
any treatment effect would only result if the participants gained a cognitive awareness
that their actions had consequences, that is, if they learned the contingency. Contingency
learning conditions have been correlated with increased movement and increased positive
affect in SMD participants and increased time in alert or behavior states in infant

participants.

69

It was necessary therefore, for the investigator to make an “educated guess”
(Darius & Portier, 1999, p. 92) of the number of intervention sessions that would be
needed for the participants to learn the contingency. Several studies have been conducted
on contingency learning in SMD individuals and researchers have reported treatment
effects after a wide range of intervention phase lengths. However, it can be argued that
because consumers of single subject design research rely on logical rather statistical
inference that the most appropriate reference studies would be those that most closely
approximate the conditions and participants of the current study.

Published studies most similar to the current study are the case studies of the use
of Soundbeam as an intervention for children with severe multiple disabilities reported by
Ellis (1995, 1996, 1997). He reported increased instances of positive affect and
improved movement control in his participants after 13 weekly 20-30 minute sessions
(Ellis, 1997) and 11 weekly 20-minute sessions with the Soundbeam. Therefore, the
investigator conservatively set the ﬁrst BC phase at 15 sessions (although on a denser
schedule than reported by Ellis).

Removal of the contingency (B) phase. During this phase, the investigator
subtracted the contingency component of the intervention in order to examine the
possibility that the novelty of the musical output was the salient factor in any difference
found between baseline and the BC (both components) condition. This phase was
conceptualized as a short probe of this factor. An early stopping rule was also
implemented, as previous studies involving the withdrawal of any contingency from
individuals with SMD have reported that removing the contingency experience can be
conﬁrsing and distressful for the participants (Fields, 1999; Watson, Hayes & Vietze,

1982). If the parent or investigator noted an increase in signs of unhappiness in the

70

participant such as crying or agitation that necessitated stopping the session, the control
of the Soundbeam was returned to the participant.

Final intervention (BC) phase. The length of the ﬁnal phase was slightly shorter
than the ﬁrst BC phase. At the conclusion of the experiment the Soundbeam was placed
in the participants’ classroom at school for their continued use.

Controls of Threats to Validity

The investigator followed several procedures to control threats to the internal
validity of the experiment. Kazdin (1982) lists several major threats to internal validity
that can be addressed either through design or procedures. The threats particularly
relevant to this study included history, maturation, instrumentation, statistical regression,
treatment integrity and order effects.

History and Maturation

History was a particularly signiﬁcant threat because the participants attended the
same classroom and followed the same school schedule. A multiple baseline design, in
which the intervention is introduced at different times for the 2 participants, was used to
address this threat. If changes in the dependent variable are observed in each participant
only after the intervention has been implemented for that participant, then the case is
strengthened for the inference that the intervention was the cause of the change.

The multiple baseline design also controls for maturation. The unequal phase
lengths of the current study (the ﬁrst intervention phase is longer than the other phases)
may contribute to the maturation confound and some researchers (Barlow & Hersen,
1984) recommend fairly equal phase lengths in designs to control for this possibility.

However, maturation was less of a threat in the current study because of the relatively

71

short term of the experiment (13 weeks) and the slow developmental rates typical of
children with SMD.
Instrumentation Error

As human observers collected the data in this experiment, there were several
possible threats related to observation as a data collection method. Salient threats were
observer drift, participant reactivity to observation, observer reactivity to co-rating
procedures, experimenter bias, and recording equipment failure.

To counteract observer drift, two retraining-recalibration sessions (Hartrnann &
Wood, 1982) with the research assistant were conducted after four and after eight weeks
of data collection. These sessions consisted of reviewing the original training tape of the
participant, discussion, and agreement on speciﬁc indicators of states for each of the
participants.

Participant reactivity was dealt with by conducting an orientation session in both
the classroom and the participants’ homes before the baseline sessions began. The
purpose of these sessions was to ensure that the family was comfortable with the
procedures and to give the participant a chance to habituate to the procedure. These
sessions were videotaped and these tapes were used by the investigator and co-rater to
reﬁne the protocols for the behavior state code.

Behavior state was coded during each session so the investigator was aware of
which sessions were also being rated by the research assistant. Therefore, the threats of
experimenter bias and observer reactivity were relevant. Experimenter bias refers to the
phenomenon of seeing what one wishes to see, which is the tendency to collect data that
conﬁrms one’s hypothesis or desired outcome. Observer reactivity refers to the

phenomenon of the primary observer (in this case the investigator) obtaining higher

72

interobserver reliability scores when the person knows that a session is being co-rated. To
control for these threats, two of the analyses were conducted from edited videotapes at
the conclusion of the entire experiment. The investigator coded the presence of
happiness indices and voluntary movement from a randomly sequenced set of videotapes
of all of the sessions. Then, the research assistant rated ten of the videotapes, also
randomly selected. In this way the investigator did not know which tapes would be cc-
rated for the measures of happiness and movement.

The coding procedure for behavior state was conducted during the experiment.
Eleven of each of the participant’s sessions were coded by both the investigator and the
research assistant for behavior state. Interobserver reliabilities were calculated for each
observational code. Reliabilities were calculated using interval to interval percent
agreements. A Kappa coefﬁcient (Salvia & Ysseldyke, 2001) which is generally
considered a more stringent measure of agreement than percent agreement was also
calculated. Interobserver reliabilities were reported for each participant for behavior state
category, for happiness intervals and movement intervals.

To minimize the possibility of equipment failure, two cuing tapes were produced
and each tape was re-timed weekly to make sure that the prompts to observe or record
were exact. The investigator took two identical video cameras and extra batteries to all of
the sessions.

Statistical Regression

Statistical regression refers to the fact that extreme scores tend to revert to the
mean on repeated assessments. Response guided experimentation (Edgington, 1984) was
not used in this study to minimize the possibility that a phase would be changed after a

high or low “bounce” in the data.

73

Treatment Integrity

Treatment integrity refers to the fact that the treatment the participant receives
should be consistently implemented. The experimenter developed and used a session log
to record factors related to treatment integrity at each session (Appendix D). The
surrounding environmental variables of light, background noise, and room temperature
were noted.
Order Effects

The musical feedback provided by Soundbeam is different than an on-off switch
to control music. The musical phrases produced are directly related to the user’s
movements, that is, the user is actually playing the instrument with movement. The music
is original each time, it is not merely replaying a song sequence. Therefore Soundbeam
music can be conceptualized as having both a contingent nature and a unique sound.
Both of these components, the contingency and unique type of sound were presented in
the same order for both participants. The use of the same sequence for all the participants
would make it difﬁcult to identify the salient factor (either contingency or unique sound)
if differences in mean measures of the variables were found between the baseline and
intervention conditions. However, the investigator considered it important to focus on
contingency as the main effect and therefore presented the contingent learning condition
ﬁrst. This A- BC-B-BC design sequence is recommended when a researcher wishes to
investigate whether some consequences do not inﬂuence behavior except under response

contingent conditions (Kratochwill, 1978).

74

Participants

Two children with SMD participated in this study.
Selection Criteria

The participants were children who: (a) were in the age range of 2 to 7 years old;
(b) received special education services under the Michigan category of severe multiple
impairment; (c) were non-ambulatory; (d) were non-verbal; (e) had no identiﬁed or
suspected moderate to severe hearing loss; (f) were reported to enjoy music; (g) were
reported to usually enjoy independent play; and (h) had parental consent to participate in
the study.

Recruitment and Informed Consent

The investigator contacted the directors of developmental centers in southern
Michigan to explain the purpose of the study and seek assistance in recruiting
participants. The center of the ﬁrst director who indicated willingness to participate in
the study was selected as the research site. A presentation about the study was made for
classroom teachers who then suggested potential participants. The investigator sent a
postcard with a brief explanation of the study to the family homes of potential
participants. Families who returned the postcard indicating possible interest were mailed
a detailed consent letter (Appendix A) and a video of the investigator demonstrating the
equipment.

Parents of potential participants then met with the investigator, were given an
opportunity to ask questions, review the assent procedures for their child, and saw a
demonstration of the equipment. Consent signatures were obtained if the parents
indicated interest in participation. The ﬁrst 2 participants for whom the investigator

attained parental consent were included in the study.

75

The parents selected the pseudonym used to identify their child participant in this
report. The child’s name was known only to the investigator, her dissertation chairperson,
research assistant, school principal, and classroom staff. Videotapes and written material
were accessible only to the investigator, parents, research assistant, and members of her
dissertation committee.

Description

Two girls who receive services under the category of severe multiple disability
participated in the study. They are classmates in a self-contained pre-primary classroom
in a school for children with severe disabilities. Their classroom of ten students is staffed
by a teacher and three teacher assistants. Table 6 provides a summary of participant
characteristics. Information was obtained from school records, observations, and
interviews with parents and educational staff. Short descriptions of each of the
participants are provided next.

Esther. Esther is a 6 year old female who is the youngest of three siblings. She is
diagnosed with genetic neurological disorder of unknown origin that resulted in
developmental delay, seizure disorder, and reactive airway disease. She has received
special education services since the age of 4 months. She receives nourishment via a
g-button gastrostomy button. Medications are prescribed to help control seizures and
muscle spasticity and to aid her breathing. Esther’s teacher reports that Esther shows
pleasure when being read to, can sometimes activate a single switch with assistance, and
will show anticipation of her turn. She can search with her eyes for the source of sounds.
She demonstrates voluntary movement of her right arm, although an ATNR reﬂex limits

her ability to explore and manipulate objects. Esther smiles at others and vocalizes

76

Table 6

Characteristics of Participants

 

 

Esther Lucy
Age 6 years 4 years
Diagnosis Spastic quadriplegic cerebral palsy, Rett Syndrome
reactive airway disease
Physical Non-ambulatory, ATNR limits Non-ambulatory,
Abilities exploratory abilities, some voluntary limited functional use
movement in arms and hands of hands, full range
of motion in arms and legs
Sensory Normal hearing, Mild hearing loss,
Abilities vision corrected for refractive error normal vision
with contact lenses/ glasses
Adaptive Vineland Scores Vineland Scores
Behavior Communication 1 mo Communication 5 mo
Daily Living Skills 3 mos Daily Living Skills 14 mos
Socialization 1 mo Socialization 9 mos
Medications Phenobarbital Depakote
Baclofen Albuterol (prn)
Rynatan
Sulfatrim
Albuterol (pm)
Educational Classroom- 30 hrs Classroom - 30 hrs
Services Occupational therapy- 30 min Occupational therapy -30 mi
(per week) Physical therapy- 30 min Physical therapy- 30 min

Speech therapy— 30 min

Music therapy— 30 min

Teacher Consultant- 30 min
Therapeutic swim- 30 min

Nurse attendant during transportation

77

Speech therapy- 30 min
Music therapy- 30 min
Therapeutic swim- 30 min

Table 6, continued

 

 

Esther Lucy

Nutrition Receives complete formula Fed fork-mashed foods,
via gastrostomy button developing rotary chew

pattern

Adaptive Custom molded seating system, Custom molded seating

Equipment dynamic ankle-foot orthoses, system, dynamic ankle-
butterﬂy shoulder harness, foot orthoses, butterﬂy
Hensinger collar shoulder harness, standing

Dani wheelstand

 

78

attitudes. She currently receives the services of a teacher consultant, physical and
occupational therapist, adapted physical education teacher and music therapist in addition
to her classroom placement. She attends school 25 to 30 hours a week from September to
June and 15 hours a week June through mid-August.

Lucy. Lucy is a 4 year old female who is the youngest of four siblings. She has
received special education services since she was 11 months old and was
diagnosed with Rett Syndrome at 2 years of age. Rett Syndrome is neurological disorder
caused by a defective gene on the X chromosome and is characterized by slow physical
grth leading to small stature for age, apraxia, display of stereotypical hand
movements, and communication difﬁculties. Lucy’s development appeared normal until
about 10 months of age, when she began to display the typical symptoms of the disorder,
including loss of acquired speech and hand skills. Although she has full range of motion
in all extremities, she is non-ambulatory and has very limited functional use of her hands.
Lucy can sit independently. Her hand stereotypies are characterized by repeated
movements of both hands toward her mouth and tongue.

Lucy will smile or fuss to express her emotional state. She can make simple
choices by looking at the preferred object. Her mother reports that Lucy enjoys watching
educational videos and will fuss when the video ends. Lucy enjoys interacting with her
siblings and has deﬁnite preferences for certain toys. She currently receives the services
of a physical and occupational therapist, adapted physical education teacher, and music
therapist in addition to her classroom placement. She attends school 25 to 30 hours a

week from September to June and 15 hours a week from June to mid-August.

79

Instrumentation
Behavior State

Description. The participants’ quality of behavior state organization was
measured with an observational code developed by Guess and his colleagues (S. Roberts,
personal communication, April 3, 2002). This observational protocol and behavior state
taxonomy (Appendix B) is the latest revision in a series of codes used by Guess and
colleagues. The researcher using the code is prompted by an audiotape recording to
observe the participant for a 5-second interval and then is given 5 seconds to record the
dominant behavior state observed.

Validity. The usefulness of the construct of behavior state organization as a
predictor of future development and as a differentiating characteristic between groups of
individuals has been demonstrated in research on infants and individuals with SMD.
Poor behavior state organization is a predictor of poor developmental outcomes. Studies
of premature infants (Davis & Thoman, 1987), infants with Down Syndrome (Prechtl,
1974), and infants with SMD (Guess, 2000) have reported state cycles that differ from
typically developing infants.

Reliability. As is customary with observational codes, interobserver reliabilities
of this code have been reported in previous studies using the percent agreement formula.
A recent study reported overall agreement scores of 90 % (Guess, Roberts, & Rues,
2002)

Measures obtained. For each 20-minute session 120 ﬁve-second intervals of
observed dominant state were observed and recorded. The results from each session
yielded a number between 0 and 120 in each of eight behavior state categories with the

total number being 120. The number of observed preferred state intervals was divided by

80

the total number of intervals to yield a percentage of preferred state intervals for each
session.
Happiness Indices

Description. To quantify the construct of happiness, the investigator used the
Indices of Happiness observational protocol (Green & Reid, 1996). An observer using
this protocol (Appendix B) is prompted by audiotape to observe the participant for a
designated interval and then records whether an index of happiness was observed in the
participant at any time during that interval. An index of happiness is deﬁned as any
facial expression or vocalization typically considered to be an indicator of happiness
among people without disabilities. Examples include smiling, laughing, and yelling while
smiling (Green & Reid, 1996).

Validity. The developers of this instrument sought to validate the relationship
between observed happiness indices and actual personal happiness in two ways (Green &
Reid, 1996). In the ﬁrst evaluation procedure, 4 participants with SMD were assessed on
preferred and least preferred stimulus using an approach and avoidance procedure.
Subsequently, both the previously assessed most and least preferred stimuli were
presented to the participant for equal time periods for two presentations each for three
sessions. These sessions were coded by an observer who was unaware of which stimuli
represented the most and least preferred stimuli. Two participants exhibited higher
percentages of happiness intervals, 50% and 25% when presented with their most
preferred stimuli as compared to percentages of happiness intervals of 7% and 3%,
respectively, when presented with their least preferred stimuli. The second 2 participants

exhibited no indices of happiness, but exhibited more frequent percentages of indices of

81

unhappiness, 68% and 47%, when presented with their non-preferred stimuli as compared
to percentages of 40% and 0% respectively, when presented with preferred stimuli.

In a second evaluation procedure, Green and Reid, (1996) asked 18 practitioners
to rate videotapes of four participants on how happy the practitioners perceived the
participant to be while completing an activity. The practitioners’ ratings were compared
to a formal count of happiness indices taken earlier by the researchers during the taping
of the activity. For two of the participants, 95% of the practitioners reported that they
appeared to be happier in a tape segments showing no observed indices of happiness
versus tapes showing previously counted indices of unhappiness. For the third
participant, 100% of the practitioners reported that he appeared happier in a tape segment
containing previously counted indices of happiness versus a tape with no previously
counted indices of happiness. For the last participant, 95% of the practitioners reported
that he appeared to be happier in the tape showing previously counted indices of
happiness versus a tape showing no observed indices of happiness.

Reliability. The developers of this code reported interobserver reliabilities of
using a percentage agreement on an interval-by-interval basis. Recent studies using this
code have reported an overall agreement of 98% across 5 participants (Green & Reid,
1999a) and an overall agreement of 99% across 3 participants (Green & Reid, 1999b).

Measures obtained. Only preferred state intervals observed during a session were
coded for happiness indices. The number of intervals in which a happiness index was
observed (called “happiness intervals”) was divided by the total number of preferred state

intervals during the session to yield a percentage of happiness intervals for the session.

82

Voluntary Movement

Description. An observational measure (Appendix B) was used to quantify the
amount of movement during a session. Voluntary movement was deﬁned differently for
each participant using information provided by the participant’s caregivers. Voluntary
movement for Esther was deﬁned as any movement of her right arm or hand. Voluntary
movement for Lucy was deﬁned as any movement of either leg or foot away from her
wheelchair foot rest.

Validity. Quantiﬁcation of movement by counting or discrete movements has
been used in previous studies as a measure of understanding of cause and effect.
Increases in speciﬁed movements under operant conditioning paradigms have been noted
in several studies of response-contingent learning (Burch, Clegg, & Bailey, 1987; Dunst,
Cushing, & Vance, 1985; Lancioni, etal., 2002).

Reliability. Interobserver reliabilities on counts of movements have been
reported in previous studies (Burch, Clegg, & Bailey, 1987; Dunst, Cushing, & Vance,
1985; Lancioni, et al., 2002) ranging from 89% to 100%.

Measures obtained. Only preferred state intervals observed during a session were
coded for presence of a voluntary movement. The number of intervals in which a
voluntary movement was observed (called “movement intervals”) was divided by the
total number of preferred state intervals during the session to yield a percentage of
movement intervals for the session.

Intervention

The contingent auditory stimulation provided in this experiment was generated by

a movement sensor system called Soundbeam (Swingler, 1998). Soundbeam is one of

several existing movement sensor technologies that provides auditory feedback to any

83

movement of the user. The Soundbeam emitter sends out an ultrasonic beam, which is
aimed at the user. The user’s movements in the beam are transformed into MIDI streams
that are immediately interpreted into musical sounds produced on an attached electronic
keyboard. The sensitivity and range of the beam can be adjusted to the type of
movements the user is capable of demonstrating. There is a direct correlation between
the speed, note range, and pattern of the musical phrase to the speed, direction and range
of the user’s movement.

Of the currently commercially available non-tactile musical interfaces,
Soundbeam has been described as the most user-friendly for the interventionist (Ellis,
1995; Jacobs, 1997). Soundbeam is marketed by TFH industries in the United States as a
therapeutic device for use by individuals with SMD. Soundbeam equipment has been
placed in hundreds of special schools and hospitals in over 20 countries (Swingler, 1998).
An illustration of the Soundbeam components is presented in Figure 1.

Data Collection

Data Collection Schedule

The children participated in three 20-minute play sessions a week over the course
of 13 weeks for a total of 39 sessions. The families selected a weekly schedule that best
suited their needs for the duration of the study. Both participants’ families selected a
Monday, Tuesday, Wednesday schedule. Esther’s sessions were held at 3:00 pm in the
afternoon and Lucy’s sessions were held at 4: 15 pm on each of these days.

Both participants completed all 39 sessions during the 13 week time frame.
During the ﬁfth week of the experiment Esther completed the third session on a Friday

due to a scheduling conﬂict for Wednesday’s session that week.

84

Keyboard
K)
O O O 0

illll'l'llll'llllllllll'll'll'l'll'l'lll'l‘lll

O
0000

(l
m

 
   

\

C3)
/©
0 /
Controller

 

Research Environment

All sessions were conducted in the participants’ homes. The sessions were
conducted in a room designated by the family. The investigator brought and set up the
equipment for each session. Esther’s sessions were conducted in an enclosed sun porch
that was separated by a door from the family’s living room. Lucy’s sessions were
conducted in a family room directly off the kitchen and dining room of her home. Both
rooms had rows of windows on three sides and were illuminated by natural light.

Preferred activities were selected by the families for both participants to be
available for the entire experiment. Esther had access to a collection of hanging toys
placed by her wheelchair. Lucy was placed near a television set to watch children’s
videos. Esther’s research environment is shown in Figure 2. Lucy’s research
environment is shown in Figure 3.

Data Collection Procedures

Equipment for data collection included a Sony digital Handycam video camera
with tripod, a tape recorder with the cueing tape, and headphones for each observer. The
headphones were connected to the same tape recorder. Observations were recorded on
protocols attached to clipboards.

The experiment began in the A or baseline condition. The participant was
provided with a preferred leisure activity (hanging toys for Esther, children’s video for
Lucy, and observed. The Soundbeam equipment was placed in the room but not
activated.

In the ﬁrst intervention phase (BC), the Soundbeam was activated. The
Soundbeam emitter was aimed at the body part where the child exhibited some voluntary

movement (Esther’s right arm and Lucy’s legs). Esther’s typical position with the

86

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

J1
l:’
I
=
_ Esther
E
E . E Camera
xercrse :: /
E ui ment \
q p Emitter 6
Shelves Observers

 

 

 

Living Room

Figure 2. Research Environment for Esther

87

 

Front
Porch

 

at

Television

 

 

 

Dining Room

OD

 

 

 

Lucy

 

 

Emitter

Observers

‘Camera

O

 

 

 

 

 

Lit nun

 

R
0

Figure 3. Research Environment for Lucy

88

Kitchen

 

Soundbeam emitter is depicted in Figure 4. Lucy’s typical position with Soundbeam
emitter is depicted in Figure 5.

During the third experimental phase (B), the control of the Soundbeam was taken
over by the investigator. The participants were told and shown that the investigator
would now control the music. The investigator moved her hand in the beam to generate
music while the participant was both moving and not moving. An attempt was made to
deliberately generate music during a similar proportion of time that the participant
generally had activated the beam in the sessions in the previous BC phase. The fourth
phase (BC) was a return to the conditions of the ﬁrst BC phase.

The investigator coded all sessions for the behavior state variable using the
behavior state protocol (Appendix B). The investigator was seated in front of each
participant and wore headphones to listen to a cueing audiotape. The 20-minute
audiotape cued the investigator to observe for a ﬁve second interval and then record the
predominate behavior state in that interval.

Eleven sessions were also observed by the research assistant in order to assess
interobserver reliability. A single jack, double headphone was used so that both the
investigator and research assistant listened to the same cueing tape.

At the conclusion of each session the investigator entered ﬁeld notes on the
session. She recorded observations on the conditions, statements and questions asked by
the parents, unusual incidents, information on antecedent events, and general impressions
and questions about the session. In addition, she prepared a summary on the session
which was e-mailed to her dissertation chair for discussion and reﬂection.

In order to complete the subsequent analysis of the second two variables of

happiness and movement, the audiotape from each session was later dubbed onto the

89

 

ition of Esther with Soundbeam Emitter

Figure 4. Typical Pos

90

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 5. Typical Position of Lucy with Soundbeam Emitter

91

videotape using the audio dubbing feature of the camera. Therefore, a speciﬁc interval
recorded on the coding sheet could be matched to the corresponding interval on the
videotape. All session videotapes were then edited at the conclusion of the experiment to
contain only the intervals in which the child displayed a preferred (awake/alert/inactive or
interact) behavior state. These videotapes were used for the analysis of happiness indices
or voluntary movements that occurred in these intervals.

Key Personnel and Qualiﬁcations

All sessions were conducted by investigator. The investigator has over 20 years
of experience working with students with several types of disabilities and their families.
Her experience includes 10 years in a position where she regularly made home visits as
part of an early intervention program for children with disabilities. She visited the
behavior state research team (Guess and colleagues) at the University of Kansas and
completed training in the use of the Behavior State coding protocol at that site. She also
attended an introductory session on the use of Soundbeam with individuals with
disabilities conducted by the TFH (USA) Ltd. Company.

The investigator recruited a research assistant from a pool of individuals who
responded to an advertisement. The researcher assistant possessed an Ed. S. degree from
the College of Education at Michigan State University and had previous research
experience using observational protocols. The assistant was familiar with informed
consent procedures, conﬁdentiality, and research procedures.

The investigator provided training procedures recommended by Roberts (1992)
for the behavior state coding protocol. This training included: (a) viewing an
introductory videotape on the construct of state; (b) memorization of the descriptions of

the eight states, demonstrated by a score of 100% on a written assessment; and (c) coding

92

to 85% agreement with the investigator on videotape samples using the 5-second interval
observation protocol form on both general training videotapes and videotapes of the 2
participants in the study. A written protocol and speciﬁc characteristics exhibited by
each participant was developed jointly by the investigator and the research assistant.

Training was also provided for the Indices of Happiness protocol following
descriptions provided in previous studies (Green & Reid, 1996, 1999a, 1999b; Green,
Gardner & Reid, 1997; Ivancic, et al., 1997). This training included reviewing the
behavior state deﬁnitions and practicing the observation system along with the
investigator.

Training for the amount of movement protocol included review of the deﬁnition
of voluntary movement for each child and practicing the coding system along with the
investigator.

Data Analyses

Data were analyzed for the three research questions in the following manner.
Research Question 1

The question on the difference of quality of behavior state organization was
analyzed by comparing the percentage of intervals that the participant was observed in a
preferred state (awake/inactive/alert or interact) across the phases of the study.

The data collected for this question was taken from the session coding forms,
which yielded 120 observations for each session. The quality or the total percentage of
intervals in preferred state was calculated by dividing proportion of intervals recorded in
awake/alert and interact states by the total 120 intervals.

The percentage of preferred state intervals in each session was graphed yielding a

graph with 39 measures across four phases. To evaluate the amount of change in the

93

mean of these measures between the phases, an effect size (ES) descriptive statistic was
calculated (Komrey & Foster-Johnson, 1996). The ES was calculated by obtaining the
difference of the means between two adjacent conditions and then dividing by the pooled
standard deviation of both conditions. Therefore, ES was calculated for the difference
between Phase A and B, the difference between B and BC and the difference between BC
and the ﬁnal B. The effect sizes were calculated after visual examination of the graphed
data for clearly visible effects or trends.

Research Question 2

For this question, preferred behavior state functioned as a control variable. All
intervals in which a preferred state was observed were viewed on an edited tape of each
session. Each interval in which an index of happiness was observed was counted as a
happiness interval. The total number of happiness intervals was divided by the total
number of preferred state intervals to yield a percentage. For example, if 100 intervals
from the session were identiﬁed as being in a preferred state and 30 of them also
contained at least one index of happiness, then the happiness indices measure for that
session was 30%. The happiness indices measures were graphed, yielding 39 data points
across the four phases.

The graph was inspected for clearly visible trends and effects. In the absence of
trend, the effect size was again calculated to help describe the magnitude of change
between adjacent phases.

Research Question 3

In this question, preferred behavior state again functioned as a control variable.

The percentage of intervals that the participant displayed a voluntary movement in the

beam was calculated and compared across phases.

94

The same edited tape of preferred state intervals was used to determine the
percentage of intervals in which the participant displayed a voluntary movement. The
data points were once again graphed and inspected for clearly visible trends and effects.
In the absence of trend, the effect size statistic was calculated to describe the magnitude

of change between adjacent phases.

95

CHAPTER 4
Results, Discussion, and Conclusions

The overall purpose of this study was to assess the effectiveness of a movement
sensor called Soundbeam on the quality life of young children with SMD. Speciﬁc
research questions focused on the impact of Soundbeam on preferred behavior states,
indices of happiness, and the level of voluntary movement across four phases of a single-
subject design. These phases were A (preferred leisure activity), BC (the leisure activity
plus contingent auditory stimulation to the child’s movements generated by Soundbeam),
B (leisure activity with noncontingent auditory stimulation), and a return to the BC
condition. Information in this chapter includes: (a) a description of the participants
beginning behavior state proﬁles; (b) presentation and discussion of the results for each
research question; and (c) conclusions based upon the results.

Behavior State Proﬁles of Participants

Previous research (Guess et al., 1993) has demonstrated that individuals with
SMD can be described by behavior state proﬁles. These proﬁles are based upon the
percentages of time that the individual spends in each of the eight behavior states of
sleep, drowse, daze, awake/alert, interact, stereotypy, crying/agitated, or self-injury. In
addition to information on physical and cognitive abilities, it has been suggested that
behavior state proﬁles can be used as a further descriptor of the individuals with SMD.

The data presented in this section represent information on the behavior state
proﬁles of the 2 participants. Speciﬁc interventions and educational programs might be
more or less appropriate for individuals with different state proﬁles.

The behavior state protocol used in this study contains 120 ﬁve-second intervals

per twenty minute session during which the observer records the predominant state

96

observed for that interval. The total number of intervals coded for each behavior state
category was summed and divided by the total number of session intervals to obtain a
percentage of time in each of the eight possible categories in the baseline phase of the
experiment.

A behavior state proﬁle is generally developed by a series of observations taken at
different times over several days (Guess et al., 1993). However, each participant in this
study was observed at a speciﬁc time each day. Esther was observed ﬁrst in the mid-
aftemoon and Lucy was observed next in the late aﬁemoon. The observations on their
behavior states at those times during the baseline session and comments from their
caretakers about their behavior states during the rest of the day were used to determine
what proﬁle could be used to describe them as they began the experiment.

Reliability of Observations

Interobserver reliability for the behavior state code was assessed by calculating
three measures of interval to interval agreement between the investigator and the research
assistant, as well as an overall Kappa statistic. These measures were overall percentage
of agreement, percentage of occurrence reliability, and percentage of non-occurrence
agreement (Foster & Cone, 1986). The results of these calculations for each participant
and behavior state category are presented in Table 7.

The ﬁrst agreement coefﬁcient, overall percentage of agreement, is calculated by
summing the number of agreements on occurrence and agreements on non-occurrence of
the behavior and dividing by the total number of intervals observed. This agreement
coefﬁcient is greatly dependent on chance and is most suitable when occurrence and
non-occurrences of behavior are relatively equivalent as it can be inﬂated by the high

number of agreements on non-occurrence of behavior. Therefore, in categories where

97

Table 7

Interobserver Agreement Percents for Individual Behavior States

 

 

Overall Percentage Percentage
Percentage Occurrence Non-Occurrence
Agreement Agreement Agreement
Sleep
Esther 99% 79% 99%
Lucy NA NA NA
Drowse
Esther 98% 68% 98%
Lucy NA NA NA
Daze
Esther 98% 43% 98%
Lucy 100% 75% 100%
Awake
Esther 89% 81% 78%
Lucy 94% 80% 92%
Interact
Esther 91% 79% 87%
Lucy 98% 36% 98%
Stereotypy
Esther NA NA NA
Lucy 94% 100% 83%

 

Note. A designation of NA in this table refers to a behavior state that was not observed

during the intervals that were co-rated for interobserver agreement. The behavior state of
self-injury was not observed in either participant.

98

there is a low incidence of observed behaviors, the value of this agreement coefﬁcient
will tend to be high.

During the entire experiment, Esther exhibited a low occurrence of the behavior
states of sleep, drowse, and daze. Lucy exhibited a low occurrence of the behavior state
of interact. High rates of agreement on the non-occurrence of those states are reﬂected in
the relatively high overall percentage agreement coefﬁcients.

The second agreement coefﬁcient, percentage occurrence agreement, is calculated
by dividing agreements on occurrence by agreements plus disagreements. This agreement
coefﬁcient has generally been used in previous behavior state research that used a similar
coding system (Guess et al., 1990; Guess et al., 1993). The values obtained in previous
research have ranged from 42% t0100% (Guess et al., 1990; Guess et al., 1993).

As was the case in previous studies of behavior state by Guess and his colleagues,
in the current study it was more difﬁcult to attain acceptable reliabilities for some
behavior states than others. For Esther, the lowest interobserver agreement was obtained
for the daze state. This is the category that low interobserver agreement was also
reported in previous studies (Guess et al., 1990; Guess et al., 1993). Examination of the
protocols indicated that the states of daze, drowse, and sleep were difﬁcult for the
investigator and research assistant to agree on. The issue of differentiating between daze
and drowse was addressed at both recalibration sessions held during the course of the
study.

The investigator and the research assistant noted that it was sometimes difﬁcult to
closely observe Esther’s eyes when her lids were lowered and her head tilted back against
her wheelchair head rest, which was a common denominator in all three of these states.

In addition, the ﬁrst appearance of sleep intervals was not seen until a session near the

99

end of the experiment and it proved difﬁcult for the investigator and research assistant to
agree on he exact occurrences of these states during each interval, as Esther moved back
and forth from sleep to drowse in that session.

Although the percentage occurrence agreement of the two preferred states,
interact and awake/alert, for Esther are higher, further examination of the protocols
indicated that when there were disagreements between the investigator and research
assistant involving these two states, it was generally when one observer marked interact
and the other coded awake/alert. These disagreements were probably caused by timing
discrepancies. As the coding protocol speciﬁes that the coded state represents the
majority of the 5-second interval, when both states were observed within one interval, the
state observed for 3 seconds was coded. Esther often rapidly alternated between these
two states as she touched and withdrew from her toys. A difference in one second of
observation time noted by the investigator and research assistant was enough to result in a
disagreement between these two states.

To summarize, further examination of the protocols for Esther indicated that
disagreements between the investigator and research assistant were generally found
within the aggregated category of non-preferred states (daze, drowse, and sleep) or within
the aggregated category of preferred states (interact and awake/alert).

The investigator and research assistant were unable to reach acceptable
reliabilities on the category of interact for Lucy. The category of interact seemed to be
problematic for coding because it was a transitional state for Lucy. The participant is
coded in interact state when she purposely touches or manipulates something. As Lucy
didn’t have many objects available for play, her physical contacts consisted of grabbing

her bib or her paciﬁer. She invariably brought these objects to her mouth. Chewing on

100

objects was coded as stereotypy. Therefore, most of Lucy’s actions that were coded as
interact were observed for just part of an interval. When this sequence happened during
the ﬁve second interval, the investigator might see 3 seconds of interact versus 2 seconds
of stereotypy, thus coding the interval as interact. The research assistant might see 2
seconds of interact versus 3 seconds of stereotypy, thus coding the interval as stereotypy.
The reach for objects could also be preceded by a period of awake/alert state, causing a
possible disagreement on whether the interval was awake/alert or interact if the timing
was different of these observations by one second.

The incidence of interact state was very low for Lucy, only 33 of the total 1320
intervals observed by both the investigator and research assistant over the course of the
experiment were marked by either the investigator or research assistant as this state. Of
those 33 intervals, the investigator and research assistant disagreed on 21 intervals.
Eleven of these disagreements were an awake/interact split and ten of the disagreements
were a stereotypy/interact split. This meant that only 10 of the intervals were a
disagreement about a preferred and non-preferred state.

The third agreement coefﬁcient, percentage of non-occurrence agreement is
calculated by dividing agreements on nonoccurrence by disagreements plus agreements
on nonoccurrence. Therefore, the values for this agreement will tend to be higher when
there are relatively fewer incidences of the behavior in that category. This was the case in
the current study for the state categories of sleep, drowse, and daze for Esther and the
state category of interact for Lucy.

The overall Kappa statistic corrects the formula for the ﬁrst agreement coefﬁcient,
overall percent agreement by subtracting change agreement on occurrence and on

nonoccurrence from both the numerator and the denominator. The Kappa values

101

obtained for Esther were .88 for sleep, .80 for drowse, .59 for daze, .77 for awake, and
.81 for interact. The Kappa values obtained for Lucy were .86 for stereotypy, .52 for
interact, .87 for awake, and .86 for daze.

Overall, the percentage agreement occurrence values and Kappa statistics
represent acceptable interobserver reliabilities for the purposes of this study. The lower
values for some of Esther’s categories represent disagreements between categories that
were aggregated into non-preferred state for further analysis. The problematic reliability
for Lucy’s category of interact is somewhat alleviated by overall low incidence of the
state for her and the fact that about half of the 21 disagreements on interact actually
resulted in coding for the aggregated preferred states that were used for further analysis.
Results for Esther

Esther spent 97% of the observed intervals in either of the preferred states during
the baseline phase. The total percentage of intervals that Esther was observed in the
various behavior state categories during the baseline sessions is displayed in Figure 6.

Esther’s overall quality of behavior state falls into Proﬁle 1 described by Guess et
al. (1993). Individuals in the Proﬁle 1 category spend more than 75% of time in either
awake/alert or interact (preferred) states with at least 20% of that time in the interact
state. This impression was conﬁrmed by her classroom teacher who reported that Esther
generally appears alert during the day and rarely sleeps or appears drowsy at school
unless she is ill.

It should be noted that Esther’s percentage of time in the interact state during the
baseline phase could be a result of the fact that toys were positioned next to her

wheelchair as her preferred leisure activity. Esther was only able to reach with her right

102

Interact
43%

 

Figure 6. Percentages of State for Esther in Baseline Phase

103

arm and hand and had limited range of motion in that arm. Similar amounts of time in
the interact state would probably not be observed during the entire day if toys and objects
are not continually placed in close proximity to her right side.

Results for Lucy

Lucy spent 45% of the observed intervals in either of the preferred states during
the baseline phase. Figure 7 displays the total percentage of intervals Lucy was observed
in various behavior states in the baseline phase.

Lucy’s overall behavior state quality falls into Proﬁle 3 as described by Guess et
al. (1993). Individuals in this proﬁle spend less than 75% of the time in one of the
preferred states and may spend much of time engaged in stereotypical behaviors. This
categorization was conﬁrmed by Lucy’s teacher who reported that Lucy spends a large
percentage of time in school engaged in stereotypical behaviors. However, she also
reported that Lucy sleeps during the morning bus ride to school and takes a two hour nap
in the afternoon. She was not observed sleeping during any of the experimental sessions
conducted in the mid afternoon, suggesting that although she does need to sleep during
the day, her sleep times are distinct and predictable.

The very low percentage of interact state observed during the experiment was
understandable considering that the preferred leisure activity provided during each
session was playing a video on the television for her. She did have access to her bib with
varied textures and attached objects but her toy tray was kept at school. Her teacher
reports that Lucy will interact more with objects if someone sits with her and holds her
non-dominant hand while encouraging her to reach for speciﬁc objects. Restraining or

weighting the non-dominant hand is a tactic that has been used successfully to increase

104

Crying/agitated
2%

    

Alert
43%

Stereotypy t:
5 3% i ’

Interact
2%

Figure 7. Percentages of State for Lucy in Baseline Phase

105

functional use of the dominant hand in some children with Rett Syndrome (Katsiyannis,
Ellenburg, Acton, & Torrey, 2001).
Discussion

The two participants began the experiment with different behavior state proﬁles.
Esther’s behavior state proﬁle is considered an optimum proﬁle for learning. Lucy’s
behavior state proﬁle is considered less optimum as stereotypical behavior often
interferes with the individual’s ability to attend to and interact with the environment.

Individuals with SMD can be categorized in one of ﬁve proﬁles (see Table 4), and
a recent study reported that these proﬁles tend to be established around age three (Guess,
Roberts, & Rues, 2002). Only two external conditions, position of the child and social
interaction, have been correlated with preferred state in earlier studies, suggesting that the
internal characteristics of an individual may exert a stronger inﬂuence over state quality
than interventions that change environmental parameters, at least after infancy.

Time in Preferred States

Research Question 1 of this study investigated whether there was a difference in
the percentage of time that the participants spent in a preferred behavior state across the
four experimental phases of the experiment. The quantity of time in either of the
preferred states is one attribute of behavior state organization. As preferred states are
presumed to be the most conducive states for observation of and interaction with the
environment, researchers posit that these states are where learning occurs.

The addition of Soundbeam to their play sessions had different effects on the two
participants. Esther’s percentage of intervals in preferred state was unaffected, while
Lucy’s percentage of intervals in preferred state actually decreased. The mean

percentage of intervals of preferred state for both participants in the four experimental

106

phases is presented in Figure 8. The session percentages were calculated by dividing the
number of preferred state intervals by the total number of intervals of the session (usually
120). Some session percentages had to be calculated with less than 120 intervals due to
recording equipment failure. Esther’s Session 1 was 103 intervals and Session 28 was 99
intervals. Lucy’s session 6 was 101 intervals and session 14 was 119 intervals due to a
problem with the camera.

Table 8 presents the actual number of intervals in preferred state across the four
phases for the 2 participants.
Reliability of Observations

Reliability data for behavior states, including the preferred states of awake/alert
and interact, can be found in Table 7. The inter-observer reliability for the behavior state
code protocol was reported in the previous section on percentages of all behavior states.
Two of the behavior state categories, awake/alert and interact, were aggregated into one
category of preferred state for analysis.
Results for Esther

The addition of Soundbeam music both contingent and noncontingent to Esther’s
movements did not change the percentage of intervals that Esther was observed in
preferred state. The individual measures of preferred state percentages for each of
Esther’s sessions are presented in Figure 9. She displayed a fairly constant and high level
of percentage in preferred state across all sessions. During sessions 35 and 39, Esther’s
displayed symptoms of a respiratory illness which worsened during the session, resulting
in a higher incidence of drowse and sleep states. Descriptive statistics for each phase are
presented in Table 9. The large standard deviation in the ﬁnal BC phase is the result of

the two sessions of illness.

107

100%

90%

70%

60%

50%

40%

30%

20%

Mean Percentage of Preferred State Intervals

0%

80% '

10% .

 

 

 

 

Phase

Figure 8. Mean Percentage of Preferred State Intervals for Both Participants

108

I Lucy
a Esther

Table 8

Number of Preferred State Intervals for Esther and Lucy

 

 

A BC B BC

Esther

Awake/Alert 186 1 l 10 495 l 162

Interact 147 652 199 489

Total Intervals 343 1800 699 l 800
Lucy

Awake/Alert 301 500 190 357

Interact l6 1 5 6 9

Total Intervals 701 1799 668 1440

 

109

mm mm mm mm

p

F _ H r p

h F L

_

mﬁommmom

3 am mm mm mm mm a C

2

2

 

 

 

 

 

 

 

 

 

.xco

o\oo_

o\oom

o\oom

.xcov

o\oom

face

Axeon

Axcow

e\eom

e\ooo_

SIB/U910] 91918 pauajord JO aﬁeruaored

Figure 9. Percentage of Preferred State Intervals for Esther in all Sessions

110

Calculation of the effect size magnitudes between the phases yielded an effect
size of .46 between phase A and phase BC, .56 between BC and B, and .39 between B
and the ﬁnal BC phase. These relatively small effect sizes conﬁrm the visual
examination of the graph. As described in Chapter 3, the effect sizes obtained in this
study are used to describe relative changes in means for each participant.

Preferred states have been described as learning states in the behavior state
literature, and observations of Esther seemed to conﬁrm this idea. When she was coded
in these states she was reaching for toys, holding them at different angles, moving them
around using different motions, and using different ﬁnger movements to explore them.
In awake/alert states when she wasn’t actively exploring her toys, she often oriented to
sounds around her by turning her head or visually examined her hand or the toys. Esther
often paused after her movements and smiled. She also could have a look of
determination on her face when she moved the toys vigorously around.

Results for Lucy

The percentage of intervals Lucy was observed in preferred state decreased after
Soundbeam was introduced in the ﬁrst BC phase. This relative decrease in preferred
state percentage was maintained for the remainder of the experiment in both the
subsequent noncontingent to movement (B) and ﬁnal contingent to movement (BC)
music phases. The individual measures of preferred state percentages for each of Lucy’s
sessions are presented in Figure 10. Descriptive statistics for each phase are presented in
Table 10. Calculation of the effect size magnitudes between the phases yielded an effect
size of 1.18 between phase A and phase BC, 0.0 between BC and B, and .33 between B

and the ﬁnal BC phase.

111

mm mm mm mm

3“

mcoﬁmom

om 5N mm mm gm 2 2

m—

 

 

 

 

 

 

 

 

 

 

o\co

o\oo_

o\oom

o\com

o\cov

o\oom

oboe

Rook.

c\oow

Room

$02

SIBAJOIUI arms parrojard jo aﬁeruaorod

Figure 10. Percentage of Preferred State Intervals for Lucy in all Sessions

112

Table 10

Descriptive Statistics for Lucy — Preferred State

 

 

A BC B BC
Mean Percentage of 45% 29% 29% 25%
Preferred State Intervals
Standard Deviation 14.57 13.48 12.06 12.43

 

113

When Lucy was coded in preferred state she often had a non-animated facial
expression and appeared to focus intently on her surroundings. The investigator noted
that longer sequences of preferred state (more than three consecutive intervals) tended to
be characterized by a passive body posture. During these sequences her hands lay quietly
in her lap and her body appeared almost limp as she slumped slightly back in her chair.
As discussed earlier, a very small percentage of the aggregated preferred state category
consisted of interact state or contact with her surroundings and these interact patterns
consisted of bringing an object to her mouth to chew. Preferred state was a relatively
small percentage of her total time in all phases as she displayed almost constant hand
stereotypies.

Discussion

The ﬁnding that Esther exhibited little change in percentage of time in preferred
state may reﬂect the fact that because she already displayed over 95% of the time in these
states there may simply have not been much room for possible improvement. The only
sessions where she displayed a lower percentage of alert state were the two sessions
where she was ill. This ﬁnding seems consistent with the Guess et al. (1993) ﬁndings
that proﬁles for individuals tend to remain stable over time.

Lucy’s decrease in preferred state after the introduction of Soundbeam raises
several questions of how this ﬁnding should be interpreted. Lucy began the experiment
with a non-optimum proﬁle and the introduction of Soundbeam further increased the
incidence of intervals of stereotypy, a non-preferred state. Reducing hand stereotypies
and increasing functional hand use is a stated goal of interventions for children with Rett
Syndrome (Sullivan, Laverick & Lewis, 1995) so this intervention may be

contraindicated for children with Rett Syndrome

114

However, Lucy’s hand stereotypies may not interfere with her ability to learn
from other movements she is capable of making. As hand stereotypies invariably occur
with Rett Syndrome, the etiology of these stereotypes is generally regarded as related
directly to the compromised nervous systems of these children (Evans & Meyer, 1999).
An organic basis for this etiology contrasts with other hypothesized reasons for the
expression of stereotypies in SMD individuals such as compensation for low sensory
input or avoidance of particular situations (Guess & Carr, 1991; Kennedy, Meyer, &
Knowles, 2000).

It may be important to shift the focus of the intervention to the encouragement of
voluntary movements that can be controlled by the child rather than trying to extinguish
the almost constant stereotypical movements. For example, a previous study
demonstrated that a child with Rett Syndrome could learn to control her head movements
(Sullivan, Laverick, & Lewis, 1995), even though her hand stereotypies did not decrease.

Children with conditions such as Rett Syndrome that invariably lead to
stereotypies may be less distracted or hampered by these stereotypies as they attempt to
interact with their social and non-social environments. This may indicate that in some
situations, a preferred learning state for some children with particular conditions might
include the presence of stereotypies. In the current study, the Soundbeam was aimed at
Lucy’s legs as she appeared to have more control of her leg movements.

The decrease in Lucy’s preferred state may also be the result of a methodological
limitation of the study. As very few toys were available for Lucy, she had little
opportunity to interact with objects. Perhaps had more toys been made available, and
movement of these toys provided the contingent music, the stereotypies might have

decreased if she used her hands to operate the Soundbeam rather than her legs.

115

Happiness Indices

Research Question 2 investigated possible differences in the percentage of
“happiness intervals” that were observed in the participants across the four phases of the
experiment. The construct of happiness was operationalized by behavioral indications of
happiness that are seen in typically developing individuals such as smiles or
vocalizations.

When a smile or vocalization was observed in a preferred state interval it was
coded as a “happiness interval.” The mean percentage of happiness intervals for both
participants in the four experimental phases is presented in Figure 11. The actual
number of happiness intervals and preferred state intervals observed in the four phases is
presented in Table 11.

Reliability of Observations

Interobserver reliability for this measure was assessed by calculating three
measures of interval to interval agreement between the investigator and the research
assistant; overall percentage of agreement, percentage of occurrence reliability, and
percentage of non-occurrence agreement (Foster & Cone, 1986). The overall percentage
of agreement for Esther was 100%, the percentage of occurrence reliability was 98% and
the percentage of non-occurrence reliability was 99%. Calculation of all three measures
for Lucy yielded a value of 100% for each.

These values indicate a high rate of agreement and are similar to those found in
previous studies (Green & Reid, 1996, 1999a, 1999b) and lend conﬁdence to the results

gained from analysis of the measures.

116

1 00%

 

90%l- e W e *‘k ——-—. r—w —————- ___# -—--— - -
80% --—-—~~ A ~- # —-— —— * i~- v- f, v---«-———- ~—
70%4 ~-- - ~-

60%-- — - f e e -—~~—---~———-e————-v~——~~--»——v—A-—

 

I Lucy
3 Esther

50%-- v e -- -- v i v -- ~4-

 

 

 

 

40%

 

30%

20% e

Mean Percentage of Happiness Intervals

10%

 

 

0%

 

Phase

Figure 11. Mean Percentage of Happiness Intervals for Both Participants

117

Table 11

Number of Happiness Intervals for Esther and Lucy

 

 

A BC B BC

Esther

Happiness 68 196 146 332

Preferred State 333 1762 694 1651

Total Intervals 343 1800 699 l 800
Lucy

Happiness 31 182 63 13 1

Preferred State 317 515 196 366

Total Intervals 701 1799 668 1440

 

118

Results for Esther

Esther’s percentages of happiness interval were generally low across all phases
and there is a slight decrease after the introduction of Soundbeam music contingent to
movement. The individual measures of happiness interval percentages for each of her
sessions are presented in Figure 12. Descriptive statistics for each phase are presented in
Table 12. Calculation of the effect size yielded an effect size of 1.17 between phase A
and BC, 1.38 between BC and B, and .14 between B and the ﬁnal BC phase.

This decrease in happiness indices seems to indicate that Esther did not ﬁnd the
contingent music an enjoyable experience during the ﬁrst phase when it was ﬁrst
introduced. The happiness index coded for her was a wide smile that often lingered on
her face for a few seconds. The investigator noted that Esther generally did not smile
while playing with her toys. Her smiles were often observed after she paused in her play
or when she heard the voices of her family in the other rooms of the home.

Results for Lucy

Lucy’s percentages of happiness intervals increased after the introduction of
Soundbeam music contingent to music. This increase was maintained across the
subsequent noncontingent and return to contingency phases of the experiment. The
individual measures of happiness interval percentages for each of her sessions are
presented in Figure 13. Descriptive statistics for each phase are presented in Table 13.
Mean effect size calculations yielded an effect size of 1.4 between phase A and BC, .09
between BC and B, and.32 between phase B and BC.

The happiness index coded for her was a happy vocalization that accompanied a
smile. Lucy’s mother described her happy vocalizations as “buzzy” sounds that Lucy

made only when she was excited or very happy. Lucy generally made these sounds

119

mm mm mm mm Tm on mm mm

b

P _ .l P p

Fir

_

_

p

_L

b

meowmmom

mmaaim—m—Zo n m

» F

 

 

 

 

 

 

0m

 

 

 

o\oo

$3

$2..

gem

30v

.xcom

Axeoc

$3

$3

Axcoo

$03

spa/trawl sseutddeH jo aﬁeruoorad

Figure 12. Percentage of Happiness Intervals for Esther in all Sessions

120

Table 12

Descriptive Statistics for Esther - Happiness Intervals

 

 

A BC B BC
Mean Percentage of 20% 1 1% 21% 19%
Happiness Intervals
Standard Deviation 15.42 6.26 9.86 14.89

 

121

am mm mm mm 3. am mm mm

_ p

_ _ _

__r_rr H__

meoamom

mm a a C

2

2

 

 

 

 

 

 

 

 

 

 

$0

$2

Xom

.xcom

.xcov

c\oom

goo

Koch

ﬁcow

Xoa

o\ooo~

S[BA.191UI ssourddeH jo oﬁeruoorod

122

Figure 13. Percentage of Happiness Intervals for Lucy in all Sessions

Table 13

Descriptive Statistics for Lucy — Happiness Intervals

 

 

A BC B BC
Mean Percentage of 11% 39% 37% 45%
Happiness Intervals
Standard Deviation 7.06 22.58 24.11 25.00

 

123

between intervals of stereotypy when she would pull her hands down, wiggle her feet and
vocahze.

The percentage of happiness intervals for session 22 is based on a partial session
of 69 intervals. The noncontingent condition was stopped when Lucy became upset and
control of the Soundbeam was returned to her and the remaining intervals were dropped
from analysis. However, she did not display any signs of non-assent for the remainder of
the ﬁve B sessions so those were presented as planned.

Discussion

There could be several reasons for the ﬁnding that Esther’s percentage of
happiness intervals decreased with the ﬁrst introduction of the contingent music provided
by Soundbeam. It is possible that she found the addition of another stimulus while she
was playing with her toys annoying. The distraction of the musical feedback to her
movements could have meant that she received less pleasure from her play and therefore
smiled less.

A second possibility could be that Esther did gain some understanding of the
contingency (that her movements caused the music) but did not enjoy having to produce
the music. In this case she might have enjoyed or had neutral feelings about the music as
long as she didn’t have to produce it. An understanding of the contingency might be the
reason for the increase in happiness intervals noted after the contingency was withdrawn
in the B phase. However, the decrease in happiness intervals was not replicated in the last
BC phase.

The ﬁnding that Esther’s percentage of happiness indices decreased may indicate
that Soundbeam would not be not considered a leisure activity for her. The possible

reduction of happiness intervals observed in Esther means that provision of the

124

Soundbeam may not be an enjoyable activity for all SMD children. Children with SMD
who are not accustomed to having control of their environments may have come to prefer
it that way; therefore, Soundbeam may not be considered a leisure activity but rather a
work activity. In other words, the child cannot simply enjoy the music with Soundbeam,
he/she has to put forth effort to make it happen. It may also be important to be cautious
about interrupting a child who is concentrating on or enjoying manipulating materials by
adding another sensory stimulation. It may be more pleasurable and productive for them
to concentrate on one activity at a time.

The happiness protocol may be helpful to distinguish leisure and non-leisure
activities from the child’s perspective. Some activities such as learning to control
switches or manipulate an object may require the student to focus her concentration on
the activity which may not be accompanied by a smile. In this study, Esther often had a
look of determined concentration on her face as she moved her toys around and activated
the Soundbeam. Even with her lack of apparent enjoyment of the activity, learning the
contingency may ultimately be a valuable use of her time if she gains experience and
comfort with control.

The decrease noted could also be a result of a methodological limitation in the
experimental design. The very short and unstable baseline limits interpretation of the
change because of the wide variance of the three measures.

In contrast, Lucy did appear to enjoy the addition of Soundbeam music to her
leisure activity. The investigator had also noted that Lucy seemed happier in the BC
phase. Lucy’s mother told the investigator after several sessions of the Soundbeam, that

Lucy just “brightened up” when Soundbeam was being set up for her.

125

Because her percentage of happiness indices did not decrease on average when
the contingency was removed in the B phase, there is no evidence that she learned or did
not learn the contingency component. She did have one adverse reaction to the removal
of the contingency in the ﬁrst B session when the Soundbeam emitter was not placed near
her. After about eight minutes of the investigator producing the music, Lucy began to
make loud, whining vocalizations and her eyes ﬁlled with tears. Her mother said that
Lucy was very upset.

In accordance with the assent procedure of the study, the investigator returned
control of the Soundbeam to Lucy by placing the emitter near her. Lucy was content for
the remainder of the session, and did not become upset during the subsequent sessions
without Soundbeam control. The mean for percentage of happiness intervals did not
decrease in the B (noncontingent phase), so there is no evidence that the contingency
component was necessary for her to enjoy the music.

Voluntary Movement

The ﬁnal research question investigated possible differences in amounts of
voluntary movements across the four experimental phases. The Soundbeam emitter was
aimed to insure that movements the participants were capable of controlling would result
in musical feedback. The emitter beam was aimed at Esther’s right arm and hand so that
her movements with her toys activated the instrument. The emitter beam was aimed at
the lower half of Lucy’s body so that her leg movements activated the instrument.

The participants had different reactions to the introduction of the contingent and
noncontingent music. The mean percentage of movement intervals for both participants

in the four experimental phases is presented in Figure 14. The actual number of

126

1 00%

 

90% 777—77 77- A 7*777777 7

80% ~ 7 77 7 77 7 i77777i7 7 7 7 7
70%
60% 7
50% ILucy
ElEsther
40%
30% 7

20% 7

Mean Percentage of Movement Intervals

10% 7

 

 

 

0%

 

Figure 14. Mean Percentage of Movement Intervals for Both Participants

127

movement intervals and preferred state intervals observed in the four phases is presented
in Table 14.
Reliability of Observations

Any preferred state interval during which the participant displayed a voluntary
movement was coded as a “movement interval.” The same three measures of
interobserver agreement that were calculated for the previous observational codes were
calculated for movement. For Esther, the overall percentage agreement was 99%, the
percentage occurrence agreement was 99%, and the percentage of non-occurrence
reliability was 97%. For Lucy, the overall percent agreement was 99%, the percentage
occurrence agreement was 95% and the percent non-occurrence reliability was 98%.
These values indicate a high rate of agreement and lend conﬁdence to the results gained
from analysis of the measures.
Results for Esther

Esther maintained a similar average of movement intervals across all phases of
the experiment. The introduction of both contingent and noncontingent music of
Soundbeam did not affect her percentages of movement. The individual measures of
movement interval percentages for each of Esther’s sessions are presented in Figure 15.
Descriptive statistics for the means of the phases are presented in Table 15. Calculation
of the mean effect sizes yielded an effect size of .15 between phase A and BC, .81
between BC and B, and .28 between B and the ﬁnal BC phase.

Although Esther, who has cerebral palsy, seemed to require a lot of energy to
move her arm, she did display a variety of purposeful movements across all phases. She

was observed to shake, bat, grasp the toys, as well as explore them with her ﬁngers.

128

Table 14

Number of Movement Intervals for Esther and Lucy

 

 

A BC B BC

Esther

Movement 252 1300 433 1 1 15

Preferred State 3 33 1762 694 165 1

Total Intervals 343 1 800 699 l 800
Lucy

Movement 120 286 l 12 190

Preferred State 317 515 196 366

Total Intervals 701 1799 668 1440

 

129

mcoamom

mm mm mm mm 3” am mm mm mm a a

~_~bLFFr_~HF_bL_»_b__

C

2 2

Z

r H

 

 

 

 

 

 

 

 

o\oo

o\co_

o\com

o\oom

{cow

o\oom

face

{con

gem

o\ooa

o\coo ﬂ

SIEArorul ruaurvow jo aﬁerueorad

Figure 15. Percentage of Movement Intervals for Esther in all Sessions

130

Table 15

Descriptive Statistics for Esther -— Movement

 

 

A BC B BC
Mean Percentage of 76% 74% 62% 66%
Movement Intervals
Standard Deviation 4.87 13.89 16.71 13.32

 

131

Although the Soundbeam did not appear to increase Esther’s amount of
movement after it was introduced, her movements that activated Soundbeam in the
contingent phases brought an enthusiastic reaction from her mother. Her mother was not
in the same room with Esther and the investigator but could hear the Soundbeam through
the living room door as she went about her household tasks. When Esther was returned
to the living room her mother often commented on the session. She made comments such
as “I heard that pretty music you made, it was even better than yesterday” or “Wow, you
were really going to town in there.” Esther smiled at the enthusiasm in her mother’s
voice.

Several researchers have written that one of the most important effects of
technologies such as Soundbeam that allow children to cause an effect may be the
opportunity for the child’s caretaker to see the child as a competent individual who can
make something happen all by herself (Dunst, et al.,l985; Sullivan & Lewis, 1993).
Esther’s mother assumed that Esther knew she was making the music, whether or not this
was truly the case. The music production gave her mother a chance to comment on her
daughter’s performance and praise her “good job.” The long term effects of seeing the
child as one who understands the contingency may help make it a self fulﬁlling prophecy.
Results for Lucy

Lucy exhibited an increase in movement interval percentages after the
introduction of the Soundbeam. This increase was maintained even after the withdrawal
of the contingent component. The individual measures of movement interval percentages
for each of Lucy’s sessions are presented in Figure 16. Descriptive statistics for the

means of the phases are presented in Table 16. Calculations of the mean effect sizes

132

om mm mm mm

in om mm mm mm

_ H b p h h b F F b h p b p b p h

mqoamom

3. m—

S

2 mg

 

 

 

 

 

 

 

 

 

 

 

o\oo

$3

o\oom

Room

o\oov

$8

o\ooo

. o\oon

.xcow

nxcom

.xcooﬁ

spa/trawl ruowaAow JO aﬁeruoorsd

133

Figure 16. Percentage of Movement Intervals for Lucy in all Sessions

Table 16

Descriptive Statistics for Lucy— Movement

 

 

A BC B BC
Mean Percentage of 40% 57% 59% 61%
Movement Intervals
Standard Deviation 16.67 24.67 20.47 23.98

134

yielded an effect size of .73 between phases A and BC, .08 between BC and B, and .09
between B and BC.

Lucy’s movements were characterized by a variety of leg movements both in
tandem and single away from the footrest of her wheelchair. She was observed lifting
one leg above waist level and often extended both legs with a ﬂutter kick of her feet. Leg
movements were observed during both preferred intervals and those intervals
characterized by hand stereotypies. Lucy was able to generate several different types of
musical sequences because she was able to move her feet in different areas of the beam
and through a wide range of motion.

Lucy’s mother also commented on Lucy’s movements, often during the session
itself which she could observe from the kitchen or dining room. Her mother noted
speciﬁc movements by saying “What a pretty sound you made, I liked that a lot.” Her
mother also noted that Lucy only made certain movements during the experimental
sessions. As the investigator was setting up the Soundbeam for one session, Lucy lifted
one foot above her head in a graceful up and down sweeping motion. Her mother said,
“See, she never makes that movement unless you (investigator) are here.”

Discussion

Examination of Esther’s graph of percentage of movement intervals revealed both
that her average amount of movement intervals remained unchanged during the entire
experiment and that there was extreme variability in these percentages from session to
session. This variability suggests that there were other factors more strongly associated
with Esther’s ability and/ or willingness to move her arm than the presence or absence of

both the contingent and noncontingent music provided by Soundbeam.

135

One factor that appeared to be strongly associated with percentage of movement
was Esther’s health. It was the investigator’s impression that the amount of movement
that Esther was able to make on any given day depended on her breathing abilities. She
was oﬁen observed to stop and rest between movement sequences to catch her breath.
When her breathing ability seemed particularly compromised on a given day by extra
congestion, each time she moved her arm, she would begin to cough. After each
coughing spell she seemed to need to rest and recover.

Although her physical ability to sustain her movements varied, the fact that there
was no change in mean levels of movement interval percentages across all phases of the
experiment may indicate that, similar to the behavior state organization, Esther had
already reached the ceiling level for this measure before the experiment began. That is,
Esther participated in her leisure activity, playing with the toys, already as much as she
was physically able to do on any given day. Her motivation to interact with her
enviromnent was already in place and it was not changed by the addition of either
contingent or noncontingent musical feedback to her movements.

Lucy did show some change in the percentage of movement intervals as they
increased after the ﬁrst introduction of the Soundbeam. The change came at the same
point in the experiment that changes in the other two variables occurred, after the
introduction of the intervention in the ﬁrst BC phase with no change after that in the
subsequent B phase.

The ﬁnding that all three measures changed only after the ﬁrst introduction of
contingent music seems to reﬂect Lucy’s understanding that the Soundbeam provided a
different experience than the auditory soundtrack of her videos and that she enjoyed the

experience. However, the fact that removal of the contingency component in the

136

subsequent B phase seems to indicate that Lucy either did not learn the contingency, or
did learn it and accepted the fact that it was removed as there was no reduction in her leg
movements that had originally activated it in the ﬁrst BC phase.

Conclusions

The three research questions asked in this study looked at the effect of a speciﬁc
movement sensor, Soundbeam, on quality of life factors in two young girls with SMD.
The questions all concerned whether there a difference in the level of measures taken of
these quality of life factors between any of the experimental phases of the experiment.
The answer to these questions seems to be a qualiﬁed yes, a difference was detected on
one measure for one participant and all three measures for the second participant.
However the differences detected were small and only occurred at the ﬁrst introduction
of the intervention for both participants.

One conclusion from the results was that Soundbeam and similar interventions
that provide contingent stimulation to the individual’s actions do not affect children with
SMD in a “one size ﬁts all” way. The effects of the intervention may be dependent on
the behavior state proﬁle, a factor that has not generally been used when describing
participants in a study.

The effect of Soundbeam seemed to be almost insigniﬁcant for the ﬁrst
participant, Esther. The only difference on the three measures was that she seemed less
happy when given contingent control of the music in the ﬁrst phase. Soundbeam should
probably not be considered a leisure activity for this participant. The use of Soundbeam
did not affect the other two quality of life measures, probably because this participant
already maintained an optimum state proﬁle and already used her movement abilities as

much as she could.

137

The second participant, Lucy, did appear to show some effects but it was unclear
which component of the intervention, the contingency or the music itself caused the
changes in her measures. In Lucy’s case, Soundbeam could be used as a leisure activity
with the caveat that this instrument increased the percentage of time she spent in the non-
preferred state of stereotypy. Using Soundbeam to decrease the incidence of stereotypies
was not successful in this study.

A second conclusion is that the use of multiple measures of quality of life
provided more insights about possible interpretations of the participants’ reactions to the
intervention than measures based on developmental or functional frameworks. Although
only small differences were found and differences were not found for all factors, the
measures proved sensitive enough to detect some differences over a relatively short
period of time. One participant showed a change in affect after the introduction of the
intervention. The change in all three measures for the second participant, strengthened
the conclusion that she realized that there was something different about the contingent
and noncontingent music conditions as compared to her preferred leisure activity alone.

A third conclusion is that even though Soundbeam may not directly affect all of
the child’s quality of life factors in the short term, the effect of this intervention on others
close to the child may ultimately prove beneﬁcial. Children are part of a family and
social system. They may be affected by other’s perceptions of their competencies and
understanding of the environment. The observation that the parents of both participants
in this study assumed their child had attained a new competency and praised them for it
may have long term effects on the parent’s expectations of the child. A series of these

interactions may be the base for an expectation of competency that would expose the

138

child to more situations in which she is expected to exercise control and choice, a
foundational idea of quality of life.

A fourth conclusion is that this study did not provide conclusive evidence that
Soundbeam is more effective than other alternatives in improving the quality of life
factors measured for young children with SMD. However, Soundbeam proved to be a
sturdy and user-friendly piece of equipment. It took less than one hour for the
investigator to learn to assemble the components and change the settings of the control
unit. The Soundbeam equipment was set up and taken down over 70 times during the
experiment, with no equipment failures. The components could be conﬁgured in
different ways to ﬁt the space available for the children’s play sessions. The concept of
moving through a beam to change output from the instrument was easily understood and
readily observed by family members who watched the child interact with the instrument.
However, as obtaining a Soundbeam represents a considerable commitment of ﬁnancial
resources, practitioners may want to explore alternative methods of providing

contingency experiences for children with SMD.

139

Chapter 5
Implications, Limitations, Recommendations, and Summary

The availability of contingent stimulation provided by Soundbeam would be
described in a systems framework as an external or exogenous parameter manipulated by
the interventionist. As contingent stimulation is generally missing in the lives of children
with SMD its role as a possible control parameter in maintaining preferred states, as well
as the children’s reactions was investigated.

The three quality of life measures used in this study are instruments that proved to
be sensitive enough to detect changes in the participants’ responses to this contingency
response intervention over relatively short time period. The use of developmental scales
such as the Bayley developmental scales or adaptive behavior instruments such as the
Vineland used in some earlier studies of contingent stimulation interventions with
individuals with SMD did not detect changes in the participants’ behavioral repertoires
even after the interventions had been provided for several months (Howell, 1992;
Landesman-Dwyer & Sackett, 1979). The use of assessment within a quality of life
framework demonstrated that some aspects of the behaviors of children with SMD can
change rather quickly. The use of the quality of life assessments can help determine if
the interventions should be continued or should be withdrawn. For example, in this
study, the fact that Esther did not seem to enjoy the contingent music when it was ﬁrst
introduced indicates that it should not be considered a leisure activity for her.

As was the case in this study, individuals with SMD have often been found to
exhibit highly variable responses to the same activity from day to day and also have been

found to respond differently to the same intervention (Lancioni et al., 2002; O’Brien, et

140

al., 1994; Realon, et al., 1988). However, several implications can be drawn from the

effect of this intervention on each of the three factors of quality life that were measured.
Implications

Behavior State

The ﬁnding that the participants’overall quality of behavior state did not change
after the introduction of contingency response reinforces earlier ﬁndings that the quality
of state proﬁles of individuals with SMD is very difﬁcult to change with environmental
intervention alone and seem to stabilize around age three (Guess, Roberts, & Rues,
2002). The possible effectiveness of contingency response interventions to improve
proﬁles might be maximally effective when implemented with younger children and
infants. This ﬁnding underscores the possible importance of providing contingent
environments to infants with SMD (Guess, 2000).

Teachers may want to plan extra activities for children who, similar to Esther in
this study, already exhibit optimum proﬁles. These children may not need the rest periods
or “down time” that are often part of classroom daily schedules (Roberts, 1992). During
the down times which occur due to other staff responsibilities, providing interactive and
contingent response activities can help optimize the child’s time.

Happiness Indices

The fact that one participant (Esther) did not ﬁnd the ﬁrst introduction of
contingency response to be enjoyable conﬁrms earlier ﬁndings that caretakers and others
may not know what activities may be preferred by an individual with SMD (Bannerman,
1990; Everson & Reid, 1997; Favell, et al., 1996). The investigator assumed that the
provision of contingency response with music would be an enjoyable experience for

Esther but it did not increase her incidence of happiness intervals. Teachers and

141

caregivers may want to conduct a preference assessment (Leatherby, et al.,1992) before
providing leisure activities.

The fact that the Happiness Indices assessment was sensitive enough to detect
changes in affect in the 2 participants even after a short intervention period may make it
useful for teachers who wish to plan both leisure and work activities for their students.
Teachers may use the observational protocol to insure that leisure activities they have
chosen for their students are, in fact, activities that are enjoyed or preferred by the
student.

Voluntary Movement

The investigator’s observation that the parents of both participants often
commented on their child’s movements implies that teachers and caretakers should
seriously consider setting up play areas in which the child’s movements cause
“something to happen.” If children are give access to contingent environments, it
increases the chance that their movements, whether intentional or not, will activate a
stimulus that invites a response from others. The resulting increased social interactions
may be just as valuable as the opportunities to experiment with cause and effect.

Although Soundbeam technology may not be commonly available, there are many
alternatives to providing opportunities for contingent responses. Different arrangements
of toys, switches, and hanging objects can be positioned near the child (Dunnet, 1999;
Kinsley & Langone, 1995; Lancioni, et al., 1998; Saunders, et al. 2001). For example, in
the current study, Lucy’s mother was impressed with her daughter’s varied foot
movements in the Soundbeam and decided to attach bells to Lucy’ shoes so that she

could continue to receive auditory feedback to her leg movements.

142

Limitations

The conclusions and implications drawn from this exploratory study must be
regarded with caution due to several limitations. Limitations in both the research design
and the instruments used are discussed next.

Research Design

The generalizability of single subject designs depends on logical inferences made
from repeated replications of the effect of the intervention across participants.
Unfortunately, no general inference can be made from this study as different results were
obtained for the 2 participants. In addition, the use of only 2 participants greatly
decreased the chance of determining a representative reaction to the intervention.

When large variances of measures within a phase are obtained, as was the case in
this study, there is concern about experimental control. There were several variables in
this study that could have been better controlled by the investigator.

One notable source of variability was the health of the participants. Children
with SMD can have widely ﬂuctuating levels of health from one day to the next (Ault, et
al., 1994). The investigator could have implemented more stringent criteria of the
participant’s daily health in order for the participant to complete a session. For example,
the nurse in the participants’ classroom could have been asked to assess the child’s health
on a predetermined scale which included such factors as body temperature, observations
on indications of sickness. A low rating for the day could have been cause for
rescheduling the session.

Another source of variability was the different antecedent events before the
experimental sessions. Two sessions during the ﬁrst BC phase were held on days when

school had been cancelled because of inclement weather. Week ﬁve of the ﬁrst BC phase

143

coincided with the children’s spring vacation from school. This type of variability could
have been controlled by scheduling the study during a period that didn’t include a school
vacation.

A third source of variability may have been changes in the preferred activity that
were offered across all the phases. Esther was given different toys to play with and
different videos were played for Lucy in order to avoid satiation with the same activity.
However, the rotation of these toys was not controlled so that the same rotation was
offered in each phase. The investigator could have offered the preferred activity for
equal amounts of time in each phase in order to insure that the possible differing levels of
attractiveness of these activities did not affect the participants’ reactions.

In addition to the existence of several possible confounding variables, other
limitations of the research design could have lead to incorrect ﬁndings. An alternative
explanation of the ﬁnding that Esther’s happiness indices decreased after the A phase
may simply be the result of the short and unstable baseline. A longer baseline might have
produced a mean more similar to the ﬁnal three phases as the third data point (possibly an
outlier) would have had less effect on the mean of a phase containing more than three
data points.

A second alternative explanation for Esther’s decrease in happiness may be that
she simply did not like the type of music produced. Although the music output was
selected by Esther’s music therapist as appropriate for her developmental level, a
preference assessment could have been conducted to insure that each participant enjoyed
this type of stimulus.

The ﬁnding that Lucy did not understand the contingency component of the

intervention (as her movements and smiles did not decrease when it was withdrawn) may

144

also have resulted from a limitation of the design. After her initial protest when the
investigator removed the emitter, Lucy may have simply accepted that it was the
investigator’s turn to make music. The music was still contingent on the investigator’s
movements and as such, Lucy may have understood that it was still enjoyable for her to
observe and listen to.

Instrumentation

The results of this study should also be interepreted with caution because of
possible concerns about the construct validity of the assessment instruments used.
Quantiﬁcation of quality of life is an emerging area of assessment and the use of these
assessments in studies such as this one may raise questions about how the constructs they
are designed to measure are operationalized.

A basic assumption of the ﬁrst quality of life measure, behavior state, is that only
two states, awake/alert and interact, are considered optimum for learning (Guess,
Roberts, & Guy, 1999). This assumption might be questioned in light of observations of
Lucy, a child with Rett Syndrome. As is characteristic of children with Rett Syndrome,
Lucy moved her hands almost continually in repetitive patterns. However, it did appear
that she discerned something was different about her environment when Soundbeam was
introduced as her leg movements which activated the instrument increased as did her
indices of happiness. Perhaps a preferred state for children whose stereotypies are
considered “organically based” (Evans & Meyer, 1999) may include the co-occurrence of
those stereotypies with other alert behaviors.

The second quality of life measure, indices of happiness, may also need further
examination of its construct validity. As the protocol requires the user to count smiles,

one must also think about other purposes of smiles. Smiles can serve other purposes than

145

to express happiness. Green and Reid (1996) discount the possibility that individuals
with SMD smile merely to conform to social norms. However, smiles can also be used
to initiate interactions with others. Both participants in this study seemed to use smiles
for the social purposes. A smile may be an effective method for SMD children to draw
positive attention and social interaction from others. The participants in this study might
have smiled at times, especially in the beginning of the study, to try and engage the
investigator in a social interaction rather than to express happiness. The necessity of the
investigator to remain in close proximity to the children in order to code behavior state
might have led to social smiling to confound this measure.

Finally, a concern could be expressed about the third quality of life measure, the
use of movement counts as an indicator of learning cause and effect relationships.
Movements of the participants in this study to alleviate physical discomfort or express
anger could all have been counted as movement intervals. These types of movements, of
course, would not reﬂect the participants’ understanding of the contingency. Also, the
participants could have learned the contingency but not consistently expressed this
understanding by increasing their movements. Satiation with the musical stimulation or
simple physical fatigue could have resulted in a reduction of movements would confound
these counts as a construct for understanding the concept of cause and effect.

Recommendations for Further Research

Further research on Soundbeam as an intervention to provide contingent
stimulation and further research on the use of quality of life measures would serve to
clarify and extend the ﬁndings of this study. The effectiveness of Soundbeam on quality

of life factors could be further explored in studies with participants who have different

146

characteristics, use Soundbeam in different ways, or make direct comparisons of
Soundbeam to other types of contingent stimulation interventions.

Providing Soundbeam experiences to infants with SMD may result in a much
larger change in quality of life assessments than was demonstrated in this study. Guess,
Roberts, and Rues (2002) report that behavior state proﬁles stabilize around three years
of age for children with SMD. The provision of the Soundbeam contingency experience
to children younger than 3 years may have more of an effect as a control parameter for
changing and improving poor behavior state proﬁles than children who receive the
contingency experience when they are older.

Soundbeam may also have a greater effect on improving state proﬁles of children
who exhibit different behavior state proﬁles than the participants in this study. Studies
that examine its effect on children with high incidences of drowse or daze states or
children who are alert but rarely interact with the environment would be informative.

A study that directly compares the effectiveness of Soundbeam to other types of
contingency experiences such as switch activated toys would provide teachers and
caregivers with speciﬁc data on the relative effectiveness of this device. Comparisons of
how quickly children with SMD become satiated with the output or how much instruction
is needed in order to experience contingent responses would provide important
information to teachers and caregivers who are considering obtaining these types of
devices.

As quality of life is emerging as an important framework in which to evaluate
interventions designed for individuals with SMD, further evaluation and use of the

quality of life assessments would be useful. Possible changes to reﬂect more accurate

147

construct validity, adaptation of the protocols for use by practitioners and the use of these
assessments to evaluate a variety of interventions would be useful.

Single subject studies on interventions for SMD individuals that report means,
standard deviations, and number of measures taken in each phase would be extremely
helpful as ranges of obtained effect sizes could be calculated and compared to other
studies of similar interventions. One reason for the tentativeness of the conclusions
reached in this study was the absence of literature on effect sizes for single subject
studies with SMD participants. Although meta-analyses of between-group studies of
interventions in the social sciences and medical ﬁelds provide a context in which to
evaluate meaningful effect sizes (Lipsey, 1990; McCartney & Rosenthal, 2000), the
absence of such information in the literature of interventions for SMD individuals leaves
no basis for comparison. In addition, studies that attempted to correlate obtained effect
sizes with other measures of improvement such as subjective impressions of caretakers or
objective measures such as muscle range of motion would be important to help determine
what the magnitude of “meaningful” effect are for this population.

Summary

A single subject multiple baseline design with was used to evaluate the effect of
providing contingent musical stimulation to voluntary movements on three quality of life
measures in 2 participants, aged 4 and 6 years, with severe multiple disabilities. The
intervention, provided by a movement sensor called Soundbeam, consisted of providing
music both contingent and noncontingent to the girls’ voluntary movements in addition to
access to their preferred leisure activity. All experimental sessions were conducted the

participants’ homes over a 13 week period.

148

Different results were obtained for the 2 participants. One participant exhibited a
slight decrease observed happiness indices when the contingent component of the
intervention was presented. The other measures of percentage of preferred state and level
of movement did not change.

The second participant demonstrated a decrease in percentage of preferred state
intervals and an increase in observed happiness indices and voluntary movements after
the introduction of the intervention. These changes were maintained for the remainder of
the experiment.

One conclusion was that the quality of life assessments used can effectively detect
change even over the relatively short term of this experiment. The construct validity of
these assessments may need further evaluation.

It was also concluded that further study is needed before Soundbeam can be
recommended as an effective intervention to improve the three quality of life factors
measured in young children with severe disability. Further research on Soundbeam with
SMD children of different ages and with different behavior state proﬁles was
recommended.

In addition, this study demonstrated the need for the development of a database
on obtained effect sizes in single subject research on interventions for individuals with
severe multiple disability. A comparison of effect sizes across studies would provide both
a context in which to assess progress and a method of synthesizing the effectiveness of

various interventions.

149

APPENDICES

150

Appendix A

Informed Consent

151

 

OFFICE 3r
RESEARCH
ETHICS AND
STANDARDS

University Committee on
Research Involving
Human Subjects

M=cmcan State University

202 Girls J-iall

EJSI Lansing, MI

48824

St LESS-2 i80

FAX SI DMZ-4503

Web. 2w.» msu eduruser/ucnns
E'Mdll rcnnS'Qisiedu

The Urchin!!! Stare UOWP'SIIY
i0£4 7: inst:.'z.t;cmi [livery/W

Excellence n Action
MSU s an diamante-action,
azum-wpoﬁuﬁy rsrituiron

MICHIGAN STATE

UNIVERSITY

February 3, 2004

TO: Gail DUMMER
132 IM Sports Circle
MSU

RE: IRB# 03.042 CATEGORY: FULL REVIEW

APPROVAL DATE: February 3, 2003
EXPIRATION DATEzJanuary 3. 2004

TITLE: THE EFFECT OF CONTINGENT AUDITORY STIMULATION OF
SELF-INITIATED MOVEMENT ON THREE QUALITY OF LIFE FACTORS IN
YOUNG CHILDREN WITH SEVERE MULTIPLE DISABILITIES

The University Committee on Research Involving Human Subjects' (UCRIHS) review of this
project is complete and I am pleased to advise that the rights and welfare of the human
subjects appear to be adequately protected and methods to obtain informed consent are
appropriate. Therefore, the UCRIHS approved this project.

RENEWALS: UCRIHS approval is valid until the expiration date listed above. Projects
continuing beyond this date must be renewed wrth the renewal form. A maximum of four such
expedited renewals are possuble. Investigators wishing to continue a project beyond that time
need to submit a S-year application for a complete review.

REVISIONS: UCRIHS must review any changes in procedures involving human subjects. prior
to initiation of the change. If this is done at the time of renewal, please include a revrsion form
with the renewal. To revise an approved protocol at any other time during the year. send your
written request with an attached revrsion cover sheet to the UCRIHS Chair, requesting revised
approval and referencrng the preject's IRB# and title. Include in your request a description of
the change and any revised instruments, consent forms or advertisements that are applicable.
PROBLEMS/CHANGES: Should either of the followrng arise during the course of the work.
notify UCRIHS promptly: 1) problems (unexpected side effects. complaints. etc.) involving
human subjects or 2) changes in the research environment or new information indicating

greater risk to the human subjects than existed when the protocol was previously reviewed and
approved.

If we can be of further assistance. please contact us at (517) 35‘12180 or via email:
UCRIHS@msu.edu. Please note that all UCRIHS forms are located on the web:
http://wwwmsu.edu/userlucrihs

Sincerely.

'W
%Z [Xi/VJ V ﬂ
5hr

Kumar, MD.
UCRIHS Chair

AK; jm

cc: Cathy McQuillan
4113 Kenya
Jackson. MI 49201

152

Consent Form for Participation in the Study

The Effect of Contingent Auditory Stimulation on Three Quality of Life Measures in
Young Children with Severe Multiple Disabilities

Dear Parent or Guardian:

My name is Cathy McQuillan and I am a doctoral candidate in special education
at Michigan State University. I will be conducting a research study for my dissertation
that involves young children with severe multiple disabilities. The purpose of this letter
is to describe the study and invite your child to participate.

What is the purpose of the study?

One reason that children with severe disabilities may have difﬁculty learning is
that it is hard for them to play with toys or participate in other learning activities by
themselves. Several computer technologies have been developed that may make it easier
for children with severe disabilities to “make something happen” by themselves. For
example, you may be familiar with toys or devices that can be started by the child when
he or she presses a switch.

The purpose of this study is to see whether a type of technology called
Soundbeam may help children with severe disabilities stay alert, help them learn what
happens when they make movements, and provide them with a fun way to play by
themselves. Soundbeam is a musical instrument that is activated by movement only. It
has an ultrasonic sensor that is connected to an electronic keyboard. Any movement in
the sensor beam causes the keyboard to play. Some pictures of children using the
Soundbeam equipment are on the last page of this letter.

Soundbeam is being used in many schools and learning centers in over 20
countries. There has been some research done about how effective this type of
equipment is and I would like to add to that research.

How much time will the study take?

A child who participates in the study will be visited by me three times a week at
home for 13 weeks in a row for a total of 39 visits. During these visits, I will bring the
Soundbeam equipment, set it up and observe and videotape your child playing for 20
minutes. The total time of each visit would be about 30-40 minutes.

I would arrange a time to visit your home after school or in the evening that
would be mutually agreeable to you and me. I would visit at the same time and on the
same days of the week (for example at 4:30 on Monday, Wednesday and Friday.)

153

What happens dung a visit?

I would ask you to choose a place in your home where I could set up equipment
for the visits. The place should be a quiet area where the child’s activity wouldn’t
interfere with normal family routines. You may observe all of the sessions if you wish.

During each visit the following things will happen. I will ﬁrst set up the
Soundbeam, your child’s toys and the video camera in the area of the home you have
selected. Then you and I will position the child in his/her wheel chair or seating system.
The arrangement of equipment and observers will look like this.

 

 

‘ Soundbeam emitter © Participant

' Music source . Observer

A Video camera

 

There will be three types of sessions. During the ﬁrst few sessions, we will give
your child his/her favorite toys to play with. During the later sessions, the child will still
have his favorite toys to play with, but will also have an opportunity to activate
Soundbeam with his/her movements or will have an opportunity to listen to some musical
output from the keyboard that does not depend on his/her movements.

During all of the sessions I will be watching your child and using a checklist to
note how alert he or she is. I will also bring a second person called my “research
assistant” to some of the sessions who will also observe your child and follow the same
checklist. Later, I’ll look at both checklists to see if our observations match. I will
videotape the sessions so that aﬁer I’ve complete all my visits, I can review the tapes to
see if the child seemed to enjoy the sessions and to record how much he/she moved
during each of the sessions. At the end of each session I will ask you some questions your
child’s previous activities and mood before my visit.

What kind of information is gained from this study?

I will compare how alert the child was, how much the child moved and how much
the child seemed to enjoy the activity across these three types of sessions. These
comparisons may tell me how useful a movement sensor such as Soundbeam is to help
children with severe disabilities learn and have ﬁm.

Will children who participate in the study beneﬁt in any way?

Children with severe multiple disabilities often don’t have many chances to
control things around them or “make things happen” by their own movements. There
have been many reports by teachers and researchers that when children with severe
multiple disabilities have an opportunity to use their movements to make music with the
Soundbeam equipment, that they seem to learn better control of their movements and
oﬁen seem to be happy while they use it. Children in this study may experience similar
increases in motor control and enjoyment.

154

Are there any risks to children who participate in the study?

There are no physical risks to your child. The Soundbeam apparatus emits
ultrasound beams toward the location in which the child is sitting. Any interruption to
those beams caused by the child's movements is rewarded with contingent music
feedback. Soundbeam has been used in over 1000 schools for the past ten years and no ill
effects have been reported. Soundbeam was developed especially for individuals with
severe multiple disabilities and is available for purchase ﬁ'om adaptive equipment
catalogs.

In order to ensure that your child does not get uncomfortable in one position, we
will only position your child in an upright position in his/her wheelchair or seating
system for 20 minutes. With your permission, I will talk with your child’s physical
therapist to see if there are any concerns about positioning your child.

Your child may become disappointed when he/she no longer has a chance to play
with Soundbeam after the study is over. For that reason, I will place the Soundbeam at
your child’s school so that your child can continue to use it there.

What happens if the child doesn’t like the sessions?
If you or I feel that the child is not comfortable or does not want to continue the
session, we will stop the session for the day.

Does it cost any thing to participate in the study?
No, there are no costs to participate.

How will the information from the study be used and how will our family and child’s
privacy be protected?

I will use information from my observations and the videotapes to write my
dissertation, and possibly, to prepare published articles and/or educational presentations.
I will keep the videotapes and written information in my immediate possession, a locked
vehicle or in a locked ﬁle cabinet in my home ofﬁce. I will keep these materials in my
possession until all research publications and presentations have been completed and as
required by my university. Then, the videotapes and written information will be
destroyed.

Your child will be described in my dissertation and in any subsequent articles or
presentations but his/her real name, school and city of residence will not be revealed. I
will ask you to give me a pseudonym or “made up” name to use when describing your
child. The only people who know the identity of the children and location of the study
will be me, your child’s school principal, classroom teacher, physical therapist, the
research assistant, my university advisor (Dr. Gail Dummer), and members of my
dissertation committee (Drs. David Stewart, Cynthia Okolo, and Tim Stocks).

In order to accurately describe your child I will ask you to allow me to review
your child’s school record in order to give your child’s medical diagnosis, educational
testing results and current educational goals in my written descriptions. When this type
of description is provided in reports, it helps the readers (such as other teachers,
researchers and parents) determine if the results that I report might apply to the children
they are working with.

155

I would like to use short video clips of the children in the study to show in
educational presentations after my dissertation is complete. Video clips often help people
understand the information that is being presented. You do not have to allow me to use
the clips in order for your child to participate in the study. If you do agree to allow me to
use the clips for a presentation, I will ask you to review the clips and then sign a separate
consent form before each presentation. An example of this separate consent form is
attached to the end of this letter.

Your privacy and your child's privacy will be protected to the maximum extent
allowable by law.

What happens ifJarents choose not to have their child participate in this study?

Participation in this study is entirely voluntary. If you choose not to participate,
this does not affect your child’s educational opportunities and current school program at
all. Soundbeam will be placed at your child’s school after the study and perhaps your
child would like to try it then.

What happens if the parents and child start to participate in the study but then later
change their minds and don’t want to participategany more?

You may tell me at any time that you don’t want to continue in the study and we
will stop the visits. The information I’ve collected until then may or may not be used for
my dissertation, depending upon the number of sessions completed. If you decide to stop
participating, this will have no effect on your child’s current educational plan.

Who should parents contact if they have Questions, now or in the ﬁJture?

If you have any questions about this study now, or in the future you may contact me
(Cathy McQuillan) or my advisor (Gail Dummer):

Cathy McQuillan

Phone (517) 784-9287
Fax (517) 768-8148
E-mail mcquilll @msu.edu

Address 4113 Kenzie
Jackson, MI 49201

Dr. Gail M. Dummer

Phone (517) 355-4744

Fax (517) 353-2944
E-mail dummer@msu.edu
Address 132 [M Sports Circle
Department of Kinesiology

Michigan State University
East Lansing, MI 48824

156

If you have questions or concerns regarding your rights as a study participant, or are
dissatisﬁed at any time with any aspect of this study, you may contact Dr. Ashir Kumar,
anonymously if you wish:

Ashir Kumar, M.D.

Chair of the University Committee on Research Involving Human Subjects (U CRIHS)
Phone (517) 355-2180

Fax (517) 432-4503

E-mail ucrihs@msu.edu

Address 202 Olds Hall

East Lansing, MI 48824

What should parents do if they think they mi g1_1t like their child to participate in this
study?

 

I will be contacting you shortly to see if you are interested. If you think you
might be interested we will arrange a meeting. I will demonstrate Soundbeam for you,
we will talk about all of the information in this letter and I’ll answer any questions you
have. Then, if you would like to participate, I will obtain your signatures on the
following page of this letter.

Sincerely,

Cathy McQuillan
Doctoral Student
Department of CEPSE
College of Education
Michigan State University

157

Appendix B

Coding Protocols

158

Behavior State

(Guess, Roberts, & Rues, 2001; Guess, Roberts, Siegel-Causey, & Rues, 1995; Roberts,
1992)

Directions to Coder

Begin auditory tape. A countdown will begin to Code 1. At the cue, begin observing the
child for the ﬁve- second interval. During the ﬁve- second interval, it is helpful to tap out
5 beats, one for each second and note the observed state at each beat. For example, you
may observe sleep, sleep, sleep, sleep, drowse during one interval as a child begins to
change state. At the auditory cue “stop”, look down at your sheet and circle the state the
child displayed for majority of the interval. A quick note may also be jotted on the line
below the box. If you are totally unsure about the observed state or missed the
observation, make a diagonal slash through the box.

Look up and wait for the number cue to begin observing the next 5-second interval.
Please make sure that the number of the interval you are coding matches the auditory cue.

Attached are three coding sheets for the 20 minute observation period consisting of 120
5-second intervals.

Abbreviations on coding sheet

SL Sleep
DR Drowse
DA Daze

AW Awake-inactive-alert
INT Interact

ST Stereotypy

CA Crying, agitated

SI Self injury

SEZ Seizure, this is not considered a state but circle if one occurs during the interval

159

Child:
Date:

 

 

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

10.

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

SL DR
AW INT

ST CA SI

DA

SEZ

 

 

 

 

13.

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

 

 

 

 

ST CA SI SEZ

 

 

 

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

11.

 

SL DR
AW INT

ST CA SI

DA

SEZ

 

 

 

 

14.

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

160

 

 

Coder: Page # 1
Session #
2 3
SL DR DA SL DR DA
AW INT AW INT

 

ST CA SI SEZ

 

 

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

12.

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

15.

 

SL DR
AW INT

ST CA SI SEZ

DA

 

 

 

 

Happiness Indices
(Green & Reid, 1996)
Directions to Coder

Put videotape number on sheet. You will view a series of intervals where the participant
has been previously identiﬁed as in a preferred state. The number of the interval will
appear on a black screen immediately before each interval. Observe the entire 5-second
interval for any happiness index displayed by the participant. A happiness index is
deﬁned as any behavioral indication of positive affect that is typically seen in individuals
without disabilities. Examples are smiling and vocalizations that seem to express
excitement or pleasure.

At the end of each interval the screen will go dark for 5 seconds. During this time, mark
“yes” if at any time during the previous observed interval the child displayed at least one
happiness index, mark “no” if no happiness index was observed. Then look back at the
screen to observe the next interval.

Please make sure you mark the correct interval space. The total number of intervals will
vary from videotape to videotape.

161

Voluntary Movements

Directions to Coder

Put videotape number on sheet. You will view a series of intervals where the
participant has been previously identiﬁed as in a preferred state. The number of the
interval will appear on a black screen immediately before each interval. Observe the
entire 5-second interval for any voluntary movement made by the participant. A
voluntary movement is deﬁned differently for each participant. For Esther, a voluntary
movement is any movement of her right arm or hand. For Lucy, a voluntary movement is
any movement of her legs away from the footrest of her wheelchair.

At the end of each interval the screen will go dark for 5 seconds. During this
time, mark “yes” if at any time during the previous observed interval the child displayed
at least one voluntary movement. Mark “no if no voluntary movement was observed.
Then look back at the screen to observe the next interval.

Please make sure you mark the correct interval space. The total number of intervals will
vary from videotape to videotape.

162

Participant Coder

 

 

Videotape #
Total intervals coded

 

Interval Happiness Index Targeted Movement
1 . Yes No Yes No
2. Yes No ‘ Yes No
3. Yes No Yes No
4. Yes No Yes No
5. Yes No Yes No
6. Yes No Yes No
7. Yes No Yes No
8. Yes No Yes No
9. Yes No Yes No
1 0, Yes No Yes No
1 1. Yes No Yes No
12 Yes No Yes No
l 3. Yes No Yes No
14, Yes No Yes No
1 5. Yes No Yes No
16. Yes No Yes No
1 7. Yes No Yes No
1 8. Yes No Yes No
19. Yes No Yes No
20. Yes No Yes No

163

Appendix C

Individual Protocols of Behavior State

164

Individual Behavior State Protocol

Participant Esther

Sleep
Teacher reports she never sleeps during the day unless ill, breathing sounds may be

“rattley ” and steady

Drowse
May be diﬂicult to tell if her glasses are on, her head will be back against head rest

Daze
Relaxed body, eyes ﬁxed, head not moving, sometimes appears “cross-eyed sometimes
appears to be looking up, sometimes hand can be “hooked " on toy

Awake/alert/inactive
Smiles, moves head, arches backward in chair, moving arms, swallowing, yawning,
cooing, sounds

Interact
Touches toys- her hand or ﬁngers must be moving or pushing against the hanging toy,
touching her own arm, fingering her clothing

Stereogypy

Not observed

Cﬂinglagitated

“Whiney vocalizations ” for more than 3 counts, active grimacing

Self-iniugy
Not observed

Seizure
Mother says Esther '5 body will suddenly stiffen, she may cry out, then does not seem
aware of surroundings at all

165

Individual Behavior State Protocol

Participant Lucy

Sleep
A very deep sleeper, her body will be totally relaxed, little movement

Drowse
Her head may be down on the armrest, may be lightly ﬁngering her chin, eyes open and
close very slowly

Daze
Not looking at the television screen or observers, glassy eyes, hands quiet in her lap

 

Awake/alert/inactive
No stereotypies, she often looks side to side, happy “buzzy " sounds, moving her legs,
sucking on her pacifier, “whiney ” vocalizations that don ’t last more than 3 counts

Interact
Grabbing at her bib before it comes to her mouth, scooping up her pacifier, rubbing her
face or hair, ﬁngering toy in her lap

Stereogypy

Hands poking at her mouth or chin, chewing her ﬁnger or leaving ﬁnger in her mouth
and sucking, touching her tongue, chewing on her bib or wheel chair armrest

C in a itated
“whiney ” vocalizations, loud cries, tears, banging her feet on footrest, throwing herself
backward in wheelchair, wrinkles up face for more than 3 counts

Self-iniugy
Not observed

Seizure
Mother reports that Lucy has not had any seizures for over a year. Mom will monitor
sessions for seizure activity.

166

Appendix D

Sample Session Log Sheets

167

Participant Esther Session 16
Date 4/8/03 (Tuesday)

Phase A B BC B

Co-rater Yes No

Normal routine
Before session
Tube feeding at school at 11:30
PM school activities (story, video, music, circle time)
Toileting at 2:30 at school
Bus ride- Home at 3:10
Mother removes her AF Os
Wears Hensinger collar
Starts session as soon as she gets off bus at home
During session
Hanging toys available on her right
Chair inclination at 110 degrees
Wears seatbelt, butterly harness; foot straps off
Soundbeam position and settings (when activated)
Sound source on her right, MHDI utilities: 11, 3/4 volume setting

Deviations from this routine
none

Conditions-
Li ght- average
Sound- no street sounds, could hear brother crying in other room for about 4
seconds
Temperature- felt a little cooler than usual, space heater was on

Health notes

A few coughs, breathing sound “ rattley” and loud but no meds needed at
school

Esther’s indications of assent or non-assent during this session
Cooing, smile, no indications of physical discomfort

Session notes, parent comments
Often arches to left when reaching for toys. Often pauses, smiles during the
silence. Is she thinking about what she did? Smiles don ’t occur with movement
but after the music stops. It might be too much for her to smile and move at the
same time. She seems to get tired and need longer rest periods when her
breathing is this loud. Mom said “I heard that nice music you were playing "
when Esther was returned to living room- she smiled

168

Participant - Lucy Session 26
Date 4/29/03 (Tuesday)

Phase A B BC B

Co-rater Yes No

Normal routine
Lunch at school at 11:30
PM nap at school
Toileting at 2:30 at school
Bus ride; home at 3:20
Remove DAF Os and shoes
Watches videos until my arrival
Seat belt on, “chewy” bib, paciﬁer available on cord
Short chat with Mom while I set up equipment
Blues Clues video during session
Soundbeam position and settings (when activated)
Sound source on her left, MIDI utilities: 11, 3/4 volume setting

Deviations from this routine
Stayed home from school (pinkeye). Mom said that Lucy hadn 't slept for more
than 30 minutes (about an hour less than her usual nap at school. Lunch was at
12:00, diaper change at 2:00
Mom put on Lucy 's butterfly harness- didn ’t want her leaning too far forward and
banging her head (she was very active yesterday).

Conditions-
Li ght- Bright
Sound- My short conversation with mom about her paciﬁer during session. Phone
rang once.
Temperature- Average

Health notes

Pinkeye diagnosed this am at doctor. Lucy hasn 't slept well but does not seem
grumpy or unduly tired.

Lucy’s indications of assent or non-assent during this session
Smiles, buzzy sounds. No protesting.

Session notes, parent comments

More movement today. Mom commented on how Lucy was watching me.
”She likes the attention. " Reactivity/.7? Is she performing for me or enjoying

169

the music/ This probably wouldn 't be diﬁrerent across phases. One interaction
with her when I had to re-attach pacifier back over harness.

170

REFERENCES

Abramson, L. Y., Seligrnan, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in
humans: Critique and reformulation Journal of Abnormal Psychology, 87, 49-74.

Alberto, P. A., & Sharpton, W. (1988). Components of instructional technology. In

L. Stemberg (Ed.), Educating students with severe or profound handicaps
(pp. 185- 218). Rockville, MD: Aspen.

Allison, D. A. (1992). When cyclicity is a concern: A caveat regarding phase change
criteria in single-case designs. Comprehensive Mental Health Care, 2,
31-149.

Allison, D. A., Franklin, R. D., & Heshka, S. (1992) Reﬂections on visual inspection,
response guided experimentation, and type I error rate in single-case designs.
Journal of Experimental Education, 61, 45-51.

Arthur, M., Hook, J ., & Butterﬁeld, N. (1995). Behaviour state: Exploring best
practices for students with the most severe and multiple disabilities.
Australasian Journal of Special Education, 19, 37-44.

Ashton, R. (1973). The state variable in neonatal research: A review. Merrill-Palmer
Quarterly, 9. 3-20.

Ault, M. M., Guy, B., Guess, D., Bashinski, S., & Roberts, S. (1995). Analyzing behavior
state and learning environments: Application in instructional settings. Mental
Retardation, 33, 304-316.

Ault, M. M., Guy, B., Rues, J., Noto, L., & Guess, D. (1994). Some educational
implications for students with profound disabilities at risk for

inadequate nutrition and the nontherapeutic effects of medication. Mental
Retardation, 32, ZOO-205.

Ayres, B. J., Meyer, L. H., Erevelles, N., & Park-Lee, S. (1994). Easy for you to say:
Teacher perspectives on implementing most promising practices. Journal of the
Association for Persons with Severe Handicaps, 19, 84-93.

Baer, D. M. (1981). A hung jury and a Scottish verdict: “Not proven.” Analysis
and Intervention in Developmental Disabilities, 1, 91-97.

Bannerrnan, D. J., Sheldon, J. B., Sherman, J. A., & Harchik, A. E. (1990). Balancing the
right to habilitation with the right to personal liberties: The rights of people with
developmental disabilities to eat too many doughnuts and take a nap. Journal of
Applied Behavior Analysis, 23, 79-89.

Barlow, D. H., & Hersen, M. (1984). Single case experimental designs: Strategies for
studying behavior change (2nd edition). Boston, MA: Allyn and Bacon.

171

Bashinski, S. M. (1996). Interactive multimedia technology as an instructional tool
for teachers of students with profound or multiple disabilities. (Doctoral
dissertation, University of Kansas, 1996). Dissertation Abstracts
International, 5 7, 2979.

Basil, C. (1992). Social interaction and learned helplessness in severely disabled
children. AAC Augmentative and Alternative Communication, 8, 188-199.

Bass, R. F. (1987). The generality, analysis, and assessment of single-subject data.
Psychology in the Schools, 24, 97-104.

Baumgart, D., Brown, L., Pumpian, I., Nisbet, J., Ford, A., Sweet, et al. (1982). Principle
of partial participation and individualized adaptations in education programs for
severely handicapped students. Journal of the Association for Persons with
Severe Handicaps, 7, 17-27.

Becker, P. T., & Thoman, E. B. (1983). Organization of sleeping and waking states in
infants: Consistency across contexts. Physiology and Behavior, 31, 405-415.

Bedini, L. A. (1993). Technology and people with disabilities: Ethical considerations.
Palaestra, 9(4), 25-30.

Behrmann, M. M., & Lahm, L. (1982). Babies and robots: Technology to assist learning
of young multiply disabled children. Rehabilitation Literature, 45, 194-201.

Beminger, V. W. (1988). Development of operational thought without a normal
sensorimotor stage. Intelligence, 12, 219-230.

Bijou, S. W. (1981). Behavioral teaching of young handicapped children: Problems
of application and implementation. In S. W. Bijou & R. Ruiz (Eds), Behavior
modiﬁcation: Contributions to education (pp. 97-110). Hillsdale, NJ: Erlbaum.

Boismier, J. D. (1976). Visual stimulation and wake-sleep behavior in human neonates.
Developmental Psychobiology, 10, 219-227.

Brazelton, T. B. (1973). Neonatal behavior assessment scale. Philadelphia: Lippincott.

Brazelton, T. B. (1978). Introduction. In A.J. Sameroff (Ed.), Monographs of the society
for research in child development (pp. 1-13). Chicago: University of Chicago.

Bredekamp, S. (1993). The relationship between early childhood education and early
childhood special education: Healthy marriage or family feud? Topics in Early
Childhood Special Education, 13, 258-273.

Brimer, R. (1990). Students with severe disabilities: Current perspectives and practices.
Mountain View, CA: Mayﬁeld.

172

Brinker, R. P., & Lewis, M. (1982a). Contingency intervention. In J. Anderson
(Ed.), Curricula for high-risk and handicapped children (pp. 37-41). North
Carolina University, Chapel Hill: Technical Assistance Department.

Brinker, R. P., & Lewis, M. (1982b). Making the world work with microcomputers: A
learning prosthesis for handicapped infants. Exceptional Children, 49,
163-170.

Brotherson, M. J., Cook, C. C., Cunconan—Lahr, R., & Wehmeyer, M. L. (1995). Policy
supporting self-determination in the environments of children with disabilities.

Education and Training in Mental Retardation and Developmental Disabilities,
30, 3-14.

Browder, D. M. (1991). Assessment of individuals with severe disabilities. Baltimore:
Paul H. Brookes.

Brown, F ., & Cohen, S. (1996). Self-determination and young children. Journal of the
Association for Persons with Severe Handicaps, 21, 22-30.

Brown, R, Gothelf, C. R., Guess, D., & Lehr, D. H. (1998). Self-determination for
individuals with the most severe disabilities: Moving beyond chimera. Journal of
the Association for Persons with Severe Handicaps, 23, ”-26.

Brown, J. W., Leavitt, L. A., & Graham, F. K. (1976). Response to auditory stimuli in 6-
and 9-week-old human infants. Developmental Psychobiology, 10, 255-266.

Brown, L., Nietupski, J ., & Hamre-Nietupski, S. (1976). The criterion of ultimate
functioning. In M. A. Thomas (Ed.), Hey, don 't forget about me! Education ’s

investment in the severely, profoundly, multiply handicapped (pp. 2-15). Reston,
VA: The Council for Exceptional Children.

Burch, M. R., Clegg, J. C., & Bailey, J. S. (1987). Automated contingent reinforcement
of correct posture. Research in Developmental Disabilities, 8, 15-20.

Campbell, P. H. (1987). Physical management and handling procedures with students
with movement dysfunction. In M. E. Snell (Ed.), Systematic instruction of
persons with severe handicaps (pp. 174-187). Columbus, OH: Charles E. Merrill.

Campos, J. J ., & Brackbill, Y. (1973). Infant state: Relationship to heart rate, behavioral
response and response decrement. Developmental Psychobiology, 6, 9-19.

Carta, J. J., Schwartz, 1. S., Atwater, J. B., & McConnell, S. R. (1991). Developmentally

appropriate practice: Appraising its usefulness for young children with
disabilities. Topics in Early Childhood Education, I 1, 1-20.

173

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ:
Erlbaum.

Cohen, M. A., Gross, P. J ., & Haring, N. G. (1976). Developmental pinpoints. In N. G.
Haring & L. J. Brown, (Eds), Teaching the severely handicapped, Vol. I (pp. 35-
110). New York: Grune & Stratton.

Columbo, J ., & Horowitz, F. D. (1987). Behavioral state as a lead variable in neonatal
research. Merrill-Palmer Quarterly, 33, 423-437.

Columbo, J ., Moss, M., & Horowitz, F. D. (1989). Neonatal state proﬁles: Reliability
and short-terrn prediction of neurobehavioral status. Child Development, 60,
1102-1110.

Cornell, J ., & Mulrow, C. (1999). Meta-analysis. In H. J. Ader & G. J. Mellenbergh,
(Eds), Research methodology in the life, behavioral and social sciences, (pp.
285-323). London: SAGE.

Darius, P., & Portier, K. (1999). Research design. In H. J. Ader & G. J. Mellenbergh,
(Eds), Research methodology in the life, behavioral and social sciences, (pp. 67-
95). London: SAGE.

Davis, D. H., & Thoman, E. B. (1987). Behavioral states of premature infants:

Implications for neural and behavioral development. Developmental
Psychobiology, 20, 25-38.

Dewson, M. R. J ., & Whitely, J. H. (1987). Sensory reinforcement of head turning with
nonambulatory, profoundly mentally retarded persons. Research in
Developmental Disabilities, 8, 413-426.

Downing, J. E. (1996). Including students with severe and multiple disabilities in typical
classrooms: Practical strategies for teachers. Baltimore: Paul H. Brookes.

Dule, K., Komer, H., Williams, J., & Carter, M. (1999). Delivering therapy services for
students with high support needs: Perceptions of roles, priorities and best

practice. [Electronic version]. Journal of Intellectual and Developmental
Disability, 24, 243-263.

Dunnet, J. (1999). Use of activity boxes with young children who are blind, deaf-
Blind or have severe learning disabilities and visual impairments. Journal
of Visual Impairment and Blindness, 93, 225-232.

Dunst, C. J ., Cushing, P. J ., & Vance, S. D. (1985). Response-contingent learning in
profoundly handicapped infants: A social systems perspective. Analysis and

Intervention in Developmental Disabilities, 5, 33-47.

Eagle, R. S. (1985). Deprivation of early sensorimotor experience and cognition

174

in the severely involved cerebral-palsied child. Journal of Autism and
Developmental Disorders, 15, 269-283.

Edgington, E. S. (1983). Response-guided experimentation. Contemporary Psychology,
28, 64-65.

Ellis, P. (1995). Incidental music: A case study in the development of sound therapy.
British Journal of Music Education, 12, 59-70.

Ellis, P. (1996). Layered analysis: A video-based qualitative research tool to support
the development of a new approach for children with special needs. Bulletin
of the Council for Research in Music Education, 130, 65-74.

Ellis, P. (1997). The music of sound: A new approach for children with severe
and profound and multiple learning difﬁculties. British Journal of Music
Education, 14, 173-186.

Ennis, C. D. (1992). Reconceptualizing learning as a dynamical system. Journal of
Curriculum and Supervision, 7, 115-130.

Evans, I. M., & Meyer, L. H. (1999). Modifying adult interactional style as
positive behavioural intervention for a child with Rett syndrome. Journal of
Intellectual and Developmental Disability, 24, 191-205.

Everson, J. M., & Reid, D. H. (1997). Using person-centered planning to determine
employment preferences among people with the most severe developmental
disabilities. Journal of Vocational Rehabilitation, 9, 99-108.

Favell, J. E., Realon, R. E., & Sutton, K. A. (1996). Measuring and increasing the
happiness of people with profound mental retardation and physical handicaps.
Behavioral Interventions, 11(10), 47-58.

Featherstone, H. (1980). A difference in the family: Life with a disabled child. New
York: Basic Books.

Felce, D., & Perry, I. (1995). Quality of life: Its deﬁnition and measurement. Research
in Developmental Disabilities, 16, 51-74.

Ferguson, D. L. (1985). The ideal and the real: The working out of public policy in
curricula for severely handicapped students. Remedial and Special Education,
6, 52-60.

Ferguson, D. L., & Baumgart, D. (1991). Partial participation revisited. Journal of the
Association for Persons with Severe Handicaps, 16, 218-227.

Fields, S. L. (1999). The effects of technology use by an individual with severe or

175

multiple disabilities within the educational environment: A case study. (Doctoral
dissertation, University of Oklahoma, 1999). Dissertation Abstracts
International, 60, 1979.

Finkelstein, N. W., & Ramey, C. T. (1977). Learning to control the environment in
infancy. Child Development, 48, 806-819.

Fogel, A., & Thelen, E. (1987). Development of early expressive and communicative
action: Reinterpreting the evidence from a dynamic systems perspective.
Developmental Psychology, 23, 747-761.

Foster, S. L., & Cone, J. D. (1986). Design and use of direct observation. In A. R.
Ciminero, K. S. Calhoun, & H. B. Adams, (Eds), Handbook of behavioral
assessment, (pp. 253-324). New York: John Wiley & Sons.

Fox, L, Hanline, M. F ., Vail, C. 0., & Galant, K. R. (1994). Developmentally appropriate
practice: Applications for young children with disabilities. Journal of Early
Intervention, 18, 243-257.

Glenn, S. M., & Cunningham, C. C. (1984). Selective auditory preferences and the use of
automated equipment by severely, profoundly and multiply handicapped
children. Journal of Mental Deﬁciency Research, 28, 281-296.

Glickman, L., Deitz, J ., Anson, D., & Stewart, K. (1995). The effect of switch control
site on computer skills of infants and toddlers. The American Journal of
Occupational Therapy, 50, 545-553.

Goldenberg, E. P. (1979). Special technology for special children. Baltimore:
University Park Press.

Green, C. W., Gardner, S. M., & Reid, D. H. (1997). Increasing indices of happiness
among people with profound multiple disabilities: A program replication and
component analysis. Journal of Applied Behavior Analysis, 30, 217-228.

Green, C. W., & Reid, D. H. (1996). Deﬁning, validating, and increasing indices
of happiness among people with profound multiple disabilities.
Journal of Applied Behavior Analysis, 29, 67-78.

Green, C. W., & Reid, D. H. (1999a). A behavioral approach to identifying sources of
happiness and unhappiness among individuals with profound multiple disabilities
[Electronic version]. Behavior Modiﬁcation, 23, 280-294.

Green, C. W., & Reid, D. H. (1999b). Reducing indices of unhappiness among
individuals with profound multiple disabilities during therapeutic exercise

routines. Journal of Applied Behavior Analysis, 32, 137-147.

Guess, D. (2000). Serving persons with severe and profound disabilities: A work in

176

progress. In M. Wehrneyer & J. Patton (Eds), Mental retardation in the 215t
century (pp. 91-111). Austin, TX: Pro-Ed.

Guess, D., & Carr, E. (1991). Emergence and maintenance of stereotypy and self-injury.
American Journal on Mental Retardation, 96, 299-319.

Guess, D., Mulligan-Ault, M., Roberts, 8., Struth, J ., Siegel-Causey, E., Thompson, B., et
al. (1988). Implications of biobehavioral states for the education and treatment of
students with the most profoundly handicapping conditions. Journal of the
Association for Persons with Severe Handicaps, 13, 163-174.

Guess, D., Roberts, 8., Behrens, G. A., & Rues, J. (1998). Response to Mudford, Hogg,
and Roberts on continuous recording of behavior state. American Journal on
Mental Retardation, 103(1), 75-79.

Guess, D., Roberts, 8., & Guy, B. (1999). Implications of behavior state for the
assessment and education of students with profound disabilities. In A.
Repp & R. Homer (Eds), Functional analysis of problem behavior:
From effective assessment to effective support (pp. 338-3 94). Belmont, CA:
Wadsworth.

Guess, D., Roberts, 8., & Rues, J. (2000). Behavior states are real and observable.
American Journal on Mental Retardation, 105, 300-306.

Guess, D., Roberts, 8., & Rues, J. (2002). Longitudinal analysis of state patterns
and related variables among infants and children with signiﬁcant

disabilities. Research and Practice for Persons with Severe Disabilities, 27,
1 1 2- 1 24.

Guess, D., Roberts, 8., Siegel-Causey, E., Ault, M., Guy, 3., Thompson, B., et al. (1991).
Investigations into the state behaviors of students with severe
and profound handicapping conditions. Monograph. University of Kansas.

Guess, D., Roberts, 8., Siegel-Causey, E., Ault, M. M., Guy, B., Thompson, 3., et a1.
(1993). Analysis of behavior state conditions and associated variables among

students with profound handicaps. American Journal on Mental Retardation, 97,
634-653.

Guess, D., Roberts, 8., Siegel-Causey, E., & Rues, J. (1995). Replication and extended
analysis of behavior state, environmental events, and related variables among

individuals with profound disabilities. American Journal on Mental Retardation,
100, 36-5 1.

Guess, D., & Sailor, W. (1993). Chaos theory and the study of human behavior:

Implications for special education and developmental disabilities. Journal of
Special Education, 27, 16-34.

177

Guess, D. & Siegel-Causey, D. (1995). Attractor dimensions of behavior state changes
among individuals with profound disabilities. American Journal on Mental
Retardation, 99, 642-663.

Guess, D., Siegel-Causey, E., Roberts, 8., Guy, B., Ault, M., & Rues, J. (1993). Analysis
of state organizational patterns among students with profound disabilities. Journal
of the Association for Persons with Severe Handicaps, I 8, 93-108.

Guess, D., Siegel-Causey, E., Roberts, 8., Rues, J ., Thompson, B. & Siegel-Causey, D.
(1990). Assessment and analysis of behavior state and related variables among
students with profoundly handicapping conditions. Journal of the Association for
Persons with Severe Handicaps, 15, 211-230.

Gutowski, S. J. (1996). Response acquisition for music or beverages in adults with
profound multiple handicaps. Journal of Developmental and Physical
Disabilities, 8, 221-231.

Guy, 8., Guess, D., & Ault, M. M. (1993). Classroom procedures for the measurement of
behavior state among students with profound disabilities. Journal of the
Association for Persons with Severe Handicaps, 18, 52-60.

Hanline, M. F ., & Fox, L. (1993). Learning within the context of play: Providing typical
early childhood experiences for children with severe disabilities. Journal of the
Association for Persons with Severe Handicaps, 18, 121-129.

Hartmann, D. P., & Wood, D. D. (1982). Observational methods. In A. Bellack, M. H.
Hersen, & A. Kazdin (Eds), International handbook of behavior modification and
therapy (pp. 109-138). New York: Plenum Press.

Haskett, J ., & Hollar, W. (1978). Sensory reinforcement and contingency awareness of
profoundly retarded children. American Journal of Mental Deficiency, 83, 60-
68.

Hersen, M. (1982). Single-case experimental designs. In A. Bellack, M. H. Hersen, & A.
Kazdin (Eds), International handbook of behavior modification and therapy
(pp. 167-203). New York: Plentun Press.

Holcombe, A., Wolery, M., & Gast, D. L. (1994). Comparative single subject research:
Description of designs and discussion of problems. Topics in Early Childhood
Special Education, 14, 119-145.

Holvoet, J ., Guess, D., Mulligan, M., & Brown, F. (1980). The individualized curriculum

sequencing model (11): A teaching strategy for severely handicapped students.
Journal of the Association for Persons with Severe Handicaps, 5, 337-351.

178

Houghton, J., Bronicki, G., & Guess, D. (1987). Opportunities to express preferences and
make choices among students with severe disabilities in classroom settings.
Journal of the Association for Persons with Severe Handicaps, 12, 18-27.

Howell, M. T. (1992). Cognitive outcomes of contingency intervention for infants with
handicaps. (Doctoral dissertation, University of Illinois, 1992). Dissertation
Abstracts International, 53, 2765.

Ittenbach, R. F., & Lawhead, W. F. (1996). Historical and philosophical foundations of
single case research. In R. D. Franklin, D. B. Allison, & B. S. Gorman (Eds),
Design and analysis of single-case research (pp. 13-39). Mahwah, NJ: Erlbaum.

Ivancic, M. T., & Bailey, J. S. (1996). Current limits to reinforcer identiﬁcation for some
persons with profound multiple disabilities. Research in Developmental
Disabilities, 17, 77-92.

Ivancic, M. T., Barrett, G. T., Simonow, A., & Kimberly, A. (1997). A replication to
increase happiness among some people with profound multiple disabilities.
Research in Developmental Disabilities, 18, 70-89.

Izen, C. L., & Brown, F. (1991). Education and treatment needs of students with
profound, multiply handicapping, and medically fragile conditions: A survey of

teachers’ perceptions. Journal of the Association for Persons with Severe
Handicaps, 16, 94-103.

Jacobs, C. (1997). Investigating non-tactile midi controllers for severely disabled
children. Retrieved June 20, 2000, from
http://www.zip.com.au/~rsterry/kbcynpap.htm

Jorgensen, C., & Calculator, S. (1994). The evolution of best practices in educating
students with severe disabilities. In S. Calculator & C. J orgensen (Eds),
Including students with severe disabilities in schools: Fostering communication,
interaction, and participation (pp. 1-25). San Diego, CA: Singular Publishing
Group.

Kaiser, A. P., & Goetz, L. (1993). Enhancing communication with persons labeled
severely disabled. Journal of the Association for Persons with Severe Handicaps,
18, 137-142.

Kratochwill, T. R. (1978). Foundations of time-series research. In T. R. Kratochwill
(Ed.) Single subject research: Strategies for evaluating change (pp. 1-100).
Orlando, FL: Academic Press.

Katsiyannis, A., Ellenburg, J. S., Acton, O. M., & Torrey, G. (2001). Addressing the

needs of students with Rett syndrome. Teaching Exceptional Children, 33(5),
74-7 8.

179

Kaufﬁnan, J. M., & Krouse, J. (1981). The cult of educability: Searching for the
substance of things hoped for; the evidence of things not seen. Analysis and
Intervention in Developmental Disabilities, 1, 53-60.

Kazdin, A. (1982). Single case research designs: Methods for clinical and applied
settings. New York: Oxford University Press.

Kearns, K. (1986). Flexibility of single-subject experimental designs. Part H: Design
selection and arrangement of experimental phases. Journal of Speech and
Hearing Disorders, 51, 204-214.

Kennedy, C. H., Meyer, K. A., & Knowles, T. (2000). Analyzing the multiple functions
of stereotypical behavior for students with autism: Implications for assessment
and treatment. Journal of Applied Behavior Analysis, 33 , 559-571.

Kinsley, T. C., & Langone, J. (1995). Applications of technology for infants, toddlers and
preschoolers with disabilities. Journal of Special Education Technology, 12,
312-324.

Komrey, J ., & Foster-Johnson, L. (1996). Determining the efﬁcacy of intervention:
The use of effect sizes for data analysis in single-subject research.
Journal of Experimental Education, 65, 73-93.

Komer, A. F. (1972). State as a variable, as obstacle, and as mediator of stimulation
in infant research. Merrill-Palmer Quarterly, 18, 77-94.

Komer, A. F., Brown, B. W., Reade, E. P., Femback, S. A., & Thom, V. A. (1988).
State behavior of preterm infants as a function of development, individual and
sex differences. Infant Behavior and Development, 11, 11 1-124.

Kratochwill, T. R., & Williams, BL. (1988). Perspectives on pitfalls and hassles in
single-subject research. Journal of the Association for Persons with Severe
Handicaps, 13, 147-154.

Kupfer, F. (1982). Before and after Zachariah: A family story about a different
kind of courage. New York: Delacorte Press.

Lancioni, G. E., O’Reilly, M. F., Singh, N. N., Oliva, D., & Groeneweg, J. (2002).
Impact of stimulation versus microswitch-based programs on indices of
happiness of people with profound multiple disabilities. Research in
Developmental Disabilities, 23, 149-160.

Lancioni, G. E., O’Reilly, M. F. Van Dijk, J. & Klaase, M. (1998). An environmental
enrichment programme to promote adaptive responding in two children with

multiple disabilities. Scandinavian Journal of Behavior Therapy, 27, 130-134.

Landesman-Dwyer, S. & Sackett, G. P. (1979). Behavioral changes in nonambulatory,

180

profoundly mentally retarded individuals. In C.E. Meyers (Ed. ), Quality of life
in severely and profoundly mentally retarded people: Research foundations for
improvement (American Association on Mental Deﬁciency Monograph No. 3),
pp. 55-144. Washington, DC: American Association on Mental Deﬁciency.

Langley, M. B. (1990). A developmental approach to the use of toys for facilitation of
environmental control. In G. Hedmann (Ed.), Rehabilitation technology
(pp. 69-91). New York: Haworth Press.

Leatherby, J. G., Gast, D. L., Wolery, M., & Collins, BC. (1992). Assessment of
reinforcer preference in multi-handicapped students. Journal of Developmental
and Physical Disabilities, 4, 15-36.

Lewis, M. D. (2000). The promise of dynamic system approaches for an integrated
account of human development. Child Development, 71, 36-43.

Lewis, M., Sullivan, M., & Brooks-Gunn, J. (1985). Emotional behavior during the
learning of a contingency in early infancy. British Journal of Development
Psychology, 3, 307-316.

Lipsey, M. W. (1990). Design sensitivity. Newbury Park, CA: Sage.

Mallory, B. L. (1992). Is it always appropriate to be developmental? Convergent models
of early intervention practice. Topics in Early Childhood Education, 11(4), 1-12.

Matyas, T. A., & Greenwood, K. M. (1990). Visual analysis of single-case time series:
Effects of variability, serial dependence and magnitude of intervention effects.
Journal of Applied Behavior Analysis, 23, 341-351.

McCartney, K., & Rosenthal, R. (2000). Effect size, practical importance, and social
policy for children. Child Development, 71, 173-180.

McClennen, S. (1991). Cognitive skills for community living: Teaching students with
moderate to severe disabilities. Austin, TX: Pro-Ed.

McCubbin, T. (1987). Routine and emergency medical procedures. In M. E. Snell (Ed.),
Systematic instruction of persons with severe handicaps (pp. 152-171).
Columbus, OH: Charles E. Merrill.

McDonnell, J ., Hardman, M., McDonnell, A., & Kiefer-O’Donnell, R. (1995). An
introduction to persons with severe disabilities: Educational and social issues.

Boston: Allyn and Bacon.

Morley, S., & Adams, M. (1991). Graphical analysis of single-case time series data.
British Journal of Clinical Psychology, 30(2), 97-115.

181

Morse, M. T. (1992). Augmenting assessment procedures for children with severe
multiple handicaps and sensory impairments. Journal of Visual Impairment and
Blindness, 86, 73-77.

Morton, K. (1978). Identifying the enemy- a parent’s complaint. In A. Tumball, H. K.
Tumball (Eds), Parents speak out: Views from the other side of the two
way mirror (pp.143-148). Columbus, OH: Charles E. Merrill.

Mount, H., & Cavet, J. (1995). Multi-sensory environments: An exploration of their
potential for young people with profound and multiple learning difﬁculties.
British Journal of Special Education, 22(2), 52-55.

Mudford, O. C., Hogg, J ., & Roberts, J. (1997). Interobserver agreement and
disagreement in continuous recording exempliﬁed by measurement
of behavior state. American Journal on Mental Retardation, 102, 54-66.

Mudford, O. C., Hogg, J ., & Roberts, J. (1999). Behavior states: Now you see them,
now you don’t. American Journal on Mental Retardation, 4, 385-391.

Mulderij, K. J. (2000). Dualistic notions about children with motor disabilities: Hands
to lean on or to reach out? Qualitative Health Research, 10(1), 39-50.

Myers, S. B. (1990). A comparison of the behavior states of preschoolers with and
without severe/profound handicaps (Master’s thesis, University of Kansas, 1990).

National Information Center for Children and Youth with Disabilities. (2002, August).
Fact sheet 10: Severe and/or multiple disabilities. Retrieved October 5, 2002,
from http:// www.nichy.org/pubs/factshe/fs10txt.htm

Newton, J. S., Ard, Jr., W: R., & Homer, R. H. (1993). Validating predicted activity
preferences of individuals with severe disabilities. Journal of Applied Behavior
Analysis, 26, 239-245.

Nielsen, L. (1991). Spatial relations in congenitally blind infants: A study. Journal of
Visual Impairment and Blindness, 85, 11-16.

Nielsen, L. (1992). Space and self. Copenhagen, Denmark: SIKON.

Noonan, M. J ., & McCormick, L. (1993). Early intervention in natural environments:
Methods & procedures. Paciﬁc Grove, CA: Brooks/Cole.

O’Brien, Glenn, 8., & Cunningham, C. (1994). Contingency awareness in infants and
children with severe and profound learning disabilities. International Journal of

Disability, Development and Education, 41, 231-243.

Orelove, F. & Sobsey, D. (1987). Educating children with multiple disabilities.
Baltimore: Paul H. Brookes.

182

Ottenbacher, K. J. (1990). When is a picture worth a thousand p values? A
comparison of visual and quantitative methods to analyze single subject
data. Journal of Special Education, 23, 436-449.

Oyama, S., Grifﬁths, P. E., & Gray, R. D. (2001). Introduction: What is developmental
systems theory? In S. Oyama, P. E. Grifﬁths, & R. D. Gray (Eds), Cycles of

contingency: Developmental systems and evolution (pp. 1-11). Cambridge, MA:
MIT Press.

Park, H., Marascuilo, L., & Gaylord-Ross, R. (1990). Visual inspection and statistical

analysis in single-case designs. Journal of Experimental Education, 58,
311-320.

Parsonson, B. S., & Baer, D. M. (1992). The visual analysis of data, and current
research into the stimuli controlling it. In T. R. Kratochwill & J. R. Levin (Eds),
Single-case research design and analysis (pp. 15-40). Hillsdale, NJ: Erlbaum.

Prechtl, H. F. R. (1974). The behavioral states of the newborn infant (a review).
Brain Research, 76, 185-212.

Prechtl, H. F. R., Theorell, K., & Blair, A. W. (1973). Behavioral state cycles in a
abnormal infants. Developmental Medicine and Child Neurology, 15, 606-615.

Pronger, N. (1995). Microtechnology. In J. Hogg & J. Cavet (Eds), Making leisure
provision for people with profound learning and multiple disabilities (pp. 197-
211). London: Chapman & Hall.

Rainforth, B. (1982). Biobehavioral state and orienting: Implications for education
profoundly retarded students. Journal of the Association for Persons with Severe
Handicaps, 6, 33-36.

Realon, R. E., Favell, J. E., & Dayvault, K. A. (1988). Evaluating the use of adapted
leisure materials on the engagement of persons who are profoundly, multiply
handicapped. Education and Training in Mental Retardation, 23, 228-237.

Reboussin, D. M.., & Morgan, T. M. (1996). Statistical considerations in the use and

analysis of single-subject designs. Medicine and Science in Sports and Exercise,
28, 639-644.

Reid, D., Everson, J., St Green, C. (1999). A systematic evaluation of preferences
identiﬁed through person- centered planning for people with profound multiple
disabilities. Journal of Applied Behavior Analysis, 32, 467-477.

Reid, D. H., Phillips, J. F., & Green, C. W. (1991). Teaching persons with profound

multiple handicaps: A review of the effects of behavioral research.
Journal of Applied Behavioral Analysis, 24, 319-336.

183

Richards, S. B., & Richards, R. Y. (1997). Implications for assessing biobehavioral
state in individuals with profound disabilities. Focus on Autism and other
Developmental Disabilities, 12, 79-86.

Richards, 8., & Stemberg, L. (1989). Assessing levels of state and arousal in individuals
with profound handicaps: A research integration. Journal of Mental Deﬁciency
Research, 33, 381-387.

Richards, S. B., & Stemberg, L. (1992). A preliminary analysis of environmental
variables affecting the observed biobehavioural states of individuals with
profound handicaps. Journal of Intellectual Disability Research, 36, 403-414.

Richards, 8., & Stemberg, L. (1993). Corroborating previous ﬁndings: Laying
stepping stones in the analysis of biobehavioral states in students
with profound disabilities. Education and Training in Mental
Retardation, 28, 262-268.

Richards, 8., & Stemberg, L. (1994). Assessing levels of state and arousal. In L.
Stemberg (Ed.), Individuals with profound disabilities: Instructional
and assistive strategies (pp. 61-78). Austin, TX: Pro-Ed.

Roberts, S. I. (1992). An investigation and analysis of behavior state conditions and
related enviromnental events among children and youth with multiple severe
and profound disabilities. (Doctoral dissertation, University of Kansas).
Dissertation Abstracts International, 54, 145.

Robinson, L. M. (1986). Designing computer intervention for very young handicapped
children. Journal of the Division for Early Childhood, 10, 209-215.

Sahni, R., Schulze, K. F., Stefansi, M., Myers, M. M., & Fifer, W. P. (1995).
Methodological issues in coding sleep states in immature infants.
Developmental Psychobiology, 28, 85-101.

Sailor, W., Gee, K., Goetz, L. & Graham, N. (1988). Progress in educating students with
the most severe disabilities: Is there any? Journal of the Association for Persons
with Severe Handicaps, 13, 87-89.

Sailor, W., & Guess, D. (1983). Severely handicapped students: An instructional
design. Boston: Houghton Mifﬂin.

Salvia, J ., & Ysseldyke, J. E. (2001). Assessment. Boston: Houghton Mifﬂin.
Saunders, M. D., Questad, K. A., Kedziorski, T. L., Boase, B. C., Patterson, E. A., &
Cullinan, T. B. (2001). Unprompted mechanical switch use in individuals with

severe multiple disabilities: An evaluation of the effects of body position.
Journal of Developmental and Physical Disabilities, 13, 27-39.

184

Schneekloth, L. H. (1989). Play environments for visually impaired children. Journal of
Visual Impairment and Blindness, 83, 196-201.

Siegel-Causey, E., Guy, B., & Guess, D. (1995). Severe and multiple disabilities. In E.
Meyen & T. Skrtic (Eds), Special Education and Student Disability: An
Introduction (pp. 415-450). Denver, CO: Love.

Silliman-French, L., French, R., Sherrill, C., & Gench, B. (1998). Auditory feedback and
time-on—task of postural alignment of individuals with profound mental
retardation. Adapted Physical Activity Quarterly, 15, 51-63.

Simeonsson, R. J ., Huntington, G. S., & Parse, S. A. (1980). Expanding the
developmental assessment of young handicapped children. New Directions for
Exceptional Children, 3, 51-74.

Simeonsson, R. J ., Huntington, G. 8., Short, R. J ., & Ware, W. B. (1982). The Carolina
record of individual behavior: Characteristics of handicapped infants and children.
Topics in Early Childhood Education, 2(2), 43-55.

Snell, M. E. (1987). What does an “appropriate” education mean? In M. E. Snell (Ed.),
Systematic instruction of persons with severe handicaps (pp. 1-6). Columbus,
OH: Charles E. Merrill.

Stemberg, L. (1988). Future educational concerns: Crucial questions. In L. Stemberg
(Ed.), Educating students with severe or profound handicaps (pp. 475-484).
Rockville, MD: Aspen.

Stemberg, L., & Richards, S. (1989). Assessing levels of state and arousal in individuals
with profound handicaps: A research integration. Journal of Mental Deﬁciency
Research, 33, 381-387.

Sullivan, M. W. (1990). Contingency intervention: A program portrait. Journal of Early
Intervention, 14, 367-375.

Sullivan, M. W., Laverick, D. H., & Lewis, M. (1995). Brief report: Fostering
environmental control in a young child with Rett Syndrome: A case study.
Journal of Autism and Developmental Disorders, 25, 215-221.

Sullivan, M. W., & Lewis, M. (1993). Contingency, means-ends skills, and the
use of technology in infant intervention. Infants and Young Children,
5(4), 58-77.

Swingler, T. (1998, March). “T hat was me! " Applications of the Soundbeam MIDI
controller as a key to creative communication, learning, independence
and joy. Paper presented at California State University —Northridge Technology
and Persons with Disabilities Conference. Los Angeles, CA. Paper retrieved

185

July 15, 2002 from
http://wwwcod.csun.edu/conf/1 998/proceedings/csun98_l 63 .htm

Switzky, H., Rotatori, A. F ., Miller, T., & Freagon, S. (1979). The developmental model
and its implications for assessment and instruction for the severely/profoundly
handicapped. Mental Retardation, 17, 167-170.

Tawney, J ., & Gast, D. (1984). Single-subject research in special education. Columbus,
OH: Merrill.

Thelen, E. (1990). Dynamical systems and the generation of individual differences.
In J. Colombo & J. Fagen (Eds), Individual differences in infancy:
Reliability, stability, prediction (pp. 19-43). Hillsdale, NJ: Erlbaum.

Thelen, E. (2001). Dynamic mechanisms of change in early perceptual-motor
development. In J. McClelland & R. Siegler (Eds), Mechanisms of
cognitive development: Behavioral and neural perspectives (pp. 161-184).
Mahwah, NJ: Erlbaum.

Thelen, E., & Smith, LB. (1994). A dynamic systems approach to the development
of cognition and action. Cambridge, MA: MIT Press.

Thoman, E. B. (1990). Sleeping and waking states in infants: A functional perspective.
Neuroscience and Biobehavioral Reviews, 14(1), 93-107.

Thoman, E. B., Davis, D. H., & Deneberg, V. H. (1987). The sleeping and waking states
of infants: Correlations across time and person. Physiology and Behavior, 41,
531-537.

Thoman, E. B., Komer, A. F., & Kraemer, H. C. (1976). Individual consistency in
behavior states in neonates. Developmental Psychobiology, 9, 271-283.

Thompson, S. B. N., & Martin, S. (1994). Making sense of multisensory rooms for
people with learning disabilities. British Journal of Occupational Therapy, 5 7,
341-344.

Tripodi, T. (1994). A primer on single-subject design for clinical social workers.
Washington, DC: NASW Press.

United States Department of Education, Ofﬁce of Special Education and Rehabilitative
Services. (1995). Twentieth annual report to congress on the implementation of
the Individuals with Disabilities Education Act (IDEA). Code of Federal
Regulations, Title 34, Section 300.7.

van Geert, P. (1991). A dynamic systems model of cognitive and language growth.
Psychological Review, 98(1), 3-53.

186

Watson, J. S., Hays, L. A., & Vietze, P. (1982). Response-contingent stimulation as a
treatment for developmental failure in infancy. Journal of Applied
Developmental Psychology, 3, 191-203.

Westlake, C., & Kaiser, A. (1991). Early childhood services for children with severe
disabilities. In L. Meyer, C. Peck, & L. Brown, (Eds), Critical issues in the lives
of people with severe disabilities (pp. 429-45 8). Baltimore: Paul H. Brookes.

White, 0. R. (1980). Adaptive performance objectives: Form versus function. In W.
Sailor, B. Wilcox, & L. Brown (Eds), Methods of instruction for severely
handicapped students (pp. 47-69). Baltimore: Paul H. Brookes.

Wisniewski, L., & Alper, S. (1994). Including students with severe disabilities in general
education settings: Guidelines for change. Remedial and Special Education, 15,
4-13.

Wolery, M., Ault, M., & Doyle, P. (1992). Teaching students with moderate to severe
disabilities. New York: Longrnan.

Wolff, P. H. (1959). Observations on newborn infants. In L. J. Stone, H. T. Smith, & L.
B. Murphy (Eds), The competent infant (pp. 257-272). New York: Basic Books.

187