Citation
Interhemispheric Transfer of Praxis Information Using Probable Alzheimer Disease as a Model for Disconnection Apraxia

Material Information

Title:
Interhemispheric Transfer of Praxis Information Using Probable Alzheimer Disease as a Model for Disconnection Apraxia
Creator:
KNIGHT, ANN MARIE ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Alzheimers disease ( jstor )
Apraxia ( jstor )
Asymmetry ( jstor )
Corpus callosum ( jstor )
Hemispheres ( jstor )
Ideomotor apraxia ( jstor )
Imitation foods ( jstor )
Information representations ( jstor )
Lesions ( jstor )
Pantomime ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Ann Marie Knight Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/1/2005
Resource Identifier:
71303057 ( OCLC )

Downloads

This item has the following downloads:

knight_a ( .pdf )

knight_a_Page_067.txt

knight_a_Page_020.txt

knight_a_Page_079.txt

knight_a_Page_095.txt

knight_a_Page_072.txt

knight_a_Page_018.txt

knight_a_Page_076.txt

knight_a_Page_099.txt

knight_a_Page_006.txt

knight_a_Page_041.txt

knight_a_Page_011.txt

knight_a_Page_017.txt

knight_a_Page_026.txt

knight_a_Page_030.txt

knight_a_Page_065.txt

knight_a_Page_022.txt

knight_a_Page_031.txt

knight_a_Page_005.txt

knight_a_Page_004.txt

knight_a_Page_063.txt

knight_a_Page_032.txt

knight_a_Page_088.txt

knight_a_Page_085.txt

knight_a_Page_034.txt

knight_a_Page_019.txt

knight_a_Page_015.txt

knight_a_Page_107.txt

knight_a_Page_010.txt

knight_a_Page_048.txt

knight_a_Page_100.txt

knight_a_Page_050.txt

knight_a_Page_014.txt

knight_a_Page_055.txt

knight_a_Page_070.txt

knight_a_Page_105.txt

knight_a_Page_057.txt

knight_a_Page_069.txt

knight_a_Page_075.txt

knight_a_Page_081.txt

knight_a_Page_108.txt

knight_a_Page_060.txt

knight_a_Page_106.txt

knight_a_Page_062.txt

knight_a_Page_044.txt

knight_a_Page_092.txt

knight_a_Page_002.txt

knight_a_Page_037.txt

knight_a_Page_109.txt

knight_a_Page_093.txt

knight_a_Page_074.txt

knight_a_Page_025.txt

knight_a_Page_013.txt

knight_a_Page_066.txt

knight_a_Page_083.txt

knight_a_Page_012.txt

knight_a_Page_102.txt

knight_a_Page_098.txt

knight_a_Page_086.txt

knight_a_Page_077.txt

knight_a_Page_027.txt

knight_a_Page_078.txt

knight_a_Page_059.txt

knight_a_Page_101.txt

knight_a_Page_024.txt

knight_a_Page_080.txt

knight_a_Page_028.txt

knight_a_Page_042.txt

knight_a_Page_035.txt

knight_a_Page_054.txt

knight_a_Page_008.txt

knight_a_Page_090.txt

knight_a_Page_068.txt

knight_a_Page_046.txt

knight_a_Page_104.txt

knight_a_Page_053.txt

knight_a_Page_096.txt

knight_a_Page_103.txt

knight_a_Page_021.txt

knight_a_Page_023.txt

knight_a_pdf.txt

knight_a_Page_064.txt

knight_a_Page_016.txt

knight_a_Page_007.txt

knight_a_Page_073.txt

knight_a_Page_040.txt

knight_a_Page_047.txt

knight_a_Page_033.txt

knight_a_Page_009.txt

knight_a_Page_036.txt

knight_a_Page_087.txt

knight_a_Page_082.txt

knight_a_Page_089.txt

knight_a_Page_097.txt

knight_a_Page_061.txt

knight_a_Page_001.txt

knight_a_Page_049.txt

knight_a_Page_094.txt

knight_a_Page_045.txt

knight_a_Page_071.txt

knight_a_Page_051.txt

knight_a_Page_056.txt

knight_a_Page_039.txt

knight_a_Page_003.txt

knight_a_Page_038.txt

knight_a_Page_029.txt

knight_a_Page_043.txt

knight_a_Page_091.txt

knight_a_Page_084.txt

knight_a_Page_058.txt

knight_a_Page_052.txt


Full Text












INTERHEMISPHERIC TRANSFER OF PRAXIS INFORMATION USING
PROBABLE ALZHEIMER'S DISEASE AS A MODEL FOR DISCONNECTION
APRAXIA
















By

ANN MARIE KNIGHT


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Ann Marie Knight

































This dissertation is dedicated to my loving husband, Travis W. Knight, Ph.D. Without
his unwavering support and encouragement, this research study could not have been
completed.















ACKNOWLEDGMENTS

I would like to begin by extending tremendous gratitude and empathy to the study

participants and their caregivers. Without their willingness to sacrifice their time, this

research could not have been completed.

I would also like to thank my parents, Marianne and Bernard Cimino and Dannis

and Frances Knight for their infinite support and wisdom throughout this process.

Without their love, enthusiasm, encouragement and help caring for my son, Connor, I

truly would not have been able to complete this academic endeavor.

To my friend and mentor, Leslie J. Gonzalez Rothi, Ph.D., I would like to extend

immense gratitude and appreciation for her extensive efforts throughout my academic

career. For the past five years, she has spent countless hours with me on the design and

implementation of numerous research projects, has taught me how to apply academic

knowledge to clinical practice, and has provided me with an example of a compassionate

and knowledgeable speech pathologist. Without her extensive knowledge of

neuropsychology, communication disorders and research design, the writing of this

manuscript could not have come to fruition. Her endless dedication to educating students

has impacted my life in numerous ways. I have learned a great deal through her expertise

and supervision.

Further acknowledgements should be extended to Kenneth Heilman, MD, Leilani

Doty, Ph.D. and the fellows and staff of the University of Florida Memory Disorder

Clinic not only for their assistance in recruiting participants for this study but for









imparting their knowledge about the diagnosis and treatment of persons with memory

disorders. Special thanks are extended to Dr. Heilman for all of his advice on the design

and implementation of this experiment. It has truly been an honor and a privilege to be

mentored by someone who has so greatly contributed to the field of behavioral

neurology. I am extremely honored to have studied with him and to call him not only my

mentor but also my friend.

Additionally, I would like to acknowledge the other members of my committee, Dr.

Russell M. Bauer and Dr. Christine M. Sapienza, for their support of this project and their

commitment to research excellence. My experience at the University of Florida has been

enhanced greatly by having had the opportunity to work with these extremely talented

individuals.

I would like to extend a special thanks to Cristina Posse, MHS and Lauren Meffen,

BA, for their willingness to collaborate with me on this project. They were responsible

for spending countless hours analyzing all of the data for this project, and this research

truly could not have been completed without their dedication and perseverance. I feel

privileged to have been given the opportunity to work with two such bright and

outstanding students. In addition, special thanks are extended to Haijing Qin, M.S. of the

VA RR&D Rehabilitation Outcomes Research Center for excellent statistical support.

Statistical analysis of these data would not have been possible without her expertise and

advice. To the support staff at the University of Florida Department of Neurology (i.e.

Doug Perkinson) and the VA RR&D Brain Rehabilitation Research Center (i.e. Susan

Nadeau, Joy McCallum, and Lisa Demanuel), I would like to extend sincere gratitude for

providing excellent clerical support. Lastly, to the health reporters at The Gainesville Sun









and The Ocala Star Banner for helping me to recruit participants by publishing study

announcements free of charge and to the Gainesville, FL Alzheimer's Association for

being proactively involved with this population and for being willing to support research

endeavors.

This study was funded by 1) the VA Office of Academic Affiliations and Patient

Care Services Predoctoral Fellowship in Speech Pathology, 2) the VA Rehabilitation

Research and Development Office Centers of Excellence Brain Rehabilitation Research

Center, 3) the National Institute of Deafness and Communication Disorders, National

Institutes of Health, 4) the Florida Department of Elder Affairs, Memory Disorder Clinic

and 5) the University of Florida, Department of Communication Sciences and Disorders

and Department of Neurology.















TABLE OF CONTENTS
Page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ........................ ....... ...................... .... ix

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT ........ .............. ............. ...... ...................... xi

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

W hat is Limb Apraxia?................................................... ........ .... .......... 2
Three Types of Disconnection Apraxia: Literature Review.......................................7
W hat is A lzheim er's D disease? ........................................................................ ... ... 12
What is Known About Limb Apraxia in AD? .....................................................15
Why Study Disconnection Apraxia in AD? ...........................................................19
Su m m ary .................. ..... ............................. ............... ................ 22
Purpose, Questions, and Hypotheses ...................................................................... 24
R research Q question 1 ........................ .. .................................. .. ........... 25
R research Question 2 .................. ........................... .... .. .. .. ........ .... 26
R research Q question 3 ........................ .. .................................. .. ........... 27
Research Question 4 .................. ........................... .... .. .. .. ........ .... 28

2 M ETHOD S ..................................... ................................. ........... 31

S u objects ...................................... .................................................... 3 1
Inclusion C criteria .......................................... ........... ... .. ............
Subject D em ographics............................................................ ............... 33
Sample Size Estimation ............. ... .... ................ ..................... 34
Sample Size Estimation-2 Samples Equal Variances (most conservative
estim ate) .............................. .. ... ........................... ....... ...... 34
Sample Size Estimation-2 Samples Unequal Variances (least conservative
estim ate) ................................... ................................ .........35
E xperim mental Tasks ................................................................. ........ 35
Data Collection Procedures ...................... ............................ 35
Task 1: Verbal Command Pantomime (VC) ....................................... .......... 36
Task 2: Pantom im e Im itation (PI) ............................... .. ....................... 37
Task 3: Conceptual Pantom im e (CP) ...................................... ............... 38
Rater Training ...................................... ................................ ..........40









R e lia b ility ............................................................................................................. 4 1
Statistical A n aly sis........... .................................................................. ........ .. ..... .. 4 2
R research Q question 1 ........................ .. .................................. .. ........... 43
R research Question 2 .................. ........................... .... .... .. ........ .... 43
R research Q question 3 ........................ .. .................................. .. ........... 43
Research Question 4 .................. ........................... ... .. ... .. ........ .... 44

3 R E S U L T S .............................................................................4 9

Subject Dem graphics ........................... .. .......... ...... ............ 49
N europsychological Screening ...........................................................................49
R e lia b ility ............................................................................................................. 5 1
D descriptive Statistics ........................................................ ............ .......... .. ..53
Task 1: Verbal Command Pantomime (VC) ................................................. 53
Task 2: Pantom im e Im itation (PI) ................................ .. ....................... 53
Task 3: Conceptual Pantom im e (CP) ...................................... ............... 54
E rror Types Task 1, 2, and 3 ........................................ .......... ............... 55
Statistical A n aly sis............................................................................. ............... 56
R research Q question 1 ........................ .. .................................. .. ........... 56
R research Q question 2 .................. ............................ .... .. .. .. ............ 56
R research Q question 3 ........................ .. .................................. .. ........... 57
R research Question 4 .................. ............................ .... .. .. .. ............ 57
S u m m a ry ......................................................................................................5 8

4 D ISC U S SIO N ............................................................................... 67

Summary and Explanation of Findings ................. ...........................................68
R research Q question 1 ........................ .. .................................. .. ........... 68
Research Question 2 .................. ........................... ... .. ... .. ........ .... 70
R research Q question 3 ........................ .. .................................. .. ........... 72
Research Question 4 .................. ........................... .... .. .. .. ........ .... 75
C o n clu sio n s..................................................... ................ 7 7
Im p location s ........................................................................... 80

APPENDIX

A LIST OF STIM ULI...................................... .. ... ............. ........ 83

B DESCRIPTION OF ERRORS........................................................ ............. 87

LIST OF REFEREN CES ............................................................ .................... 89

B IO G R A PH IC A L SK E TCH ..................................................................... ..................97







viii
















LIST OF TABLES


Table pge

1-1: D diagnostic criteria for A D ............................................................... .....................29

2-1: Individual subject demographics for AD group. ....................................................45

2-2: Individual subject demographics for HC group. .............................. ................46

2-3: Strength of observer agreement for ranges of kappa statistic values. ........................47

3-1: Scores for screening measures for individual subjects in HC group.......................59

3-2: Scores for screening measures for individual subjects in AD group.......................60

3-3: Inter-rater reliability using % agreement and the Kappa statistic for task 1, 2, and 3 .....61

3-4: Intra-rater reliability using % agreement and the Kappa statistic for task 1, 2, and 3.....62

3-5: Response accuracy (percent) data with difference scores and asymmetry ratios for
individual subjects in HC group for Tasks 1, 2, and 3....................................63

3-6: Response accuracy (percent) data with difference scores and asymmetry ratios for
individual subjects in AD group for Tasks 1, 2, and 3...........................................64

3-7: Error analysis descriptive data for Task 1 (VC) ....................................................65

3-8: Error analysis descriptive data for Task 2 (PI) .....................................................65

3-9: Error analysis descriptive data for Task 3 (CP).......................................................66

3-10: Error totals for tasks 1, 2, and 3....................................... ........................... 66
















LIST OF FIGURES


Figure pge

1-1: Cognitive neuropsychological model of limb apraxia............... ...........................30

2-1: Examples of pictures used in the Florida Action Recall Test (FLART) ..................48















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

INTERHEMISPHERIC TRANSFER OF PRAXIS INFORMATION USING
PROBABLE ALZHEIMER'S DISEASE AS A MODEL FOR DISCONNECTION
APRAXIA

By

Ann Marie Knight

May 2005

Chair: Leslie J. Gonzalez Rothi
Major Department: Communication Sciences and Disorders

Praxis, or the ability to perform skilled movements, is essential to independent

living. Most skilled movements require the use of both hands and the inability to perform

skilled movements effectively can significantly impact quality of life. Despite the

importance of being able to perform skilled movements effectively and efficiently, little

is known about how the brain processes praxis movement information. What is known

about this type of processing has been learned from an ablation model. However, this

model impairs the motor function of the contralesional limb and does not allow the study

of bimanual praxis mechanisms. The purpose of this study was to investigate how praxis

information processing is represented in the brain by examining the interhemispheric

transfer of different types of praxis information. This was accomplished by examining

bimanual praxis mechanisms in individuals with Alzheimer's disease because individuals

in this population can perform praxis tasks with both hands and demonstrate both limb

apraxia and neural degeneration. This model allowed us to study how praxis information









is transferred between the two brain hemispheres and differentiate what type of praxis

information is being transferred across the corpus callosum.

In order to accomplish the goals of this study, it was necessary to confirm the

presence of limb apraxia in individuals with Alzheimer's disease. This study also

attempted to determine whether the limb apraxia that is present in this population is due

to the degradation of left hemisphere movement representations or the interruption of

interhemispheric transfer of praxis information. Another purpose of this study was to

differentiate whether the interhemispheric disconnection of praxis information was due to

the inability to transfer verbal or motor information across the corpus callosum. Findings

indicated that individuals with Alzheimer's disease have ideomotor and conceptual

apraxia in the nondominant hand and that information from praxis movement

representations in the left hemisphere are not transferred across the corpus callosum

adequately in individuals with Alzheimer's disease.














CHAPTER 1
INTRODUCTION

Humans use skilled movement in nearly every aspect of independent functioning

from preparing food to getting dressed to using hands and arms to gesture in combination

with verbal communication. When the ability to perform skilled movements is disrupted

functional independence can be severely compromised. Skilled movement is extremely

important to everyday life and movement precision in both hands is necessary for

effective and efficient completion of actions. Despite the importance of skilled

movement, little is known about how the praxis system that governs skilled movement

execution is organized in the brain. To perform skilled movements, motor cortex in each

hemisphere must access praxis movement representations that are thought to be localized

in the left hemisphere. To perform skilled movements with the left hand, information

from left hemisphere praxis movement representations must be transferred across the

corpus callosum to right hemisphere motor cortex. To perform skilled movements with

the right hand, information does not need to cross the corpus callosum but rather must be

transferred to left hemisphere motor cortex by intrahemispheric connection fibers.

Previous studies have typically relied on an ablation model to study the mechanisms of

the praxis system. However, unilateral stroke patients commonly demonstrate

contralesional hemiplegia, which may mask the presence of apraxia in the weak hand.

Therefore, stroke does not provide an ideal model for studying bimanual praxis

mechanisms, which rely on the transfer of motoric and conceptual praxis information

across brain hemispheres. Because we use both hands to perform skilled movements, it is









necessary to study praxis mechanism using a model that allows us to study bimanual

performance. The purpose of this study is to investigate how praxis information

processing is represented in the brain by examining the interhemispheric transfer of

different types of praxis information. This chapter defines limb apraxia and provides a

rationale for studying praxis mechanisms using Alzheimer's disease (AD) as a

pathological model.

What is Limb Apraxia?

Limb apraxia is an acquired disorder of skilled, learned, purposive movements

resulting from neurologic disease or injury that cannot be explained by language deficits

or primary sensorimotor disturbance (Maher & Ochipa, 1997; Rothi & Heilman, 1997).

In order to perform movements, sensory input (auditory, tactile, visual) must interact with

stored movement representations that are translated into patterns of innervation. Both

disconnection (Geschwind, 1965; Liepmann, 1980) and representational (Rothi, Ochipa,

& Heilman, 1991) models of apraxia have been proposed.

Liepmann (1980) (as described by Rothi, Ochipa, and Heilman, 1997a) proposed

that in right handed individuals, the left hemisphere guides skilled movements of both the

left and right hands and that the acquisition of skilled limb movements required the

acquisition of "movement formulae," "innervatory patterns," and "kinetic memories" for

learned movements. Liepmann proposed that "movement formulae" contain spatial and

temporal patterns for the production of movement sequences. "Innervatory patterns" are

acquired through practice and provide a method for transforming movement formulae

into muscle innervation patterns for correct limb positioning. "Kinetic memories" are

associations between innervatory patterns for action, which are highly practiced and can

be performed without spatial or visual feedback.









Geschwind (1965a, 1965b) also proposed that skilled movements of both hands

were mediated by the left hemisphere in right handed individuals. He suggested that

pantomime to command requires processing by left hemisphere language mechanisms.

For right handed movements, information is transferred from left hemisphere language

areas to left motor association cortex for programming of movements and left primary

motor cortex for motor innervation of the right hand. For left handed movements,

information is transferred from left hemisphere language areas to right hemisphere motor

cortex via the corpus callosum for motor innervation of the left hand. Disconnections can

occur which interfere with transfer of information from left hemisphere language areas to

left motor cortex for control of the right hand (left hemisphere lesions) or right motor

cortex for control of the left hand (corpus callosum lesions).

According to a representational model of limb apraxia developed by Rothi et al.

(1991) (Figure 1-1), the praxis system can be divided into conceptual and production

subsystems. This model can account for dissociations in praxis performance including

separate systems for receptive and expressive praxis, selective dissociation of sensory

input modalities from praxis movement representations, a direct route for praxis

imitation, and the notion that there is a separate system for action semantics (Rothi et al.,

1991).

The production system can be divided into a store of learned spatial-temporal

movement representations or praxicons and a mechanism for translating these

representations into motor programs or innervatory patterns, state that:

In order to perform a skilled learned act, one must place particular body parts in
certain spatial positions in a specific order at specific times. The spatial positions
assumed by the relevant body parts depend not only on the nature of the act but
also on the position and size of an external object with which the body parts must









interact. Skilled acts also require orderly changes in the spatial positions of the
body parts over time. These movement formulas command the motor systems to
adopt the appropriate spatial positions of the relevant body parts over time.
(Heilman & Rothi, 1993, p. 146).

A disruption of the production system, or ideomotor apraxia, is characterized

predominantly by spatiotemporal errors during pantomime to command and imitation of

gestures (Poizner, Mack, Verfaille, Rothi, & Heilman, 1990; Rothi, Mack, Verfaellie,

Brown, & Heilman, 1988). Performance in ideomotor apraxia may improve with

manipulation of the actual tool as a result of increased tactile and visual cuing as well as

contextual information. However, there is some evidence that actual tool use remains

defective (Poizner, Soechting, Bracewell, Rothi, & Heilman, 1989). Spatial errors are the

most characteristic errors of ideomotor apraxia and there are three forms of spatial errors

(Poizner et al., 1990; Rothi et al., 1988). Postural errors reflect an abnormality of the

required finger or hand posture and its relationship to the target tool (Rothi et al., 1997b).

Spatial orientation errors occur when the hand movement that is produced does not

appear to direct the tool toward an imagined object (Heilman & Rothi, 1993). Spatial

movement errors are disturbances of the characteristic joint movements necessary to

produce the correct action (Heilman & Rothi, 1993). Temporal errors in ideomotor

apraxia may occur in the form of a delay in the initiation of movement, occasional pauses

during the movement, or a failure to coordinate the speed of the movement with the

spatial components of the movement (Heilman & Rothi, 1993). Thus, ideomotor apraxia

may result from deficits of action implementation or degradation of praxis movement

representations (Heilman, Rothi, & Valenstein, 1982; Rothi, Heilman, & Watson, 1985).

In the first case, the patient is able to recognize gestures but gesture production is

impaired. This is thought to result from an impaired ability to execute skilled movements









despite intact movement representations. In the second case, the patient is not able to

recognize gestures and gesture production is impaired. This is thought to result from a

degradation of the movement representations (Cimino-Knight, Hollingsworth, Maher,

Raymer, Foundas, Heilman, & Rothi, 2002).

The conceptual subsystem involves three types of knowledge: knowledge of tool

and object functions, knowledge of actions independent of tools, and knowledge about

the organization of single actions into goal oriented sequences (i.e., action semantics)

(Rothi et al., 1991; Roy & Square, 1985). A tool is used to provide a mechanical

advantage in an action and an object is the recipient of an action (Rothi et al., 1997a).

Knowledge of the functions of objects and tools may have internalized linguistic referents

and externalized function knowledge (Roy & Square, 1985). The internalized linguistic

referents contain semantic descriptions of objects and actions. The externalized function

knowledge provides information about the perceptual attributes of the object and action

and the environmental context in which tools are used. It has been proposed that the

semantic system has specialized subsystems for all modalities and modes of processing

which contain specific conceptual representations (i.e., action semantics, verbal

semantics, visual semantics, auditory semantics). These multiple semantic systems are

thought to communicate with each other such that visual or verbal input can result in

action output (Rothi et al, 1997a; Raymer & Ochipa, 1997).Conceptual apraxia is a

disruption of the conceptual system that interferes with the knowledge of tool and object

functions and their associated actions (Ochipa, Rothi, & Heilman, 1992). Patients with

conceptual apraxia cannot recall the type of actions associated with specific tools or

objects and thus exhibit content errors (DeRenzi & Lucchelli, 1988; Ochipa, Rothi, &









Heilman, 1989; Ochipa et al, 1992). Other types of errors that are possible in patients

with conceptual apraxia include the inability to recall which tool is associated with an

object, lack of awareness of the mechanical advantage of particular tools, or the inability

to create novel tools to solve mechanical problems (Heilman & Rothi, 1993). It is

hypothesized that the basic deficit underlying conceptual apraxia is a degradation of

action semantics (Schwartz, Adair, Raymer, Williamson, Crosson, Rothi, Nadeau, &

Heilman, 2000).

Ideational apraxia is a disruption of the conceptual system that interferes with

knowledge about the organization of single actions into sequences. Patients with

ideational apraxia demonstrate an inability to carry out a series of actions and have

difficulty sequencing actions in the proper order (Heilman & Rothi, 1993). Essentially,

ideational apraxia is a loss of ability to conceptualize, plan, and execute a complex

sequence of motor actions involving the use of tools or objects (LeClerc & Wells, 1998).

Lesions that disconnect various forms of sensory input from praxis movement

representations have also been described in the literature (DeRenzi, Faglioni, & Sorgato,

1982; Gazzaniga, Bogen, & Sperry, 1967; Geschwind & Kaplan, 1962; Heilman, 1973).

Gazzaniga et al. (1967), Geschwind and Kaplan (1962), Geschwind (1965a, 1965b) and

Heilman (1973) described individuals who demonstrated a disconnection between

language areas necessary for comprehension of commands and movement representations

necessary for selecting and programming the appropriate actions (Heilman & Rothi,

1993).

Furthermore, DeRenzi et al. (1982) described modality-specific apraxias that result

from a disconnection of praxis movement representations and specific sensory input









(visual, verbal, or tactile). These types of dissociation apraxia are known as verbal-

motor, visuo-motor, and tactile-motor dissociation apraxias (Heilman & Rothi, 1993).

Callosal apraxia refers to apraxia that is more severe in the left hand than in the

right hand due to a lesion of the corpus callosum (in some patients apraxia may be absent

in the right hand) (Geschwind, 1965; Geschwind & Kaplan, 1962; Graff-Radford, Welsh,

& Godersky, 1987; Watson & Heilman, 1983). This type of lesion disconnects the

movement representations and action semantics in the left hemisphere (selection of

appropriate actions from a store of learned movement patterns) from right hemisphere

motor association areas and primary motor cortex (programming of innervatory patterns

for movements of the left hand). This can be explained by an interaction between left

hemisphere localization of praxis movement representations and control of the left hand

and arm by contralateral primary motor cortex. Because the left hand is controlled by

right primary motor cortex, lesions of the corpus callosum disconnect the left hand from

left hemisphere movement and semantic representations. Therefore, if there is damage to

or degeneration of the corpus callosum, right primary motor cortex may not be able to

access left hemisphere praxis movement representations and action semantics.

Three Types of Disconnection Apraxia: Literature Review

Thus far an overview of what is known about the types and mechanisms of limb

apraxia has been presented. The majority of studies of limb apraxia to date have used

unilateral stroke patients to investigate intrahemispheric transfer of praxis information.

These studies have examined praxis performance in the ipsilesional (left) hand only due

to the presence of contralesional hemiplegia. This population does not provide an

adequate model for studying interhemispheric transfer of praxis information. Following

is a discussion of what is known about the transfer of praxis information across the









corpus callosum from individuals with callosal lesions. According to the literature, at

least three types of apraxia are possible as a result of callosal disconnection: ideomotor

apraxia, conceptual apraxia, and verbal-motor dissociation apraxia (Degos, Gray, Louarn,

Ansquer, Poirier, & Barbizet, 1987; Gazzaniga et al., 1967; Geschwind, 1965;

Geschwind & Kaplan, 1962; Goldenberg, Wimmer, Holzner, & Wessely, 1985; Graff-

Radford et al., 1987; Kazui & Sawada, 1993; Tanaka, Iwasa, & Obayashi, 1990; Watson

& Heilman, 1983).

Heilman (1973) described three individuals with left hemisphere lesions who could

not perform actions to command with either hand but could imitate gestures and use

objects flawlessly with both hands. When asked to pantomime to command the

participants "appeared as if they did not understand the command" (p.862) but the spared

ability to imitate gestures and use objects suggests that "the engrams for motor sequences

are intact" (p.863). Heilman (1973) explained the deficit in these individuals as a deficit

in the transfer of information between language comprehension and motor encoding.

Similarly, Geschwind and Kaplan (1962) and Gazzaniga et al. (1967) reported

individuals with callosal lesions who could not perform actions to verbal command with

the left hand but could imitate gestures and use objects. Based on the patient described

by Geschwind and Kaplan (1962), Geschwind (1965a, 1965b) hypothesized that a lesion

of the corpus callosum would result in the disconnection of right hemisphere motor

cortex from left hemisphere language processing areas. This would result in the inability

to perform pantomime to command actions with the left hand, while gesture imitation and

actual object use are relatively preserved. This has been interpreted as a disconnection

between language areas necessary for comprehension of commands and movement









representations necessary for selecting and programming the appropriate actions

(Heilman & Rothi, 1993).

The patient described by Graff-Radford et al. (1987) was similar to those described

by Geschwind and Kaplan (1962) and Gazzaniga et al. (1967) in that she demonstrated

impaired pantomime to command but relatively spared gesture imitation with the left

hand. Praxis performance in this individual, however, was worst when she held the

object in her hand and attempted to perform the action. The authors explained the

deficits in this individual as resulting from a verbal-motor disconnection.

The patient described by Watson and Heilman (1982) experienced a lesion of the

corpus callosum that was vascular in nature. The anterior extent of the lesion was at the

junction between the genu and body while the posterior one-fourth to one-fifth of the

body and splenium as well as the supplementary and cingulate cortex remained intact. In

the course of recovery from the lesion, this patient demonstrated conceptual, ideomotor,

and verbal-motor dissociation apraxia. Initially, the patient was unable to pantomime to

command, imitate gestures or use objects with the left hand and could not demonstrate

the intent of actions. Because she was unable to demonstrate that she understood the

intent of the required action, this was considered evidence of conceptual apraxia. During

the course of recovery, the ability to imitate gestures and use objects improved (although

spatiotemporal movement errors were present), but pantomime to command with the left

hand remained impaired. Because the patient was able to imitate gestures and use

objects, this was considered evidence of a verbal-motor dissociation apraxia. Finally,

praxis testing showed that pantomime to command, gesture imitation and object use

improved with the left hand and the patient demonstrated the correct intent of actions but









continued to make spatiotemporal movement errors. Because praxis performance

remained impaired for all tasks with the left hand, this was considered evidence of

persistent ideomotor apraxia.

Furthermore, DeRenzi et al. (1982) also described individuals who demonstrated

modality-specific apraxias. These individuals performed better with certain input

modalities (visual, verbal, or tactile). For example, six participants performed better with

tactile and visual input than with verbal input and six patients performed better with

verbal and tactile input than with visual input (this could not be attributed to receptive

aphasia or visual agnosia). Two participants who performed more poorly with tactile

input than verbal or visual input were also reported. To explain this disconnection

between modality-specific input pathways and the center where movements are

programmed, it has been stated that

in the majority of patients the lesion will result in apraxia appearing in every
modality, either because it destroys the programming center or because it interrupts
all the pathways connecting it with other sensory or motor areas; however, a
discrete injury may well isolate the programming center from one type of
information and render the patient unable to execute the gesture when it is elicited
by a given sensory center but capable of performing it under the guidance of other
modalities. (DeRenzi et al., 1982, p. 310).

As evidenced by patients described by Gazzaniga et al. (1967), Geschwind and

Kaplan (1962), and Graff-Radford et al. (1987), disruption of the transfer of information

from left hemisphere language processing centers and praxis movement representations

from right hemisphere motor areas that control the left hand results in verbal-motor

dissociation apraxia. Furthermore, the patient described by Watson and Heilman (1983)

demonstrated that conceptual and ideomotor forms of apraxia are possible from a lesion

of the corpus callosum. Finally, DeRenzi et al. (1982) provided evidence that apraxia can

be modality-specific.









The literature from individuals with callosal lesions has provided evidence that

several different types of praxis information are transferred across the corpus callosum

for skilled movements of the left hand. Movement representations that are translated into

innervatory patterns and action semantics information that guides the selection of the

appropriate action must be transferred from left hemisphere praxis areas to right

hemisphere motor areas across the corpus callosum. In addition, the input modality (such

as verbal input), must interact with both praxis movement representations and action

semantics for the production of skilled movement.

As mentioned previously, unilateral stroke does not provide a favorable model for

studying bimanual praxis mechanisms due to the presence of hemiplegia. The literature

described above has utilized individuals with callosal lesions to investigate

interhemispheric transfer of praxis information at the single case level of evidence.

Because callosal lesions are rare and heterogeneous, this population also does not provide

the best model for studying the mechanisms of praxis information transfer at the clinical

trial level of evidence. This study proposes to use individuals with AD to study

interhemispheric transfer of praxis information because this population can perform

praxis tasks with both hands (unlike individuals with unilateral strokes), this disease is

prevalent among the elderly population (unlike specific callosal lesions), and there is

evidence of callosal atrophy in the areas of the corpus callosum that are suspected to

carry praxis information (similar to individuals with callosal lesions). Following is an

overview of the symptoms, diagnosis, and pathology of AD, a summary of what is known

about limb apraxia in AD and a review of the literature regarding callosal atrophy in AD.









This will lead to a rationale for studying interhemispheric transfer of praxis information

using praxis asymmetries in individuals with AD as a pathological model.

What is Alzheimer's Disease?

Alzheimer's disease (AD) is a degenerative disease of the central nervous system

(Boller & Duyckaerts, 1997). It is characterized clinically by progressive dementia and

cognitive decline and histologically by senile placques and neurofibrillary tangles. There

are many factors that are thought to contribute to the development of AD; advanced age,

genetic predisposition, the presence of the apolipoprotein E4 allele, gender (female/male

ratio, 2:1), low education level and previous head trauma have been implicated (Barclay,

Zemcov, Blass, & Sanson, 1985; Rocca, Bonaiuto, Lippi, Luciani, Turtu, Cavarzeran, &

Amaducci, 1990).

A clinical diagnosis of AD requires the presence of dementia, cognitive decline and

functional impairment. The Diagnostic and Statistical Manual of Mental Disorders

(DSM-IV) and the National Institute of Neurological and Communicative Disorders and

Stroke (NINCDS) and the Alzheimers Disease and Related Disorders Association

(ADRDA) (NINCDS/ADRDA) have published criteria for the diagnosis of AD (Table 1-

1) (McKhann, Drachman, Folstein, Katzman, Price, Stadlan, 1984). The DSM-IV

defines dementia as "the development of multiple cognitive deficits that include memory

impairment and at least one of the following: aphasia, apraxia, agnosia or a disturbance in

executive functioning." According to the NINCDS/ADRDA criteria a diagnosis of

probable AD requires the following criteria: dementia established by neurologic

examination and documented by objective testing, deficits in two or more cognitive areas,

progressive worsening of memory and other cognitive functions, no disturbance in

consciousness, absence of systemic disorders or other brain diseases that could account









for the progressive deficits in memory and cognition, and onset between 40 and 90 years

of age. The diagnosis of probable AD is supported by progressive deficits in language

(aphasia), perception (agnosia), and motor skills apraxiaa), impaired activities of daily

living and altered patterns of behavior, family history of similar disorders, and consistent

laboratory results. Possible AD is diagnosed when the patient has a variation in the

typical presentation of dementia or when another potentially dementing disorder is

present but is not the primary source of the dementia symptoms. Definite AD is reserved

for clinically diagnosed patients with histopathological confirmation by cerebral biopsy

or autopsy.

Histopathologic evidence of AD as confirmed by cerebral biopsy or postmortem

autopsy requires the presence of senile placques and neurofibrillary tangles. Senile

placques (SP) are extracellular amyloid deposits extracellularr byproducts of neuronal

degeneration) (Afifi & Bergman, 1998; Guilmette, 1997). Neurofibrillary tangles (NFT)

are intracellular aggregates of cytoskeletal filaments (tangles of fine fibers found in cell

bodies) (Afifi & Bergman, 1998; Guilmette, 1997). The NFTs represent the

accumulation of abnormal components of the neuronal cytoskeleton that form paired

helical filaments (Hof& Morrison, 1999). In individuals with AD, SPs and NFTs are

morphologically and topographically distinct, have different histological compositions,

and are present in specific cortical areas and layers. Specifically, pyramidal neurons in

Layer II and III of the cortex project to other ipsilateral and contralateral cortical areas,

respectively, via intra- and inter- hemispheric projection fibers including the corpus

callosum. In the cortex, SPs and NFTs are found in all cortical areas and are numerous

in layer III of the cortex. So the presence of SPs and NFTs in layer III could potentially









disrupt interhemispheric transfer of neuronal signals. Therefore, cognitive functions that

require interhemispheric transfer of information across the corpus callosum, like praxis,

could possibly be impaired in this patient population.

Clinically, patients with AD typically progress through three stages of the disease

process (Boller & Duyckaerts, 1997). As the individual with AD progresses through

these stages, significant cognitive decline occurs resulting in decreased functional

independence. The first stage, amnestic, is characterized by semantic and episodic

memory impairments and the presence of aphasia. The second stage, dementia, involves

a progressive decline in intellectual abilities that significantly impacts the ability to live

independently. The third stage, vegetative, is characterized by the inability to perform

activities of daily living as well as an inability to express wants and needs through

communication. During the second and third stages, memory and language become more

impaired, significantly impacting the individual's ability to communicate and remember.

The person is not able to understand verbal instructions or communicate basic needs and

may become disoriented in familiar places and unable to recognize familiar people.

Additionally, individuals at these stages of AD may demonstrate ideomotor, ideational,

conceptual and constructional apraxia, which interfere with the ability to manipulate tools

and objects in the environment. Individuals with AD experience significant cognitive

decline, decreased functional independence, and the need for more supervised care,

which ultimately increase the costs of their care.

Cognitive decline has a significant impact on impairment of functional abilities in

individuals with AD. Functional impairment in individuals with AD, evidenced by the

loss of the ability to perform activities of daily living (ADLs) and instrumental activities









of daily living (IADLs), has a major impact on the quality of life of patients and

caregivers and is an important predictor of institutionalization (Canadian Study of Health

and Aging, 1994). The presence of amnesia, aphasia, apraxia, agnosia, and/or

visuospatial impairments contribute to functional disability and functional disability

contributes to dependence, which in late stages of AD may result in institutionalization

(Tekin, Fairbanks, O'Connor, Rosenberg, & Cummings, 2001). It is beneficial to health

care providers to understand both the neurobehavioral mechanisms and clinical

implications of cognitive deficits, like limb apraxia, in order to understand how these

cognitive deficits, especially limb apraxia, interfere with the ability to function

independently in individuals with AD.

What is Known About Limb Apraxia in AD?

Limb apraxia is prevalent in all stages of the disease process in individuals with

dementia and the presence of limb apraxia has been demonstrated to have a significant

impact on functional abilities. In a study by Edwards, Deuel, Baum, and Morris (1991),

22% of subjects with suspected dementia, 47.1% of patients with mild dementia, 58.6%

of participants with moderate dementia, and 98.1% of individuals with severe dementia

demonstrated evidence of limb apraxia. Several studies have shown that limb apraxia has

a significant impact on the ability to perform activities of daily living (Foundas,

Macauley, Raymer, Maher, Heilman, & Rothi, 1995; Giaquinto, Buzzelli, DiFrancesco,

Lottarini, Montenero, Tonin, & Nolfe, 1999; Saeki, Ogata, Okubo, Takahashi, &

Hoshuyama, 1995). In a study of individuals with AD (Cho, Cho, Cho, Choi, Oh, & Bae,

2001), 56.6% of participants were dependent for one or more ADLs including bathing

(54.7%), dressing (47.2%), and feeding (5.7%), and for IADLs patients with AD

demonstrated dependence in cooking (66.0%), cleaning (64.2%), housework (79.2%),









and laundry (71.7%), all of which require skilled movement and the ability to manipulate

tools and objects (praxis).

There is also evidence that individuals with AD who develop apraxia may decline

more rapidly (Yesavage, Brooks, Taylor, Tinklenberg, 1993) and that apraxia may be

more predictive of early death than aphasia or amnesia (Burs, Lewis, Jacoby, & Levy,

1991). Furthermore, studies of limb apraxia in acute stroke have shown that the presence

of limb apraxia is a significant predictor of failure to return to work (Saeki et al., 1995),

poor functional recovery following stroke (Giaquinto et al., 1999), and poor performance

of IADLs (Foundas et al., 1995).

There is significant evidence that the presence of limb apraxia has an impact on

functional abilities in individuals with AD (Cho et al., 2001; Foundas et al., 1995;

Giaquinto et al., 1999; Saeki et al., 1995). Functional impairment has been shown to be a

predictor of institutionalization and thus increases the costs of care for individuals with

AD. A review of investigations of limb apraxia in individuals with AD indicated the

presence of 3 types of apraxia: ideomotor apraxia, ideational apraxia, and conceptual

apraxia.

Ideomotor apraxia has been reported frequently as a cognitive sequela of AD (Della

Sala, Lucchelli, & Spinnler, 1987; Derouesne, Lagha-Pierucci, Thibault, Baudouin-

Madec, Lacomblez, 2000; Foundas et al., 1999; Giannakopoulos, Duc, Gold, Hof,

Michel, & Bouras, 1998; Jacobs, Adair, Williamson, Na, Gold, Foundas, Shuren, Cibula,

& Heilman, 1999; Kato, Meguro, Sato, Shimada, Yamazaki, Saito, Yamaguchi, &

Yamadori, 2000; Rapcsak, Croswell, & Rubens, 1989; Travniczek-Marterer, Danielczyk,

Simanyi, & Fischer, 1993; Willis, Behrens, Mack, & Chui, 1998). According to recent









literature, individuals with AD demonstrate impaired performance with the dominant

(right) hand on both gesture to verbal command and imitation tasks (Travniczek-Marterer

et al., 1993). The severity of dementia has an impact on praxis performance in

individuals with AD (Foundas et al., 1999) and praxis performance degrades with the

progression of AD (Della Sala et al., 1987). With regards to error types, individuals with

AD produce more content (100%) than spatial-temporal (0%) errors with intransitive

pantomimes and more spatial-temporal (96%) than content (4%) errors with transitive

pantomimes when the dominant (right) hand is tested (Foundas et al., 1999). Patients

with AD also produce significantly more body-part-as-tool responses with the right hand

when compared with normal controls (Kato et al., 2000). The current literature on limb

apraxia in AD has examined ideomotor and conceptual apraxia in the dominant hand

only. None of the previously published studies have examined the left hand performance

of individuals with AD on praxis production or conceptual tasks. This study proposes to

investigate the mechanisms of left hand praxis performance in right handed individuals

with AD.

It should also be noted that the majority of studies examining ideomotor apraxia in

AD have not typically utilized a standardized battery for the assessment of praxis.

Furthermore, most of the studies scored responses as either correct or incorrect and did

not analyze error types. Therefore, a clinically efficient and standardized praxis measure

might be helpful in the assessment of individuals with AD.

Studies of conceptual apraxia in AD have focused on determining the

characteristics of the disorder and attempting to clarify the nature of the semantic system

(Dumont, Ska, & Joanette, 2000; Ochipa et al., 1992; Schwartz et al., 2000). Thus far 54









individuals with AD have been tested for conceptual apraxia by various authors (Dumont

et al., 2000; Ochipa et al., 1992; Schwartz et al., 2000) and 50/52 (96%) participants were

found to have deficits of the praxis conceptual system.

Ochipa et al. (1992) hypothesized that there could be three types of conceptual

apraxia in individuals with AD due to disruptions of different cognitive mechanisms of

the praxis conceptual system. First, there may be a loss of knowledge of the type of

actions associated with tools or objects (tool-object action knowledge) resulting in

content errors in tool use. Second, there may be an inability to associate tools with the

appropriate objects (tool-object associative knowledge) leading to the inappropriate

selection of tools. Finally, there may be impairment in the ability to understand the

mechanical nature of problems and the mechanical advantages of particular tools

(mechanical knowledge) leading to an inability to solve mechanical problems and an

inability to develop novel tools. In this study (Ochipa et al., 1992), the 32 participants

with AD were divided into four experimental subgroups: good ideomotor praxis without

semantic language impairment, poor ideomotor praxis without semantic language

impairment, good ideomotor praxis with semantic language impairment, poor ideomotor

praxis with semantic language impairment. Each element of the praxis conceptual

system mentioned above was tested in the patients and controls. The results indicated

that individuals with AD have an impairment of the praxis conceptual system and that

conceptual apraxia can be differentiated from both ideomotor apraxia and semantic

language deficits. Additionally, AD participants were significantly impaired in all three

proposed domains of the praxis conceptual system (tool-object action knowledge, tool-

object associative knowledge, and mechanical knowledge) so it is not known if these









three components of the praxis conceptual system are functionally or neurologically

distinct.

Rapcsak et al. (1989) examined ideational apraxia in individuals with AD by

testing serial actions requiring the use of several objects to achieve an intended goal (i.e.

prepare a cup of instant coffee with cream and sugar). The serial actions were scored by

counting the number of component actions correctly executed in the appropriate

sequence. When compared to controls, participants with AD were significantly impaired

on measures of ideational apraxia.

Why Study Disconnection Apraxia in AD?

Individuals with AD loose pyramidal neurons from layer III of the cortex that

project to the corpus callosum from analogous areas of the contralateral hemisphere.

Therefore, in addition to the presence of ideomotor, ideational, and conceptual apraxia

with the dominant hand, it is likely that patients with AD will demonstrate

interhemispheric disconnection syndromes that include the presence of ideomotor,

conceptual, and verbal-motor dissociation apraxias with the nondominant hand.

Several studies have found cortical atrophy in the temporal and parietal lobes in

individuals with AD (Foundas, Eure, & Seltzer, 1996; Halliday, Double, & Macdonald,

2003; Pantel, Schonknecht, Essig, & Schroder, 2004; Thompson, Hayashi, Zubicaray,

Janke, Rose, Semple, Herman, Hong, Dittmer, Doddrell, & Toga, 2003; Thompson,

Mega, Woods, Zoumalan, Lindshield, Blanton, Moussai, Holmes, Cummings, & Toga,

2001). Atrophy of these regions correlates with the cognitive symptoms that are seen in

the early stages of AD (i.e. apraxia and aphasia). Of interest in this study is atrophy of

areas that are critical to spoken language processing and praxis movement

representations. Thompson et al. (2003) found highly significant decreases in gray matter









in bilateral temporal and parietal cortices and that atrophy of these regions was

asymmetric with greater atrophy of left hemisphere as compared to the right hemisphere.

Additionally, the precentral and postcentral gyri (important for execution of movement

and perception of movement) were relatively spared compared with the parietal

association cortex located immediately posterior. Pantel et al. (2004) also found a

significant decrease in temporal and parietal cortical volume bilaterally and showed a

correlation between left temporal and parietal volumes and performance on tests of

naming and praxis. Thompson et al. (2001) noted relative sparing of occipital cortex

bilaterally suggesting preserved processing of visual sensory input in individuals with

dementia. These findings explain the presence of apraxia in the dominant hand of right

handed individuals with AD but would not be sufficient to explain a right hand to left

hand asymmetry in praxis performance.

Studies that have measured the corpus callosum in individuals with AD have found

atrophy in specific regions (Janowsky, Kaye, & Carper, 1996; Lyoo, Satlin, Lee, &

Renshaw, 1997; Pantel, Schroder, Jauss, Essig, Minakaran, Schonknecht, Schneider,

Schad, Knopp, 1999; Teipel, Hampel, Alexander, Schapiro, Horwitz, Teichberg, Daley,

Hippius, Moller, & Rapoport, 1998; Vermersch, Roche, Hamon, Daems-Monpeurt,

Pruvo, Dewailly, & Petit, 1996; Vermersch, Scheltens, Barkhof, Steinling, & Leys, 1993;

Weis, Jellinger, & Wenger, 1991). However, these reports have yielded conflicting

results regarding which areas of the corpus callosum are decreased in AD. Several

studies have reported a reduction in the total area of the corpus callosum in individuals

with AD as compared to normal controls (Biegon, Eberling, Richardson, Roos, Wong,

Reed, & Jagust, 1994; Black, Moffat, Yu, Parker, Stanchev, & Bronskill, 2000; Hampel,









Teipel, Alexander, Horwitz, Teichberg, Schapiro, & Rapoport, 1998; Pantel, Schroder,

Essig, Minakaran, Schad, Friedlinger, Jauss, & Knopp, 1998; Teipel, Bayer, Alexander,

Zebuhr, Teichberg, Kulic, Schapiro, Moller, Rapoport, & Hampel, 2002; Teipel, Hampel,

Pietrini, Alexander, Horwitz, Daley, Moller, Schapiro, & Rapoport, 1999). Teipel and

colleagues (2002, 1999) reported a significant reduction in the area of the rostrum and

splenium with sparing of the body of the corpus callosum while others have reported

significant reductions in the genu (Biegon et al., 1997; Black et al., 2000) and body

(Lyoo et al., 1997; Black et al., 2000). Hampel et al. (1998) noted decreased area in the

most rostral and most caudal regions of the corpus callosum in patients with AD with no

reduction of the posterior body. Weis et al. (1991) attempted to differentiate callosal

degeneration patterns in normal aging and AD. Results indicated a significant decrease

in the anterior portions (rostrum, genu, anterior body) of the corpus callosum with no

change in the posterior portions (posterior body, isthmus, and genu) in normal aging.

However, in individuals with AD, a significant decrease in the body of the corpus

callosum occurred with no change in the anterior and posterior portions.

Furthermore, the patients described by Kazui and Sawada (1993) and Watson and

Heilman (1983) demonstrated apraxia that was more severe when performing gestures

with the left hand than the right hand due to a lesion of the anterior portion of the body of

the corpus callosum. The case reported by Degos et al. (1987) presented with left apraxia

without agraphia following a lesion of the posterior portion of the body and splenium of

the corpus callosum. This dissociation suggests that callosal fibers for writing are

concentrated in the posterior portion of the corpus callosum while callosal fibers for









praxis are concentrated in the anterior portion of the corpus callosum (Kazui & Sawada,

1993).

Although a systematic investigation of the interhemispheric transfer of praxis

information using AD as a pathological model has not been completed to date, several

studies have reported differences in praxis performance with the right hand (dominant)

versus the left hand (nondominant) in this population (Ball, Lantos, Jackson, Marsden,

Scadding, & Rossor, 1993; Derouesne et al., 2000; Willis et al., 1998). Derouesne et al.

(2000) found that praxis performance was better with the right hand than with the left

hand in patients with AD. Willis et al. (1998) found that while performance accuracy

between the right and left hands was not significantly different, gesture response latencies

were significantly longer for the AD group when the left hand was used. Rapcsak et al.

(1989) found no difference in praxis performance between the right and left hands in

individuals with AD. Furthermore, due to the presence of contralesional hemiplegia,

stroke does not provide an ideal model for studying the praxis abilities of the left and

right hands independently. Therefore, a disease process which affects the fibers of the

corpus callosum that transfer praxis information across the hemispheres would provide a

superior model for studying apraxia asymmetries. Because there is evidence of callosal

atrophy in patients with AD, this disease may provide a more useful model for studying

interhemispheric transfer of praxis information.

Summary

Alzheimer's disease (AD) is a costly and debilitating condition. It causes

numerous cognitive and behavioral impairments including limb apraxia. Limb apraxia is

a disorder that disrupts skilled movements of the arms and hands and has a negative

affect on the performance of activities of daily living. Thus far, the mechanisms of limb









apraxia have primarily been studied in groups of individuals with unilateral strokes while

the mechanisms of interhemispheric transfer of praxis information have primarily been

studied in single cases of individuals with specific callosal lesions. Unilateral stroke does

not provide a good model for the study of bimanual praxis mechanisms because the

presence of hemiplegia in these individuals interferes with the examination of praxis in

the contralesional hand. Callosal lesions provide an excellent model for studying

bimanual praxis mechanisms but these lesions are extremely rare and physiologically

heterogeneous and therefore this population is not well suited for a group study. Perhaps

AD could provide a comparable model for studying the mechanisms of interhemispheric

transfer of praxis information.

It has been shown that individuals with AD demonstrate limb apraxia (like

unilateral stroke patients) and display callosal degeneration (like the callosal lesion

patients). Individuals with AD are able to use both hands to perform praxis tasks and AD

is a fairly common diagnosis within the elderly population. For these reasons, AD is

presented as a potentially superior model for studying interhemispheric transfer of praxis

information; specifically production, conceptual, and verbal-motor praxis information.

The corpus callosum is responsible for transferring information from one cerebral

hemisphere to the other. The movement representations that govern skilled movement

and the semantic representations that relate sensory input to motor output are thought to

be localized in the left hemisphere. In order for the left hand to correctly perform skilled

movements praxis representations and action semantics in the left hemisphere must be

transferred to right motor cortex via the corpus callosum. Therefore, if individuals with

AD have degeneration of the corpus callosum fibers that transfer praxis information from









the right hemisphere to the left hemisphere and if individuals with AD have been shown

to have different types of limb apraxia, it can be hypothesized that individuals with AD

will demonstrate disconnection apraxia of the ideomotor, conceptual, and verbal-motor

types (i.e. better performance of praxis tasks with the right hand than the left hand).

Purpose, Questions, and Hypotheses

The purpose of the present study is to investigate how praxis information

processing is represented in the brain by examining the transfer of different types of

praxis information from praxis movement representations in the left hemisphere to motor

cortex in the right hemisphere across the corpus callosum. This will be accomplished by

examining bimanual praxis performance in individuals with AD because individuals in

this population can perform praxis tasks with both hands (i.e. they do not have

hemiplegia), they are prevalent within the elderly population (i.e. this is not a rare

syndrome), and they demonstrate both limb apraxia and callosal atrophy (i.e. can

potentially differentiate what type of information is being transferred via the corpus

callosum).

First, it will be necessary to confirm that individuals with AD demonstrate limb

apraxia. Second, this study will attempt to determine whether the limb apraxia that is

present in individuals with AD is due to the degradation of left hemisphere movement

representations or the interruption of interhemispheric transfer of praxis information

across the corpus callosum. Third, examination of the transfer of praxis conceptual and

production information will provide information about what types of praxis information

are dissociated due to the degradation of callosal fibers in individuals with AD. Finally,

this study will attempt to differentiate whether the interhemispheric disconnection of

praxis production information is due to the inability to transfer verbal or motor









information across the corpus callosum in individuals with AD. The following research

questions will address each of these issues.

Research Question 1

Do individuals with AD have conceptual and/or ideomotor apraxia in the left hand?

Hypothesis. This research question will be examined by comparing the left hand

performance of individuals with AD to the left hand performance of healthy elderly

individuals on a verbal command pantomime task and a conceptual pantomime task.

Previous studies provide evidence that individuals with AD have conceptual and

ideomotor apraxia in the dominant (right) hand but it is also necessary to examine the

presence of conceptual and ideomotor apraxia in the nondominant (left) hand. If there

were a significant difference between individuals with AD and healthy elderly

individuals on the verbal command task (left hand), this would suggest the presence of

ideomotor apraxia in the AD group. It is predicted that there will be a significant

difference between the two groups for left hand performance on the verbal command

pantomime task (i.e., individuals with AD will demonstrate ideomotor apraxia in the left

hand). If there were a significant difference between individuals with AD and healthy

elderly individuals on the conceptual pantomime task (left hand), this would suggest the

presence of conceptual apraxia in the AD group. It is predicted that there will be a

significant difference between the two groups for left hand performance on the

conceptual pantomime task (i.e., individuals with AD will demonstrate conceptual

apraxia in the left hand). Only left hand performance is being compared to answer this

question because left hand performance requires the recruitment of both left hemisphere

praxis movement representations and right hemisphere motor areas which requires the

transfer of praxis movement information across the corpus callosum. Further questions









will address the contributions of degraded movement representations and

interhemispheric callosal disconnection to the apraxia in individuals with AD.

Research Question 2

What is the contribution of degraded praxis movement and conceptual

representations (due to cortical atrophy) to the limb apraxia in individuals with AD?

Hypothesis. This issue will be examined by comparing the right hand performance

of individuals with AD to the right hand performance of healthy elderly individuals on a

verbal command pantomime task and a conceptual pantomime task. If there is a

significant difference between individuals with AD and healthy elderly individuals on the

verbal command task (right hand), it can be assumed that the praxis movement

representations are degraded in individuals with AD. The prediction is that there will be

a significant difference between the two groups for right hand performance on the verbal

command pantomime task (i.e., there will be evidence of degraded movement

representations in individuals with AD). If there is a significant difference between

individuals with AD and healthy elderly individuals on the conceptual pantomime task

(right hand), it can be assumed that the praxis conceptual representations are degraded in

individuals with AD. The prediction is that there will be a significant difference between

the two groups for right hand performance on the conceptual pantomime task (i.e., there

will be evidence of degraded praxis conceptual representations in individuals with AD).

The comparison of right hand performance answers this question because right hand

performance does not require the transfer of praxis information across the corpus

callosum but requires within hemisphere access to praxis movement representations.

Further questions will address the role of interhemispheric disconnection in the transfer









of different types of praxis information across the corpus callosum in individuals with

AD.

Research Question 3

What is the contribution of interhemispheric disconnection (due to callosal atrophy)

to the limb apraxia in individuals with AD?

Hypothesis. The disparity or asymmetry between right hand and left hand

performance of individuals with AD and healthy elderly individuals on praxis production

and conceptual tasks will be compared to answer this research question. If the

performance asymmetry of the two groups on the conceptual pantomime task is

significantly different, the conclusion would be that praxis conceptual information is not

being transferred across the corpus callosum in individuals with AD. The prediction is

that performance asymmetry of the two groups on the conceptual pantomime task will

not be significantly different (i.e. there will not be evidence of a callosal disconnection

that is specific to praxis conceptual information in individuals with AD). If the

performance asymmetry of the two groups on the verbal command pantomime task and

the pantomime imitation task is significantly different, the conclusion would be that

information from praxis movement representations is not being transferred across the

corpus callosum in individuals with AD. The prediction is that performance asymmetry

of the two groups on the verbal command pantomime and pantomime imitation tasks will

be significantly different (i.e., there will be evidence of a callosal disconnection that is

specific to praxis movement information in individuals with AD). Because the verbal

command pantomime task requires transfer of language and motor information, it is

necessary to attempt to differentiate whether verbal information or motor information is

being interrupted by the proposed callosal disconnection in individuals with AD.









Research Question 4

Is the disruption of praxis information transfer a result of an intrahemispheric

verbal-motor disconnection or an interhemispheric corpus callosum disconnection?

Hypothesis. Answering this question will involve two comparisons. First, right

hand performance of both groups on a verbal command pantomime task and a pantomime

imitation task will be compared. If there were a significant difference between the two

groups for right hand performance on these two tasks, this would suggest that impaired

performance of individuals with AD results from an intrahemispheric verbal motor

disconnection. If there were not a significant difference between right hand performance

of the two groups on these two tasks, this would suggest that impaired performance of

individuals with AD results from an interhemispheric callosal disconnection. Second, the

asymmetry between the right and left hand performance of the experimental group will

be compared. If verbal command pantomime performance were more asymmetric (right

hand performance greater than left hand performance), this would provide evidence that

verbal input interferes with the transfer of praxis movement representations across the

corpus callosum in individuals with AD. If pantomime imitation performance were

more asymmetric (right hand performance greater than left hand performance), this

would provide evidence that deficient transfer of praxis information is specific to the

transfer of movement information across the corpus callosum in individuals with AD. It

is predicted that there will be evidence of an interhemispheric callosal disconnection in

individuals with AD that is specific to the transfer of information from praxis movement

representations.









Table 1-1: Diagnostic criteria for AD.
DSM-IV: dementia Alzheimer type
Development of multiple cognitive deficits:
Memory impairment
At least one of the following:
Aphasia
Apraxia
Agnosia
Disturbed executive functioning (planning, organizing, sequencing, abstracting)
Course characterized by continued gradual cognitive and functional decline
Deficits sufficient to interfere significantly in social and occupational functioning and
representing a decline from past functioning
Other causes of dementia excluded (medical, neurologic, psychiatric)
NINCDS-ADRDA: probable Alzheimer disease

Dementia established by examination and documented by objective testing
Deficits in two or more cognitive areas
Progressive worsening of memory and other cognitive functions
No disturbance in consciousness
Onset between 40 and 90 years of age
Absence of systemic disorders or other brain disease that could account for the
progressive deficits in memory and cognition
Diagnosis supported by:
Progressive deficits in language (aphasia), perception (agnosia),
and motor skills apraxiaa)
Impaired activities of daily living and altered patterns of behavior
Family history of similar disorders
Consistent lab results
Morris, J.C. (1999). Clinical presentation and course of Alzheimer disease. In R.D.
Terry, R. Katzman, K.L. Bick, & S.S. Sisodia (Eds.), Alzheimer disease (2nd ed)
(pp. 11-24). Philadelphia: Lippincott, Williams, & Wilkins.





























Figure 1-1: Cognitive neuropsychological model of limb apraxia.
Rothi, L.J.G., Ochipa, C., & Heilman, K.M. (1997a). A cognitive neuropsychological
model of limb praxis and apraxia. In L.J.G. Rothi & K.M. Heilman (Eds.),
Apraxia: the neuropsychology of action (pp.29-49). East Sussex, UK:
Psychology Press.














CHAPTER 2
METHODS

The goals of this study are to determine whether praxis information is transferred

from left hemisphere movement representations to right hemisphere motor areas via the

corpus callosum and to examine what types of praxis information are transferred using

this neural pathway. This study proposed to use Alzheimer's disease as a model for

investigating the interhemispheric transfer of praxis information. The following sections

describe the methods for answering the proposed research questions.

Subjects

Two groups of participants were recruited for participation in this study. A group

of healthy elderly control subjects (HC) and a group of individuals with Alzheimer's

disease (AD) participated in the study.

Inclusion Criteria

Inclusion criteria consisted of: 1) for the AD group, a medical diagnosis of AD with

no history of other neurologic disease (i.e. stroke, tumors, TBI, seizures, etc.) and for the

HC group, no history of neurologic disease, 2) no history of upper extremity mobility

problems, severe hearing loss or severe visual impairment, 3) no history of drug or

alcohol abuse by self-report, caregiver report and/or medical record (exclude participants

who have experienced alcohol or drug abuse related disease or social or vocational

interference as a result of alcohol or drug use), 4) no history of psychiatric problems by

self-report, caregiver report and/or medical records (exclude participants who have been

hospitalized for psychiatric illness) 5) because the experimental stimuli involve object









recognition to perform pantomimes, absence of visual object agnosia, as measured using

the Associative Match subtest of the Birmingham Object Recognition Battery (Riddoch

& Humphreys, 1993) (i.e., for the AD group, at least a score of 21/30 or 70% accuracy

and for the HC group, at least a score of 27/30 or 90% accuracy), 6) because the

experimental stimuli require processing of verbal commands to perform pantomimes,

absence of severe auditory comprehension deficits, as measured using the Sequential

Commands subtest of the Western Aphasia Battery (Kertesz, 1982) (i.e., for the AD

group, at least a score of 40/80 or 50% accuracy and for the HC group at least a score of

72/80 or 90% accuracy), 7) English as native language per self-report or caregiver report,

8) right handed (determined by the Waterloo Handedness Questionnaire).

The participants with AD were required to provide documentation of a medical

diagnosis of AD. All participants with AD also met the DSM-IV / NIDCD/ADRDA

criteria for probable AD. Scores from the Mini Mental State Exam (MMSE) (Folstein,

Folstein, & McHugh, 1975; Tombaugh & McIntyre, 1992) were used to verify the

presence of dementia in the AD group and to group the participants with AD by severity

level (a score of less than 27/30 was considered impaired). A shortened version of the

Boston Naming Test (Fastenau, Denburg, & Mauer, 1998; Kaplan, Goodglass, &

Weintraub, 1983) was administered to verify cognitive deficits in the AD group and to

verify normal naming function in the HC group. Short form 3 from Fastenau et al. (1998)

was used in this study. The correlation of this form with the original version of the BNT

was r = 0.69 which was significant at the p < 0.005 level. In the Fastenau et al. (1998)

study, the total sample scored a mean of 13.6 (SD=1.3) on this version. When the

performance of the total sample of healthy older adults was broken down by age, the









following results were reported: 57-68 years (n=35) the mean was 14.3 (SD=0.8), 69-76

years (n=38) the mean was 13.3 (SD=1.3), and 77-85 years (n=35) the mean was 13.3

(SD=1.5). As this study has received Institutional Review Board approval (IRB # 166-

02), each participant signed an Informed Consent Form.

Subject Demographics

Twenty-two (see sample size estimation below) individuals who had been

diagnosed with AD, were recruited for this study from the University of Florida Memory

Disorder Clinic (UFMDC) and the surrounding community. The individuals with AD

who were recruited from the UFMDC participated in a neuropsychological assessment, a

neurologic exam, and a physical exam prior to being enrolled in the study. The

participants who were recruited from the community were required to provide

documentation of a medical diagnosis of AD from a physician. Only patients who met

the DSM-IV/NINCDS-ADRDA criteria for a diagnosis of probable AD were enrolled in

the study as experimental subjects.

The AD group consisted of 14 women and 8 men with an age range of 61-90 years,

mean age of 79.23 years (SD = 6.4 years), and a mean education level of 13.59 years (SD

= 2.6 years). Of the individuals with AD enrolled in the study, 17 completed all three

experimental tasks and 5 completed 2 out of 3 tasks due to the inability to comprehend

the instructions for task 3 (Conceptual Pantomime).

In addition to the AD group, a group of 24 healthy elderly control subjects (HC)

was recruited to serve as a comparison group for the performance of the AD participants

on the experimental tasks described below. The HC group was matched with the AD

group for age and gender (see results, Chapter 3). The experimenter attempted to match

the HC and AD groups for education level, however this was not accomplished (see









results, Chapter 3). The HC group consisted of 15 women and 9 men with an age range

of 63-85 years, mean age of 76.10 years (SD = 6.8 years), and mean education level of

15.52 years (SD = 2.4 years).

Sample Size Estimation

Data from normal controls and individuals with AD for performance on the Florida

Action Recall Test (FLART) (Schwartz et al., 2000) were used to estimate the group

means ([t) for right hand and left hand performance of individuals with AD. These data

were chosen because the FLART stimuli were used for all of the apraxia measures in this

study and because there are no published data for group means to use as estimates for the

performance of individuals with AD on the proposed measures. The performance of the

normal controls on the FLART in the Schwartz et al. (2000) study was used as an

estimate for right hand performance of the individuals with AD in this study. The

performance of the experimental group on the FLART in the Schwartz et al. (2000) study

was used as an estimate for left hand performance of the individuals with AD in this

study. Because, left hand performance was expected to be significantly more apraxic

than right hand performance, the control group was used to estimate right hand

performance of the AD group. UCLA Department of Statistics Power Calculator

(http://calculators.stat.ucla.edu/powercalc/) was used to perform the following sample

size estimates.

Sample Size Estimation-2 Samples Equal Variances (most conservative estimate)

The mean for left hand performance was estimated to be 56.9% and the mean for

right hand performance was estimated to be 86.9% (Schwartz et al., 2000). Standard

deviation for both hands was estimated to be 25%, which is the most conservative

estimate of variance (Marks, 1999). A two sided hypothesis was proposed, Ho: ptA = |N









and Ha: ptA # [[N. Significance level (c) was set at 0.05 and Power was set at 0.80.

With the aforementioned parameter estimates, sample size was estimated to be n = 14 for

left hand performance and n = 14 for right hand performance, resulting in a total n=28 for

the AD group sample size estimate.

Sample Size Estimation-2 Samples Unequal Variances (least conservative estimate)

The mean for left hand performance was estimated to be 56.9% and the mean for

right hand performance was estimated to be 86.9% (Schwartz et al., 2000). Standard

deviation for left hand performance was estimated to be 17.8% and standard deviation for

right hand performance was estimated to be 7.6% (Schwartz et al., 2000), which is the

least conservative estimate of variance (Marks, 1999). A two sided hypothesis was

proposed, Ho: ptA = [N and Ha: ptA # [[N. Significance level (U) was set at 0.05 and

Power was set at 0.80. With the aforementioned parameter estimates, sample size was

estimated to be n = 7 for left hand performance and n = 3 for right hand performance,

resulting in a total n=10 for the AD group sample size estimate.

It was decided that a reasonable sample size for the AD group would be

approximately 19 subjects as this is a compromise between the most conservative

estimate and the least conservative estimate. However, a total of 22 participants were

enrolled in the study.

Experimental Tasks

Data Collection Procedures

All subjects with AD produced each stimulus item of the following tasks with the

right hand and left hand. Right hand and left hand performance within a task was

randomized across subjects in both groups. To balance for order effects tasks were

presented in random order and task order was counterbalanced for all of the subjects.









Each session was videotaped and analyzed offline for correct or incorrect performance

and error types as described by Rothi et al. (1997b). The conceptual apraxia measure

(task 3) was scored according to the criteria set forth by Schwartz et al. (2000) (i.e. each

item will be scored based on the concept conveyed by each pantomime regardless of the

quality of the movement itself).

All of the apraxia tasks consisted of 45 items. These were the stimulus items that

are included in the stimuli from the Florida Action Recall Test (FLART) (see below,

conceptual pantomime task). These stimuli were chosen in order to provide consistency

across tasks for statistical comparison. For example, if the FLART shows a picture of a

lock on a door knob and the participant is required to pantomime key, the patient will

also be required to imitate the pantomime for "key" produced by the examiner

(pantomime imitation) and to pantomime to verbal command "Show me how you hold

and use a key to open a door" (verbal command pantomime). Within each task, each

stimulus item was performed once with the right hand and once with the left hand in

random order.

Task 1: Verbal Command Pantomime (VC)

The verbal command task was used to investigate the role of production

information in praxis processing and the contribution of verbal input to the transfer of

praxis information across the corpus callosum.

Task 1 Procedures. The examiner provided the participant with the following

instructions: "I am going to ask you to pretend to use different tools. I want you to show

me how you would use each tool if you were actually holding the tool in your hand and

using it. I am going to ask you to use either your left hand or your right hand. Listen for

this cue and use only the hand I ask you to use." The examiner presented the subject with









a verbal command for each of the stimuli and the subject performed pantomimes to

verbal command with each hand. See Appendix A for verbal command stimuli.

Task 1 Scoring. Two independent raters were trained (see Rater Training, next

section) to score each of the experimental tasks. Step 1 of the scoring process was to

judge the accuracy of each individual response according to the target stimulus and

experimental task being scored. A correct production received a score of 1 while an

incorrect production received a score of 0. For task 1 (VC), each production was scored

as correct (1) or incorrect (0) according to the semantic content of the production and the

spatial and temporal aspects of the movement. For example, if the target stimulus was

scissors, the participant was required to produce a pantomime for scissors that correctly

represented the semantic content and spatiotemoral specifications of the movement for

using scissors. A production was considered correct if it did not contain any error types.

Step 2 of the scoring process was to determine whether an incorrect response was

recognizable for the target stimulus. If the production was deemed unrecognizable for

the target stimulus, no further categorization of error types was conducted. If the

production was deemed recognizable for the target stimulus but contained praxis errors,

each error was categorized into one or more of the error types described in Appendix B.

Task 2: Pantomime Imitation (PI)

The pantomime imitation task was used to investigate the role of production

information in praxis processing and the contribution of verbal input to the transfer of

praxis information across the corpus callosum.

Task 2 Procedures. The examiner provided the participant with the following

instructions: "I am going to make a movement with my hand and you are going to try to

copy my movement. I want you to watch me and wait until my movement is completely









finished before you move your hand. I am going to ask you to use either your right hand

or your left hand. Listen for this cues and only use the hand I ask you to use." The

examiner pantomimed each of the stimuli in random order and with each hand for the

subject to imitate and the subject imitated the gestures produced by the clinician with

each hand. See Appendix A for pantomime imitation stimuli.

Task 2 Scoring. Two independent raters were trained (see Rater Training, next

section) to score each of the experimental tasks. Step 1 of the scoring process was to

judge the accuracy of each individual response according to the target stimulus and

experimental task being scored. A correct production received a score of 1 while an

incorrect production received a score of 0. For task 2 (PI), each production was scored as

correct (1) or incorrect (0) according to the semantic content of the production and the

spatial and temporal aspects of the movement. For example, if the target stimulus was

scissors, the participant was required to imitate exactly both the semantic content and

spatiotemporal aspects of the movement for scissors that was produced by the examiner.

A production was considered correct if it did not contain any error types. Step 2 of the

scoring process was to determine whether an incorrect response was recognizable for the

target stimulus. If the production was deemed unrecognizable for the target stimulus, no

further categorization of error types was conducted. If the production was deemed

recognizable for the target stimulus but contained praxis errors, each error was

categorized into one or more of the error types described in Appendix B.

Task 3: Conceptual Pantomime (CP)

The verbal command task was used to investigate the transfer of action semantics

information across the corpus callosum.









Task 3 Procedures. The Florida Action Recall Test (FLART) consists of 45 black

and white line drawings of objects placed in scenes implying an action. The subject is

instructed to imagine what tool is needed to act upon each object or scene and to

pantomime the action associated with that tool in relation to the drawing. For example, a

drawing of an unshaven face requires a shaving action and a drawing of a cooked turkey

requires a carving action. The targeted tool is not shown in the drawing. For this study,

conceptual praxis was tested using the stimuli from the FLART (for examples of stimuli

see Figure 2-1).

The examiner provided the participant with the following instructions: "I am going

to show you some drawings of objects in scenes that imply an action. You must imagine

what tool is needed to act upon object in the picture. Then pretend to do the action

associated with the tool that would be used to act on the object shown. A tool is any item

that can be held in one hand and can be used to act on a pictured object. Tools may

include personal care items, kitchen utensils, household items, garage tools, sports

equipment, or musical instruments. The tool is not shown in the drawing. I will tell you

which hand to use to perform the action. Do not name the tool and do not name the

object. Using your hand to complete the action without the assistance of a tool is

incorrect." See Appendix A for the stimuli used in this task.

Task 3 Scoring. Two independent raters were trained (see Rater Training, next

section) to score each of the experimental tasks described above. Step 1 of the scoring

process was to judge the accuracy of each individual response according to the target

stimulus and experimental task being scored. A correct production received a score of 1

while an incorrect production received a score of 0. For task 3 (CP), each stimulus item









was classified as correct (1) or incorrect (0) based on the semantic content of the

production only (i.e. the presence of a spatial and/or temporal errors) without the

presence of a content error, was not considered an incorrect production). For example,

for stimulus item #45, the pictured object is a paper doll and the target pantomime is

scissors. If the subject pantomimes scissors using their fingers as the blades of the

scissors (BPT error), the semantic content of the production is correct so the production

would be scored as correct. If the subject pantomimes coloring the paper doll with a

crayon (R error), the semantic content of the production is incorrect, so the production

would be scored incorrect. A production was considered correct if it did not contain any

conceptual errors. Step 2 of the scoring process was to determine whether an incorrect

response was recognizable for the target stimulus. If a response was considered

recognizable, it was considered incorrect only if it contained content errors (i.e., hand

error (H), related error (R), nonrelated error (N), or concretization error (C)). Step 3of

the scoring process involved categorizing each content error into one of the content error

types described in Appendix B. The presence of temporal, spatial, or other errors (see

Appendix B) was also noted.

Rater Training

Two independent raters were trained to score the responses of each participant for

each task. Rater 1, the primary rater, scored all of the data for statistical analysis and

scored a percentage of the data again for reliability purposes. Rater 1 was a graduate

student in Occupational Therapy at the University of Florida. Before training she had

extensive experience viewing and scoring videotapes of apraxic research participants but

had little experience in the clinical assessment and treatment of limb apraxia. Rater 2, the

reliability rater, scored a percentage of the data for reliability purposes. Rater 2 was an









undergraduate student in Speech Pathology at the University of Florida. Before training

she was inexperienced in viewing and scoring videotapes of apraxic research participants

but had received some instruction in the clinical assessment and treatment of limb

apraxia. Rater 1 and Rater 2 received extensive training by the experimenter with regards

to judging the accuracy of responses and classification of error types. They both

participated in several sessions (approximately 4 hours) of focused instruction on the

judgment of correct and incorrect response and identification of error types. These

sessions involved viewing several practice tapes (individuals who were not included in

the study producing pantomimes) and attempting to judge accuracy and identify errors

with discussion following each production between the raters and the experimenter.

Following these sessions, the two raters watched several videotapes of pantomime

productions. They were required to score each production independently then discuss

their scoring until they were able to reach 90% agreement onlO consecutive productions.

For reliability, the two raters judged the praxis productions of each participant

independently and were not permitted to confer regarding their judgements.

Reliability

Two independent raters (Rater 1 and Rater 2) analyzed 20% of the data to

determine inter- and intra- rater reliability. Intra-rater reliability: In order to determine

whether scoring of the apraxia tasks was reliable when scored multiple times by the same

judge, 20% of the data were re-scored by Rater 1 who scored the entire data sample for

statistical analysis. Inter-rater reliability: In order to determine whether scoring of the

apraxia tasks was reliable when scored by independent judges, 20% of the data were re-

scored by Rater 2 who scored only a small sample of the data for reliability purposes.

Reliability scoring was completed from a videotape of the original test administration.









In order to describe the reliability of the two raters, percent reliability was

calculated as the number of agreements minus the number of disagreements divided by

the total number of stimuli multiplied by 100 (# of agreements # of disagreements / total

number of stimuli) x 100 = % correct. Although percent agreement reflects the

proportion of agreements among the total number ofjudgments, it does not take into

account the amount of agreement expected by chance (Kramer & Feinstein, 1981).

Therefore, statistical analysis of the reliability data was completed using the k (kappa)

statistic because this is considered the index of choice for measurement of observer

agreement and corrects for agreement expected by chance (Kramer & Feinstein, 1981).

Kappa is ordinarily used to measure the concordance between two observers.

According to Kramer and Feinstein (1981), the magnitude or value of kappa is

more descriptive than the associated p value and they state that "p<.05 is a necessary but

not sufficient criterion for meaningful observer agreement. Therefore, the following

guidelines were suggested (see table 2-4) for the strength of observer agreement.

Statistical Analysis

Before statistical analysis was completed, the data were collapsed within subject.

For the current study, statistical analysis was performed on the response accuracy

variable only. Descriptive data for each error type is provided but statistical analysis of

this data will be reserved for future studies.

For the dependent variable response accuracy, a percentage was calculated for each

participant. For the error types a percentage was calculated and error total represents the

total number of errors. The percentages and averages for the dependent variables were

calculated as follows: percent response accuracy = number of correct responses / total

number of stimuli x 100; percentage of each error type = number of errors present / total









number of errors x 100. In addition, an asymmetry ratio was calculated for response

accuracy for each subject in both groups. The asymmetry ratio was calculated as right

hand performance minus left hand performance divided by right hand performance plus

left hand performance multiplied by 100. The probability level for significance for all

statistical analyses was set a p < 0.05

Research Question 1

Do individuals with AD have conceptual and/or ideomotor apraxia in the left hand?

Analysis. Separate nonparametric Mann-Whitney U tests were used to compare

left hand performance of the AD and HC groups on the verbal command pantomime (task

1) and the conceptual pantomime (task 3) tasks. The test variable was response accuracy

and the grouping variable was group (AD and HC).

Research Question 2

What is the contribution of degraded praxis movement and conceptual

representations (due to cortical atrophy) to the limb apraxia in individuals with AD?

Analysis. Separate nonparametric Mann-Whitney U tests were used to compare

right hand performance of the AD and HC groups on the verbal command pantomime

(task 1) and conceptual pantomime (task 3) tasks. The test variable was response

accuracy and the grouping variable was group (AD and HC).

Research Question 3

What is the contribution of interhemispheric disconnection (due to callosal atrophy)

to the limb apraxia in individuals with AD?

Analysis. Separate nonparametric Mann-Whitney U tests were used to compare

the asymmetry ratios of the AD and HC groups on the verbal command pantomime (task

1), pantomime imitation (task 2), and conceptual pantomime (task 3) tasks. The test









variable for each analysis was response accuracy and the grouping variable was group

(AD and HC)

Research Question 4

Is the disruption of praxis information transfer a result of an intrahemispheric

verbal-motor disconnection or an interhemispheric corpus callosum disconnection?

Analysis. A 2x2 ANOVA procedure was used to compare right hand performance

of the AD and HC groups on the verbal command pantomime (task 1) and pantomime

imitation (task 2) tasks. For the 2x2 ANOVA, factor one was task with two levels (verbal

command pantomime and pantomime imitation) and factor two was group with two

levels (AD group and HC group). A Mann-Whitney U test was used to compare

asymmetry ratios of the AD group for the verbal command pantomime (task 1) and

pantomime imitation (task 2) tasks. For this test, the test variable was response accuracy

and the grouping variable was group (AD and HC).










Table 2-1: Individual subject demographics for AD group.


subject
#


gender


education
(# of years)


01-001 F 85 10
01-003 F 78 14
01-004 F 61 12
01-005 F 80 18
01-006 F 82 15
01-007 F 73 12
01-008 M 80 10
01-009 M 77 12
01-010 M 80 15
01-011 M 90 20
01-012 M 80 13
01-013 F 76 12
01-014 M 85 12
01-015 M 88 14
01-016 F 81 12
01-017 M 83 18
01-018 F 71 16
01-019 F 80 12
01-020 F 85 12
01-021 F 81 13
01-022 F 76 15
01-024 F 71 12
mean 79.23 13.59
SD 6.4 2.6


male, SD = standard deviation


F = female, M










Table 2-2: Individual subject demographics for HC group.


subject
#


gender


education
(# of years)


02-001 F 84 15
02-002 F 78 16
02-003 F 65 18
02-004 F 73 16
02-005 F 67 12
02-006 F 63 19
02-007 F 68 12
02-009 F 76 16
02-010 M 76 18
02-011 M 70 18
02-012 M 81 12
02-013 M 80 18
02-014 F 83 12
02-017 M 85 18
02-018 M 83 16
02-019 F 82 14
02-020 F 79 14
02-021 M 76 12
02-022 F 74 16
02-023 F 70 16
02-025 M 85 18
02-026 M 77 12
02-028 F 79 12
02-029 F 85 12
mean 76.10 15.52
SD 6.8 2.4


male, SD = standard deviation


F = female, M:









Table 2-3: Strength of observer agreement for ranges of kappa statistic values.
Value of k Strength of
agreement
< 0 Poor
0 .20 Slight
.21 .40 Fair
.41 .60 Moderate
.61 .80 Substantial
.81 1.00 almost perfect
Kramer, M.S. & Feinstein, A.R. (1981). Clinical biostatistics: the biostatistics of
concordance. Clinical pharmacology and therapeutics, 29, 111-117.









AB





CD






E F





Figure 2-1: Examples of pictures used in the Florida Action Recall Test (FLART)
Target gesture (tool): A. carving (knife), B. chopping (hatchet), C. sharpening (pencil
sharpener), D. spreading (knife), E. opening (bottle opener), F. painting (paint brush).
Schwartz, R.L., Adair, J.C., Raymer, A.M., Williamson, D.J.G., Crosson, B., Rothi,
L.J.G., Nadeau, S.E., & Heilman, K.M. (2000). Conceptual apraxia in probable
Alzheimer's disease as demonstrated by the Florida Action Recall Test. Journal
of the International Neuropsychological Society, 6, 265-270.














CHAPTER 3
RESULTS

This study examined whether praxis information crosses the corpus callosum to

inform right hemisphere motor pathways by comparing right hand and left hand

performance on three praxis tasks in individuals with AD and healthy elderly control

subjects. Descriptive statistics and statistical analyses are presented in an attempt to

answer the proposed research questions.

Subject Demographics

As stated in the methods section, an attempt was made to match the HC and AD

groups for age and education level. Mann-Whitney U tests were used to compare the two

groups for age and education level. There was not a significant difference between the

AD and HC groups for age (U = 204.000, p = 0.186) but there was a significant

difference between the two groups for education level (U = 179.500, p = 0.055). Reasons

for this difference in education level will be addressed later (see discussion, Chapter 4).

Neuropsychological Screening

As described in the previous chapter, each participant was evaluated using several

cognitive screening measures prior to participating in the experimental protocol described

above. The Mini-Mental State Exam (MMSE), the Associative Match subtest of the

Birmingham Object Recognition Battery (BORB), the Sequential Commands subtest of

the Western Aphasia Battery (WAB), and a 15-item short form of the Boston Naming

Test (BNT) were administered to each participant. The purpose of this

neuropsychological screening was threefold. 1) The performance of the AD group on









these measures was used to support the medical diagnosis and to verify the presence of

memory and cognitive deficits. Therefore, in order to be included in this study, all of the

subjects in the AD group were required to score below a certain level in order to be

considered impaired on a particular measure (see Chapter 2 for cut-off scores). 2) The

performance of the HC group on these measures was used to determine the current level

of cognitive functioning for each participant and to verify that each participant was

performing at age appropriate levels for measures of cognition and memory. Therefore,

in order to be included in this study, all of the subjects in the HC group were required to

score above a certain level in order to be considered within normal limits for a particular

cognitive domain (see Chapter 2 for cut-off scores). 3) For the AD group, it was hoped

that performance on these neuropsychological measures could be used to subdivide the

group for further data analysis. However, due to the small sample size, this type of post-

hoc analysis was not feasible. Following are the results of the neuropsychological

screening.

For the MMSE, a score of 27 or higher (out of 30) was considered within normal

limits while a score of 25 or lower (out of 30) was considered consistent with dementia

(Lezak, 1995). The subjects in the HC group scored a range of 27-30 (mean=28.25,

SD=1.1) while the AD group scored a range of 12-25 (mean=20.82, SD=3.9). In the AD

group, 0 participants had severe (MMSE < 10), 7 participants had moderate (MMSE >10

but <20) and 15 participants had mild (MMSE >20 but <25) dementia.

On the Sequential Commands subtest of the WAB, inclusion into the study required

a score of at least 40/80 or 50% accuracy and for the AD group and a score of at least

72/80 or 90% accuracy for the HC group. On this measure, the score range for the HC









group was 75-80 (mean=79.58, SD=1.3) and the score range for the AD group was 39-80

(mean=73.91, SD=9.5). One of the participants in the AD group scored below the 40/80

cut-off for inclusion (subject #01-004 scored 39/80 on this measure). However, due to

difficulty recruiting subjects for the study, the data was nevertheless included.

On the Associative Match subtest of the BORB, inclusion into the study required a

score of at least 21/30 or 70% accuracy for the AD group and a score of at least 27/30 or

90% accuracy for the HC group. On this measure, the score range for the HC group was

28-30 (mean=29.75, SD=0.5) and the score range for the AD group was 19-30

(mean=27.41, SD=2.6). One of the participants in the AD group scored below the 21/30

cut-off for inclusion (subject #01-004 scored 19/30 on this measure). However, due to

difficulty recruiting subjects for the study, the data was nevertheless included.

A 15-item version of the BNT (Fastenau et al., 1998) was also administered to each

participant. A score of 12 or below was considered impaired for this measure. Subjects

in the HC group scored a range of 13-15 items correct (mean=13.96, SD=0.8). Subjects

in the AD group scored a range of 1-14 items correct (mean=8.77, SD=3.1). Although

subject #01-009 scored within normal limits on the 15-item BNT, he was included in the

study because he had been previously diagnosed with AD by a neurologist and his

MMSE score (23/30) was below that considered to be consistent with dementia. See

Table 3-1 and 3-2 for scores on these screening measures for each individual subject.

Reliability

Because of the subjective nature of the scoring method used for this study (Rothi et

al., 1988), it was important establish the reliability of the accuracy judgments and error

categorization made by the primary rater (Rater 1). This was accomplished by requiring

Rater 1 to score 20% of the data on two separate occasions (intra-rater reliability) and









requiring Rater 2 to score 20% of the data independent of the primary rater (inter-rater

reliability). The following results suggest that high inter- and intra- rater reliability was

established thereby lending credibility to the data.

For inter- and intra- rater reliability, % agreement was greater than 80% for all

response variables and categories with the exception of inter-rater reliability of IC in task

1 (VC) (75.9%), IC in task 2 (PI) (77.1%), and IC and M in task 3 (CP) (78.2% and

79.6%, respectively). Overall, intra-rater reliability was slightly better than inter-rater

reliability in that there were no instances in which the percentage agreement of Rater 1 as

compared to Rater 2 was less than 80%.

The kappa (k) statistic was significant at the 0.05 level for all response variables,

where applicable. Kappa was greater than 0.40 for all reliability comparisons with the

exception of inter-rater reliability of C (k=0, poor) in Task 3 (CP), intra-rater reliability of

P (k=0, poor), R (k=0.282, fair), H (k=0.284, fair), and UR (k=0.402, fair) in task 1 (VC),

intra-rater reliability of R (k=0, poor) in task 2 (PI), and intra-rater reliability of P

(k=0.328, fair), N (k=0.332, fair), and C (k=0, poor) in task 3 (CP). In the instances in

which the % agreement was relatively high (i.e. > 90%) but the value of kappa was

relatively low (i.e. < .40) (see bolded variables in Tables 3-3 and 3-4), k is perhaps not a

valid measure of concordance. The reason for this disparity between percent agreement

and the kappa statistic, is that for the variables in question, the two observers did not

disagree enough to account for the possibility that they were agreeing by chance. See

Tables 3-3 and 3-4 for inter- and intra- reliability data for each independent variable in

tasks 1, 2, and 3.









Descriptive Statistics

Task 1: Verbal Command Pantomime (VC)

HC group. For the HC group, mean percent accuracy on this task was 50.0% (SD

= 7.2) with the right hand and 47.0% (SD = 7.7) with the left hand with a mean difference

between the two hands of 3.0% (SD = 7.2%) and a mean asymmetry ratio of 3.26 (SD =

8.12). In the HC group (see table 3-5), 13/24 (54.2%) participants performed better with

the right hand than the left hand (a positive difference) while 8/24 (33.3%) participants

performed better with the left hand than the right hand (a negative difference) and 3/24

(12.5%) participants showed no difference between the hands (see table 3-5).

AD group. Mean percent accuracy for the AD group on task 1 (VC) was 28.7%

(SD = 11.7) with the right hand and 24.7% (SD = 10.7) with the left hand with a mean

difference between the two hands of 4.0% (SD = 6.5%) and a mean asymmetry ratio of

7.97 (SD = 13.55). For task 1 (VC) in the AD group (see table 3-6), 15/22 (68.1%)

participants performed better with the right hand than the left hand (a positive difference)

while 6/22 (27.3%) participants performed better with the left hand than the right hand (a

negative difference) and 1/22 (4.5%) participant showed no difference between the hands

(see table 3-6).

On task 1 (VC), the difference in performance of the HC and AD groups was

21.3% with the right hand and 23.0% with the left hand. In summary, the HC group

performed pantomimes more accurately and demonstrated less performance variability

than the AD group on this task.

Task 2: Pantomime Imitation (PI)

HC group. Mean percent accuracy for the HC group, on this task was 49.8% (SD

= 11.7) with the right hand and 42.4% (SD = 13.1) with the left hand with a mean









difference between the two hands of 7.3% (SD = 7.2%) and a mean asymmetry ratio of

8.85 (SD = 8.68). In the HC group (see table 3-5), 19/24 (79.2%) participants performed

better with the right hand than the left hand (a positive difference) while 4/24 (16.7%)

participants performed better with the left hand than the right hand (a negative difference)

and 1/24 (4.2%) participant showed no difference between the hands (see table 3-5).

AD group. Data from the AD group for Task 2 showed a mean percent accuracy

of 31.1% (SD = 14.1) with the right hand and 20.4% (SD = 9.4) with the left hand with a

mean difference between the two hands of 10.7% (SD = 8.7%) and a mean asymmetry

ratio of 20.57 (SD = 21.96). For task 2 (PI) in the AD group (see table 3-6), 20/21

(95.2%) participants performed better with the right hand than the left hand (a positive

difference) while 1/21 (4.8%) participants performed better with the left hand than the

right hand (a negative difference) and 0/22 (0%) participant showed no difference

between the hands (see table 3-6).

On task 2, the difference in performance of the HC and AD groups was 18.7% with

the right hand and 22.0% with the left hand. Overall, the HC group performed

pantomimes more accurately but with similar variability in comparison to the AD group

on this task

Task 3: Conceptual Pantomime (CP)

HC group. For the HC group, mean percent accuracy on this task was 86.2% (SD

= 7.2) with the right hand and 84.9% (SD = 7.5) with a mean difference between the two

hands of 1.3% (SD = 4.6%) and a mean asymmetry ratio of 0.79 (SD = 2.78). In the HC

group (see table 3-5), 11/24 (45.8%) participants performed better with the right hand

than the left hand (a positive difference) while 7/24 (29.1%) participants performed better









with the left hand than the right hand (a negative difference) and 5/24 (20.8%)

participants showed no difference between the hands (see table 3-5).

AD group. Mean percent accuracy for the AD group on task 3 (CP) was 62.8%

(SD = 13.9) with the right hand and 61.1% (SD = 13.4) for the left hand with a mean

difference between the two hands of 1.7% (SD = 4.5%) and a mean asymmetry ratio of

1.38 (SD = (3.87). For task 3 (CP) in the AD group (see table 3-6), 11/18 (61.1%)

participants performed better with the right hand than the left hand (a positive difference)

while 6/18 (33.3%) participants performed better with the left hand than the right hand (a

negative difference) and 1/18 (5.5%) participant showed no difference between the hands

(see table 3-6).

On Task 3, the difference in performance of the HC and AD groups was 23.2%

with the right hand and 24.0% with the left hand. In summary, the HC group performed

pantomimes more accurately and demonstrated less performance variability than the AD

group on this task.

Error Types Task 1, 2, and 3

Descriptive data for error types can be found in table 3-7 (task 1), 3-8 (task 2), 3-9

(task 3) and 3-10 (error totals for tasks 1, 2, and 3). For Tasks 1 and 2, both groups

showed a high percentage of sequencing (S), internal configuration (IC), external

configuration (EC), and movement (M) errors with both hands relative to other error

types. The AD group also demonstrated a higher percentage of body part as tool (BPT)

errors than the HC group with both hands on task 1 and a higher percentage of

unrecognizable errors (UR) with both hands on task 1 and 2.

For task 3 (CP), only content errors are reported since this is a conceptual task.

The HC group showed a higher percentage of related (R) than hand (H) errors while the









AD group showed a higher percentage of hand (H) than related (R) errors on task 3 (CP).

The percentage of perseverative (P) and nonrelated (N) errors on this task was relatively

low for both groups with both hands.

When the total number of errors for task 1 (VC) and task 2 (PI) was calculated, the

HC group produced fewer errors than the AD group and both groups produced fewer

errors with the right hand than the left hand. For task 3 (CP), the AD group made more

errors than the HC group but there was little difference between the right hand and the

left hand in the total number of errors for both groups.

Statistical Analysis

Research Question 1

Do individuals with AD have conceptual and/or ideomotor apraxia in the left hand?

Results. To answer this question, Mann-Whitney U tests were performed using

data from left hand performance of the AD and the HC groups on the verbal command

pantomime (task 1) and conceptual pantomime (task 3) tasks. The Mann-Whitney U test

for left hand performance on the verbal command pantomime task (ideomotor apraxia)

was significant at the p < 0.01 level (U = 25.500, p < 0.001). The Mann-Whitney U test

for left hand performance on the conceptual pantomime task (conceptual apraxia) was

significant at the p < 0.01 level (U = 17.500, p < 0.001)].

Research Question 2

What is the contribution of degraded praxis movement and conceptual

representations (due to cortical atrophy) to the limb apraxia in individuals with AD?

Results. To answer this question, Mann-Whitney U tests were performed using

data from right hand performance of the AD and the HC groups on the verbal command

pantomime (task 1) and conceptual pantomime (task 2) tasks. The Mann-Whitney U test









for right hand performance on the verbal command pantomime task was significant at the

p < 0.01 level (U = 50.000, p < 0.001). The Mann-Whitney U test for right hand

performance on the conceptual pantomime task was significant at the p < 0.01 level (U=

25.000, p < 0.001).

Research Question 3

What is the contribution of interhemispheric disconnection (due to callosal atrophy)

to the limb apraxia in individuals with AD?

Results. To answer this question, Mann-Whitney U tests were performed using

asymmetry ratio data from the performance of the AD and HC groups on the praxis

production and praxis conceptual tasks. The Mann-Whitney U test for asymmetry ratios

on the verbal command pantomime task (task 1) was significant at the p < 0.05 level (U=

171.000, p = 0.041). The Mann-Whitney U test for asymmetry ratios on the pantomime

imitation task (task 2) were significant at the p < 0.01 level (U = 104.500, p = 0.001).

The Mann-Whitney U test for asymmetry ratios on the conceptual pantomime task (task

3) were not significant (U = 181.000, p = 0.373).

Research Question 4

Is the disruption of praxis information transfer a result of an intrahemispheric

verbal-motor disconnection or an interhemispheric corpus callosum disconnection?

Results. To answer this question, a 2x2 ANOVA procedure was used to compare

right hand performance of the AD and HC groups on the verbal command pantomime

(task 1) and pantomime imitation (task 2) tasks and a Mann-Whitney U test was used to

compare asymmetry ratios of the AD group for the verbal command pantomime (task 1)

and pantomime imitation (task 2) tasks. The 2x2 ANOVA was significant for the main

effect of group [F(1) = 68.498, p < 0.001, effect size = .441] but was not significant for









the main effect of task [F(1) = 0.250, p = 0.619] and there was not a significant

task*group interaction [F(1) = 0.237, p = 0.627]. The Mann-Whitney U test was also

significant (U = 100.000, p = 0.001)

Summary

With regards to the neuropsychological screening, the AD group demonstrated

memory and cognitive deficits consistent with a diagnosis of dementia while the HC

group demonstrated normal performance on memory and cognitive tests. Both intra and

inter rater reliability were determined to be relatively high lending credibility to the

scoring system utilized to analyze the data.

With regards to descriptive statistics, overall, the HC group demonstrated greater

response accuracy, less performance variability, and fewer errors than the AD group. For

both groups with both hands, spatial and temporal errors were the most common types of

errors produced during task 1 (VC) and task 2 (PI) while content errors were the most

common type of error in task 3 (CP) for both groups with both hands. The AD group

also produced more unrecognizable responses than the HC group.

The statistical analyses that were conducted in order to answer the research

questions showed that individuals with AD demonstrated ideomotor and conceptual

apraxia in both the right and left hands. Additionally, the results suggested that callosal

degeneration in individuals with AD interrupts the interhemispheric transfer of praxis

production information but not praxis conceptual information. Finally, it can be

concluded that the interruption of interhemispheric transfer of praxis information in

individuals with AD is specific to the transfer of motor information from left hemisphere

praxis movement representations to right hemisphere motor areas. A discussion of the

clinical and empirical implications of these results follows.









Table 3-1: Scores for screening measures for individual subjects in HC group.


subject #MMSE WAB BORE BNT
02-001 27 80 30 13
02-002 30 79 30 15
02-003 30 80 30 15
02-004 29 80 30 14
02-005 28 80 30 14
02-006 30 80 30 14
02-007 27 80 30 15
02-009 28 80 29 15
02-010 28 75 29 15
02-011 27 80 29 14
02-012 28 80 30 15
02-013 27 80 30 13
02-014 30 80 30 14
02-017 28 80 30 13
02-018 30 80 30 13
02-019 27 80 30 14
02-020 28 80 28 13
02-021 27 80 30 14
02-022 29 76 30 14
02-023 29 80 30 15
02-025 27 80 30 13
02-026 28 80 30 14
02-028 28 80 30 13
02-029 28 80 29 13
Mean 28.25 79.58 29.75 13.96
SD 1.1 1.3 0.5 0.8
MMSE = Mini Mental State Exam, WAB = Western Aphasia Battery, Sequential
Commands Subtest, BORB = Birmingham Object Recognition Battery, Semantic
Matching Subtest, BNT = 15-item short form of Boston Naming Test
SD = standard deviation









Table 3-2: Scores for screening measures for individual subjects in AD group.


subject #IMMSE


WAB


BORB


BNT


01-001 20 76 29 13
01-003 19 80 29 11
01-004 16 39 19 6
01-005 23 80 29 9
01-006 23 80 27 6
01-007 23 80 25 11
01-008 24 80 29 9
01-009 23 78 27 14
01-010 12 71 30 10
01-011 25 80 29 10
01-012 24 68 27 11
01-013 18 80 30 9
01-014 16 62 24 1
01-015 14 67 27 4
01-016 22 72 24 8
01-017 16 70 26 6
01-018 23 80 29 5
01-019 23 80 29 9
01-020 21 80 28 11
01-021 24 70 29 9
01-022 24 75 29 9
01-024 25 78 28 12
mean 20.82 73.91 27.41 8.77
SD 3.9 9.5 2.6 3.1
MMSE = Mini Mental State Exam, WAB = Western Aphasia Battery, Sequential
Commands Subtest, BORB = Birmingham Object Recognition Battery, Semantic
Matching Subtest, BNT = 15-item short form of Boston Naming Test
SD = standard deviation









Table 3-3: Inter-rater reliability using % agreement and the Kappa statistic for task 1, 2,
and 3


response
variable % k k strength % K k strength % k k strength
accuracy 92.1 0.829 almost perfect 92.7 0.833 almost perfect 96.8 0.912 almost perfect
P 100.0 1.000 almost perfect 100.0 1.000 almost perfect 99.9 0.767 substantial
R 99.9 0.856 almost perfect 100.0 1.000 almost perfect 99.6 0.943 almost perfect
N 100.0 1.000 almost perfect 100.0 1.000 almost perfect 99.9 0.908 almost perfect
H 99.7 0.856 almost perfect 100.0 1.000 almostperfect 98.2 0.906 almostperfect
content 99.9 N/A 100.0 N/A 99.3 N/A
S 99.7 0.821 almost perfect 95.5 0.815 almostperfect 97.1 0.781 substantial
T 96.8 0.690 substantial 97.0 0.603 moderate 96.3 0.575 moderate
O 96.6 0.795 substantial 98.4 0.585 moderate 98.9 0.812 almostperfect
temporal 97.1 N/A 97.0 N/A 97.4 N/A
A 98.2 0.688 substantial 97.6 0.626 substantial 98.6 0.699 substantial
IC 84.2 0.708 substantial 86.6 0.756 substantial 86.8 0.799 substantial
EC 87.9 0.565 moderate 91.1 0.736 substantial 94.0 0.656 substantial
BPT 97.4 0.538 moderate 99.0 0.627 substantial 98.9 0.846 almost perfect
M 88.2 0.732 substantial 91.2 0.813 almostperfect 89.1 0.696 substantial
spatial 91.5 N/A 93.3 N/A 93.7 N/A
C 100.0 1.000 almost perfect 100.0 1.000 almost perfect 99.9 0.000 poor
NR 99.9 0.799 substantial 99.9 0.888 almost perfect 99.7 0.908 almost perfect
UR 99.6 0.867 almostperfect 99.6 0.912 almostperfect 99.3 0.926 almostperfect
other 99.8 N/A 99.8 /A 99.6 N/A


% agreement


(# of agreements # of disagreements)/total # of stimuli x 100


VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime
% = percent of agreement, k = kappa statistic
P = Perseverative error, R = Related error, N = Non-related error, H = Hand error,
S = Spatial error, T = Timing error, O = Occurrence error, A = Amplitude error,
IC = Internal Configuration error, EC = External Configuration error,
BPT = Body-part-as-tool error, M = Movement error, C = Concretisation error,
NR = No Response, UR = Unrecognizable response
accuracy = percentage of correct responses, content = sum of P, R, N, and H errors,
temporal = sum of S, T, and O errors, spatial = sum of A, IC, EC, BPT, and M errors,
other = sum of C, NR, and UR errors


Task 1-VC


Task 2-PI


Task 3-CP









Table 3-4: Intra-rater reliability using % agreement and the Kappa statistic for task 1, 2,
and 3.


response
variable % k k strength % k k strength % k k strength
accuracy 82.7 0.661 substantial 85.4 0.685 substantial 92.7 0.810 almost perfect
P 99.9 0.000 poor 100.0 1.000 almost perfect 98.9 0.328 fair
R 99.3 0.282 fair 99.9 0.000 poor 97.7 0.713 substantial
N 100.0 1.000 almost perfect 100.0 1.000 almost perfect 99.4 0.332 fair
H 99.3 0.284 fair 100.0 1.000 almost perfect 95.0 0.769 substantial
Content 99.6 N/A 100.0 N/A 97.8 N/A
S 94.0 0.704 substantial 93.3 0.733 substantial 95.3 0.662 substantial
T 95.8 0.653 substantial 96.1 0.523 moderate 94.6 0.539 moderate
O 97.6 0.755 substantial 98.1 0.543 moderate 97.7 0.608 substantial
temporal 95.8 N/A 95.9 N/A 95.9 N/A
A 96.8 0.461 moderate 97.3 0.445 moderate 97.9 0.535 moderate
IC 75.9 0.580 moderate 77.1 0.621 substantial 78.2 0.553 moderate
EC 85.0 0.484 moderate 82.5 0.564 moderate 88.4 0.452 moderate
BPT 98.0 0.622 substantial 99.3 0.702 substantial 97.7 0.626 substantial
M 80.0 0.593 moderate 84.2 0.690 substantial 79.6 0.498 moderate
Spatial 87.9 N/A 88.8 N/A 89.0 N/A
C 100.0 1.000 almost perfect 100.0 1.000 almost perfect 99.9 0.000 poor
NR 99.7 0.749 substantial 99.7 0.832 almost perfect 99.2 0.696 substantial
UR 97.6 0.402 fair 97.7 0.489 moderate 96.0 0.630 substantial
Other 99.1 N/A 99.2 N/A 98.4 N/A
% agreement = (# of agreements # of disagreements)/total # of stimuli x 100
VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime
% = percent of agreement, k = kappa statistic
P = Perseverative error, R = Related error, N = Non-related error, H = Hand error,
S = Spatial error, T = Timing error, O = Occurrence error, A = Amplitude error,
IC = Internal Configuration error, EC = External Configuration error,
BPT = Body-part-as-tool error, M = Movement error, C = Concretisation error,
NR = No Response, UR = Unrecognizable response
accuracy = percentage of correct responses, content = sum of P, R, N, and H errors,
temporal = sum of S, T, and O errors, spatial = sum of A, IC, EC, BPT, and M errors,
other = sum of C, NR, and UR errors


Task 1- VCP


Task 2- PI


Task 3- CP









Table 3-5: Response accuracy (percent) data with difference scores and asymmetry ratios
for individual subjects in HC group for Tasks 1, 2, and 3.


Task 1-VCP


Task 2-PT


Task 3-CP


% % % % % %
acc acc acc acc acc acc
subj# RH LH diff ratio RH LH diff ratio RH LH diff ratio
02-00151.1 46.7 4.4 4.50 44.4 28.9 15.5 21.15 93.3 86.7 6.6 3.67
02-00244.4 46.7 -2.3 -2.52 40.0 40.0 0.0 0.00 91.1 91.1 0.0 0.00
02-00351.1 44.4 6.7 7.02 51.1 40.0 11.1 12.18 88.9 84.4 4.5 2.60
02-00442.2 44.4 -2.2 -2.54 33.3 26.7 6.6 11.00 80.0 82.2 -2.2 -1.36
02-00562.2 48.9 13.3 11.97 53.3 44.4 8.9 9.11 88.9 88.9 0.0 0.00
02-00646.7 46.7 0.0 0.00 46.7 53.3 -6.6 -6.60 93.3 91.1 2.2 1.19
02-00744.4 35.6 8.8 11.00 33.3 26.7 6.6 11.00 91.1 93.3 -2.2 -1.19
02-00955.6 55.6 0.0 0.00 44.4 37.8 6.6 8.03 93.3 93.3 0.0 0.00
02-01053.3 44.4 8.9 9.11 60.0 64.4 -4.4 -3.54 86.7 84.4 2.3 1.34
02-01144.4 48.9 -4.5 -4.82 42.2 31.1 11.1 15.14 80.0 73.3 6.7 4.37
02-01248.9 51.1 -2.2 -2.20 57.8 35.6 22.2 23.77 86.7 88.9 -2.2 -1.25
02-01348.9 48.9 0.0 0.00 68.9 55.6 13.3 10.68 75.6 66.7 8.9 6.25
02-01446.7 48.9 -2.2 -2.30 51.1 37.8 13.3 14.96 84.4 88.9 -4.5 -2.60
02-01746.7 44.4 2.3 2.52 46.7 37.8 8.9 10.53 88.9 91.1 -2.2 -1.22
02-01846.7 51.1 -4.4 -4.50 64.4 55.6 8.8 7.33 77.8 82.2 -4.4 -2.75
02-01944.4 26.7 17.7 24.89 31.1 33.3 -2.2 -3.42 84.4 91.1 -6.7 -3.82
02-02048.9 35.6 13.3 15.74 35.6 20.0 15.6 28.06 71.1 68.9 2.2 1.57
02-02151.1 42.2 8.9 9.54 51.1 46.7 4.4 4.50 88.9 84.4 4.5 2.60
02-02255.6 51.1 4.5 4.22 46.7 44.4 2.3 2.52 75.6 80.0 -4.4 -2.83
02-02366.7 55.6 11.1 9.08 67.4 57.8 9.6 7.67 97.9 93.3 4.6 2.41
02-02566.7 64.4 2.3 1.75 68.9 73.3 -4.4 -3.09 88.9 88.9 0.0 0.00
02-02644.4 57.8 -13.4-13.11 46.7 42.2 4.5 5.06 95.6 84.4 11.2 6.22
02-02851.1 46.7 4.4 4.50 66.7 53.3 13.4 11.17 91.1 84.4 6.7 3.82
02-02937.8 42.2 -4.4 -5.50 42.2 31.1 11.1 15.14 75.6 75.6 0.0 0.00
mean 50.0 47.0 3.0 3.26 49.8 42.4 7.3 8.85 86.2 84.9 1.3 0.79
stdev 7.2 7.7 7.2 8.12 11.7 13.1 7.2 8.68 7.2 7.5 4.6 2.78
VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime
% acc RH = percent response accuracy with the right hand
calculated as number of correct responses / total number of stimuli x 100
% acc LH = percent response accuracy with the left hand
calculated as number of correct responses / total number of stimuli x 100
diff = difference between % acc RH and % acc LH (i.e. RH minus LH)
ratio (asymmetry ratio) = (% acc RH % acc LH) / (% acc RH + % acc LH)
DNT = did not test, SD = standard deviation









Table 3-6: Response accuracy (percent) data with difference scores and asymmetry ratios
for individual subjects in AD group for Tasks 1, 2, and 3.


Task 1-VCP


Task 2-PT


Task 3-CP


%% % % %
acc acc acc acc acc
subj# RH LH diff ratio RH LH diff ratio RH LH diff ratio
01-00113.3 8.9 4.4 19.82 15.6 6.7 8.9 39.91 51.1 42.2 8.9 9.54
01-00311.1 15.6 -4.5 -16.85 4.4 15.6 -11.2-56.00 60.0 64.4 -4.4 -3.54
01-00424.4 11.1 13.3 37.46 DNT DNT N/A N/A 26.7 26.7 0.0 0.00
01-00522.2 15.6 6.6 17.46 37.8 20.0 17.8 30.80 68.9 66.7 2.2 1.62
01-00620.0 20.0 0.0 0.00 15.6 8.9 6.7 27.35 51.1 46.7 4.4 4.50
01-00726.7 28.9 -2.2 -3.96 8.9 2.2 6.7 60.36 57.8 53.3 4.5 4.05
01-00824.4 17.8 6.6 15.64 37.8 24.4 13.4 21.54 71.1 66.7 4.4 3.19
01-00922.2 17.8 4.4 11.00 48.9 28.9 20.0 25.71 82.2 77.8 4.4 2.75
01-01020.0 24.4 -4.4 -9.91 17.8 15.6 2.2 6.59 62.2 66.7 -4.5 -3.49
01-01144.4 26.7 17.7 24.89 51.1 22.2 28.9 39.43 64.4 60.0 4.4 3.54
01-01228.9 24.4 4.5 8.44 46.7 24.4 22.3 31.36 62.2 57.8 4.4 3.67
01-01335.6 37.8 -2.2 -3.00 37.8 33.3 4.5 6.33 62.2 60.0 2.2 1.80
01-01431.1 24.4 6.7 12.07 37.8 24.4 13.4 21.54 DNT DNT N/A N/A
01-01531.1 22.2 8.9 16.70 26.7 22.2 4.5 9.20 DNT DNT N/A N/A
01-01617.8 22.2 -4.4 -11.00 17.8 11.1 6.7 23.18 DNT DNT N/A N/A
01-01733.3 40.0 -6.7 -9.14 44.4 28.9 15.5 21.15 DNT DNT N/A N/A
01-01853.3 40.0 13.3 14.26 33.3 17.8 15.5 30.33 66.7 68.9 -2.2 -1.62
01-01931.1 26.7 4.4 7.61 28.9 22.2 6.7 13.11 48.9 55.6 -6.7 -6.41
01-02015.6 13.3 2.3 7.96 22.2 11.1 11.1 33.33 57.8 60.0 -2.2 -1.87
01-02151.1 46.7 4.4 4.50 46.7 42.2 4.5 5.06 88.9 82.2 6.7 3.92
01-02246.7 42.2 4.5 5.06 26.7 20.0 6.7 14.35 75.6 77.8 -2.2 -1.43
01-02426.7 15.6 11.1 26.24 46.7 26.7 20.0 27.25 73.3 66.7 6.6 4.71
Mean 28.7 24.7 4.0 7.97 31.1 20.4 10.7 20.57 62.8 61.1 1.7 1.38
SD 11.7 10.7 6.5 13.55 14.1 9.4 8.7 21.96 13.9 13.4 4.5 3.87
VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime
% acc RH = percent response accuracy with the right hand
calculated as number of correct responses / total number of stimuli x 100
% acc LH = percent response accuracy with the left hand
calculated as number of correct responses / total number of stimuli x 100
diff = difference between % acc RH and % acc LH (i.e. RH minus LH)
ratio (asymmetry ratio) = (% acc RH % acc LH) / (% acc RH + % acc LH)
DNT = did not test, SD = standard deviation









Table 3-7: Error analysis descriptive data for Task 1 (VC


group hand %P %R %N %H %S %T %O %A %IC %EC %BPT%M %C %NR %UR
LH 0.00 0.23 0.00 0.00 9.74 5.80 3.48 1.62 27.49 14.39 0.93 32.48 0.12 0.93 2.78
HC
RH 0.12 0.00 0.12 0.00 10.64 5.57 4.21 2.35 30.69 13.37 1.61 29.46 0.00 0.25 1.61
SLH 0.13 0.91 0.21 0.34 8.92 4.68 3.60 2.72 27.32 17.01 4.27 26.02 0.86 1.19 5.89
AD34 8.48 4.58 3.39 .97 26.04 16.88 3.73 25.61 0.85 0.93 4.92
RH 0.17 1.02 0.08 0.34 8.48 4.58 3.39 2.97 26.04 16.88 3.73 25.61 0.85 0.93 4.92


VC = verbal command, HC


healthy control, AD


Alzheimer's disease, RH


right hand, LH


left hand


% = percentage of..., P=Perseverative errors, R=Related, N=Non-related, H=Hand, S=Spatial, T=Timing, O=Occurrence,
A=Amplitude, IC=Internal Configuration, EC=External Configuration, BPT=Body-part-as-tool, M=Movement, C=Concretisation,
NR=No response, UR=Unrecognizable response
Percentages for error types were calculated as number of error present / total number of errors x 100

Table 3-8: Error analysis descriptive data for Task 2 (PI)
%UR

group hand %P %R %N %H %S %T %O %A %IC %EC %BPT%M %C %NR
LH 0.21 0.21 0.21 0.00 10.10 2.23 0.53 0.74 33.48 18.49 0.11 31.03 0.00 0.21 2.44
HC
RH 0.13 0.00 0.00 0.00 11.41 2.41 0.63 0.76 34.98 14.83 0.38 31.05 0.00 0.76 2.66
LH 0.00 0.07 0.00 0.00 11.01 3.20 1.04 1.26 33.11 19.20 0.67 25.52 0.00 0.67 4.24
AD00 13.49 2.66 0.83 1.19 33.12 17.25 0.64 26.51 0.00 0.09 4.04
RH 0.09 0.00 0.09 0.00 13.49 2.66 0.83 1.19 33.12 17.25 0.64 26.51 0.00 0.09 4.04


PI = pantomime imitation,


HC = healthy control, AD = Alzheimer's disease, RH = right hand, LH


left hand


% = percentage of..., P=Perseverative errors, R=Related, N=Non-related, H=Hand, S=Spatial, T=Timing, O=Occurrence,
A=Amplitude, IC=Internal Configuration, EC=External Configuration, BPT=Body-part-as-tool, M=Movement, C=Concretisation,
NR=No response, UR=Unrecognizable response
Percentages for error types were calculated as number of error present / total number of errors x 100







66


Table 3-9: Error analysis descriptive data for Task 3 (CP).


Group


SLH 1.98 56.44 4.95 36.63
HC
RH 0.98 53.92 5.88 39.22
AD LH 1.96 31.37 3.27 63.40
AD
RH 1.27 29.30 1.91 67.52
CP = conceptual pantomime, HC = healthy control, AD = Alzheimer's disease
RH = right hand, LH = left hand
% = percentage of..., P=Perseverative errors, R=Related, N=Non-related, H=Hand,
Percentages for error types were calculated as number of error present / total number of
errors x 100


Table 3-10: Error tctal 3


HC = healthy control, AD = Alzheimer's disease, RH = right hand, LH = left hand
task 1 = verbal command pantomime, task 2 = pantomime imitation, task 3 = conceptual
pantomime
% = percentage of..., P=Perseverative errors, R=Related, N=Non-related, H=Hand
Error total is the sum of all errors produced by each group for each task. For task 3, only
content errors are included in the sum of errors.


hand %P


error
group task hand total

HC LH 862
1 RH 808
AD LH 3605
AD
_H 1179

HC LH 941
SRH 789
AD LH 1344
AD
_RH 1090
C LH 101
3 RH 102

AD LH 153
RH 157


I I














CHAPTER 4
DISCUSSION

Apraxia is a movement disorder in which voluntary movement is impaired without

muscle weakness. This impairment affects the ability to select and sequence previously

learned skilled movements. Limb apraxia specifically refers to an acquired disorder of

skilled movement that affects hand and arm function. In order to perform skilled

movements, sensory input must interact with stored movement representations that are

translated into patterns of innervation. Empirical evidence has shown that the neural

representations for skilled movement are located in the parietal lobe of the left

hemisphere. In order to perform skilled movements with the right hand, praxis

movement representations and innervatory patterns in the left hemisphere must transfer

motor program information to left primary motor cortex via intrahemispheric white

matter projections. In order to perform skilled movements with the left hand, praxis

movement representations and innervatory patterns in the left hemisphere must transfer

motor program information to right primary motor cortex via interhemispheric white

matter fibers. This study proposed to investigate the neural mechanisms of limb apraxia

by examining the transfer of different types of praxis information from the left

hemisphere to the right hemisphere via the corpus callosum. AD was proposed as a

model for studying this process because individuals in this population can perform praxis

tasks with both hands (i.e. they do not have hemiplegia), this diagnosis is prevalent

among the elderly population (i.e. this is not a rare syndrome), and individuals in this









population demonstrate both limb apraxia and callosal atrophy (i.e. can potentially

differentiate what type of information is being transferred across the corpus callosum).

Previous studies have provided evidence of neuronal loss in the areas of the brain

that govern skilled movement systems (i.e. left parietal lobe) and this likely contributes to

the presence of apraxia in the right hand of right-handed individuals with AD. Other

studies have suggested neuronal loss in the cortical layers that project to contralateral

motor areas (i.e. corpus callosum) and this could explain the presence of apraxia in the

left hand of right-handed individuals with AD. Therefore, the goal of this study was to

examine whether praxis information is transferred across the corpus callosum and what

type of praxis information is transferred across the corpus callosum. Investigations of

callosal apraxia use asymmetries in right and left hand performance on praxis tasks

(pantomime to command, pantomime imitation and conceptual pantomime) to examine

the mechanisms of transfer of praxis information in terms of white matter disconnections.

It has been shown previously that praxis performance in individuals with AD is

significantly different than praxis performance of healthy elderly individuals. Therefore,

this study attempted to investigate if there was a greater disparity between right hand and

left hand performance on praxis tasks in individuals with AD as compared to healthy

elderly individuals. Based on descriptions of individuals with callosal disconnection (De

Renzi et al., 1982; Gazzaniga et al., 1967; Geschwind & Kaplan, 1962; Graff-Radford et

al., 1987; Watson & Heilman, 1983), investigation of the transfer of praxis production

and conceptual information was proposed.

Summary and Explanation of Findings

Research Question 1

Do individuals with AD have ideomotor and/or conceptual apraxia in the left hand?









Summary. In order to answer this question, left hand performance of each group

on the verbal command pantomime and conceptual pantomime tasks was compared. A

significant difference in performance (with the left hand) between the two groups on the

verbal command pantomime task would indicate the presence of ideomotor apraxia in the

left hand of individuals with AD. Statistical analysis of the data revealed that there was a

significant difference (with the left hand) between the two groups on the verbal command

pantomime task indicating that individuals with AD have ideomotor apraxia in the left

hand. A significant difference in performance (with the left hand) between the two

groups on the conceptual pantomime task would indicate the presence of conceptual

apraxia in the left hand of individuals with AD. Statistical analysis of the data revealed

that there was a significant difference (with the left hand) between the two groups on the

conceptual pantomime task indicating that individuals with AD have conceptual apraxia

in the left hand. Based on these findings, it can be concluded that individuals with AD

demonstrated both ideomotor and conceptual apraxia with the left hand.

Explanation. Left hand performance was examined to answer this research

question for two reasons. First, previous studies have reported ideomotor and conceptual

apraxia in the right hand of individuals with AD but there are no reports in the literature

that address left hand performance. Since the goal of this study was to examine bimanual

praxis mechanisms, the first step was to establish patterns of apraxia in the left hand that

were similar to previously reported patterns of apraxia in the right hand. Second, left

hand performance requires recruitment of both left hemisphere praxis representations and

right hemisphere motor areas and requires the transfer of praxis movement and

conceptual information across the corpus callosum. If there is ideomotor and conceptual









apraxia in the left hand, it is unknown whether this results from degradation of left

hemisphere praxis movement representations or deficient transfer of information from

praxis movement representations across the corpus callosum. The next two research

questions were aimed at illuminating which of these two processes contributes to the

ideomotor and conceptual apraxia in individuals with AD.

The individuals with AD in this study demonstrated limb apraxia with the left hand

that was similar to the limb apraxia in the right hand described in previous studies. Like

the participants with AD in previous studies, the individuals with AD in this study

demonstrated impaired performance on verbal command pantomime and conceptual

pantomime tasks (Ochipa et al., 1992; Schwartz et al., 2000; Travniczek-Marterer et al.,

1993). On verbal command pantomime tasks, previous studies have reported that

individuals with AD produce more body part as tool errors than healthy elderly

individuals (Kato et al., 2000) and more spatial and temporal than content errors for

transitive gestures (Foundas et al., 1999) (all of the stimuli in this study required

transitive gestures, i.e., required using a tool to act on an object) and the individuals with

AD in this study showed these same characteristics with both the right and the left hands.

With regards to the conceptual pantomime task, the AD group made more total errors,

more conceptual errors, and more unrecognizable errors than the HC group with both

hands.

Research Question 2

What is the contribution of degraded praxis movement and conceptual

representations (due to cortical atrophy) to the limb apraxia in individuals with AD?

Summary. This issue was examined by comparing right hand performance of each

group on the verbal command pantomime and conceptual pantomime tasks. A significant









difference in performance (with the right hand) between the two groups on the verbal

command pantomime task would provide evidence that praxis movement representations

are degraded in individuals with AD. Statistical analysis of the data revealed that there

was a significant difference (with the right hand) between the two groups on the verbal

command pantomime task indicating that praxis movement representations are degraded

in individuals with AD. A significant difference in performance (with the right hand)

between the two groups on the conceptual pantomime task would provide evidence that

action semantics representations are degraded in individuals with AD. Statistical analysis

of the data revealed that there was a significant difference (with the right hand) between

the two groups on the conceptual pantomime task providing evidence that action

semantics representations are degraded in individuals with AD. Based on these findings,

there is evidence to suggest that both praxis movement representations and action

semantic representations are degraded in individuals with AD.

Explanation. Right hand performance was examined to answer this question

because right hand performance does not require the transfer of praxis information across

the corpus callosum but requires within hemisphere access to praxis information. The

results of this study have supported the notion that praxis movement representations and

action semantics representations in the left hemisphere are degraded such that individuals

with AD demonstrate ideomotor and conceptual apraxia in both hands. Several studies

have found cortical atrophy in the temporal and parietal lobes in individuals with AD

(Foundas et al., 1996; Halliday et al., 2003; Pantel et al., 2004; Thompson et al.,2001;

Thompson et al., 2003). Because these areas are important for praxis information

processing (production and conceptual), it is likely that the apraxia in the right hand of









individuals with AD can be attributed to cortical atrophy in the regions that subserve

praxis production and conceptual information processing. However, the question remains

whether the disruption of interhemispheric transfer of praxis information also contributes

to the ideomotor and conceptual apraxia in the left hand of individuals with AD.

Therefore, it was necessary to examine the role of interhemispheric transfer of different

types of praxis information and these analyses could potentially refute the proposed

localization of these functions within the left hemisphere of individuals with AD.

Research Question 3

What is the contribution of interhemispheric disconnection (due to callosal atrophy)

to apraxia in individuals with AD?

Summary. This issue was examined by comparing the disparity or asymmetry

between right hand and left hand performance of the two groups on praxis production and

conceptual tasks. A significant difference in praxis asymmetry between the two groups

on the verbal command pantomime task and the pantomime imitation task would indicate

that information from praxis movement representations is not being transferred across the

corpus callosum in individuals with AD. Statistical analysis of the data revealed that

there was a significant difference in praxis asymmetry between the two groups on the

verbal command pantomime task and the pantomime imitation task indicating that praxis

movement representations are not being adequately transferred across the corpus

callosum in individuals with AD. A significant difference in praxis asymmetry between

the two groups on the conceptual pantomime task would indicate that information from

action semantics representations is not being transferred across the corpus callosum in

individuals with AD. Statistical analysis of the data revealed that there was not a

significant difference in praxis asymmetry between the two groups on the conceptual









pantomime task indicating that right hemisphere motor areas are able to access

information from action semantics representations. There is evidence from these

analyses to suggest that information from praxis movement representations is not being

transferred across the corpus callosum in individuals with AD. Additional findings

suggest that the right hemisphere is able to access information from action semantics

representations in individuals with AD.

Explanation. The disparity between right hand and left hand performance was

used to answer this question because praxis performance with the left hand relies on

interhemispheric transfer of praxis information. A loss of pyramidal neurons in the third

cortical layer that project to analogous areas of the contralateral hemisphere via the

corpus callosum results in atrophy of specific regions of the corpus callosum in

individuals with AD. With regards to the corpus callosum, several studies have reported

a reduction in the total area of the corpus callosum while other studies have suggested

reductions in specific regions of the corpus callosum (Biegon et al., 1994; Black et al.,

2000; Pantel et al., 1998; Teipel et al., 1998; Teipel et al., 1999). Based on patients

described by Kazui and Sawada (1993), Watson and Heilman (1983), and Degos et al.

(1987), fibers in the anterior portion of the corpus callosum are thought to be important

for interhemispheric transfer of praxis information. Evidence from Weis et al. (1991)

indicated a significant decrease in volume of the anterior corpus callosum without a

significant decrease in volume of the posterior corpus callosum. Hampel et al. (1998)

also noted decreased area in the most rostral (genu and anterior body) and caudal

(splenium) regions of the corpus callosum without reduction of the posterior body. These









findings suggest that degeneration of callosal fibers could interfere with the

interhemispheric transfer of praxis information in individuals with AD.

The results of this study indicated a disruption in the transfer of information from

praxis movement representations but not in the transfer of information from action

semantics representations. It could be that the callosal disconnection that results in

deficient transfer of praxis information is specific to production information such that

information specific to the conceptual attributes of the movement is able to be transferred

across the corpus callosum while information specific to the temporal and spatial

specifications of the movement is not being adequately transferred across the corpus

callosum. Alternatively, these findings might provide evidence that praxis movement

representations are localized within the left hemisphere but actions semantics

representations may have a bihemispheric distribution that allows the right hemisphere to

access praxis conceptual information. So even if praxis conceptual information that is

stored in the left hemisphere cannot be transferred across the corpus callosum, the right

hemisphere may be able to access whatever action semantics representation is needed to

complete a given task. These findings suggest that it is likely that the conceptual apraxia

in both hands of individuals with AD is related solely to degraded action semantics

representations due to bilateral cortical atrophy while the ideomotor apraxia in the left

hand of individuals with AD can be attributed to a combination of degraded left

hemisphere praxis movement representations and deficient interhemispheric transfer of

praxis information. However, we have not addressed whether the ideomotor apraxia in

individuals with AD could result from an intrahemispheric verbal motor disconnection or

whether it is the verbal input or motor representations that are not being adequately









transferred across the corpus callosum in individuals with AD. The final research

question will address these two issues.

Research Question 4

Is the disruption of praxis information transfer a result of an intrahemispheric

verbal motor disconnection or an interhemispheric corpus callosum disconnection?

Summary. Right hand performance of both groups on the verbal command

pantomime and pantomime imitation tasks were compared to address this issue. A

significant difference between right hand performance of the two groups on these tasks

would suggest that impaired performance of individuals with AD results from an

intrahemispheric verbal motor disconnection. The results of this analysis were significant

for the main effect of group but not task and there was not a significant interaction

between group and task. This suggests that the impaired performance of individuals with

AD on the verbal command pantomime and pantomime imitation tasks results from an

interhemispheric disconnection rather than an intrahemispheric disconnection. These

results also imply that the interruption of praxis movement representation transfer across

the corpus callosum is not dependent on the transfer of verbal information.

The asymmetry between right and left hand performance of individuals with AD on

the verbal command pantomime and pantomime imitation tasks was also compared. A

significant difference between these two tasks would provide further evidence that

individuals with AD are unable to transfer information from praxis movement

representations across the corpus callosum. There was a significant difference in praxis

asymmetry of individuals with AD on the verbal command pantomime and pantomime

imitation tasks indicating that the presence of verbal input cannot account for the









deficient transfer of movement information across the corpus callosum in individuals

with AD.

Explanation. Thus far, we have examined the mechanisms of praxis information

transfer using the verbal command pantomime task. The verbal command pantomime

task requires processing of verbal information and an interaction between language

processing centers and praxis movement representations. The pantomime imitation task

requires processing of visual information and involves solely the transfer of information

from praxis movement representations to motor areas for movement execution.

Therefore in order to determine whether verbal input interferes with interhemispheric

transfer of praxis information it was necessary to compare the performance of individuals

with AD on these two tasks.

If individuals with AD demonstrated impaired performance on the verbal command

pantomime task but not the pantomime imitation task, this would indicate that verbal

input was interfering with interhemispheric transfer of praxis information. However, the

individuals with AD in this study demonstrated impaired performance on both the verbal

command pantomime and pantomime imitation tasks meaning that the information that is

unable to cross the corpus callosum is motor in nature (i.e. information from praxis

movement representations that contain the temporal and spatial specifications of a

movement). Furthermore, if the disparity or asymmetry between hands is significantly

different in individuals with AD, this would provide further evidence that information

from praxis movement representations is inadequately transferred across the corpus

callosum. If verbal command performance was more asymmetric than pantomime

imitation performance (based on group mean asymmetry), this would be considered an









intrahemispheric verbal motor disconnection but because pantomime imitation

performance was more asymmetric than verbal command pantomime performance (based

on group mean asymmetry), it can be concluded that there is an interhemispheric callosal

motor disconnection in individuals with AD.

Conclusions

According to the results of this study, individuals with AD have conceptual and

ideomotor apraxia in both the dominant (right) and nondominant (left) hands. Based on

the finding of a significant difference in right hand performance on the verbal command

pantomime and conceptual pantomime tasks, it can be concluded that praxis movement

representations and action semantics representations are degraded in individuals with AD

and that degraded movement representations contribute to the ideomotor apraxia while

degraded semantic representations contribute to the conceptual apraxia in individuals

with AD. Previous studies have provided evidence of neuronal loss in the areas of the

brain that govern skilled movement systems which likely results in the degradation of

praxis movement and conceptual representations. A significant difference in the

asymmetry of performance of the two groups on the praxis production tasks but not the

praxis conceptual task indicates that deficient transfer of praxis information across the

corpus callosum contributes to the ideomotor but not the conceptual apraxia in the left

hand of individuals with AD. Other studies have suggested neuronal loss in the cortical

layers that project to contralateral motor areas (i.e., corpus callosum) and this could

explain the deficient transfer of praxis production information.

So while the apraxia in the right hand of individuals with AD can be attributed to

degraded representations, a combination of degraded movement representations and

deficient interhemispheric transfer of praxis information most likely explains the









ideomotor apraxia in the left hand. According to the results of this study, the conceptual

apraxia in the left hand of individuals with AD is related to the degradation of action

semantics representations but left hand performance on the conceptual pantomime task

was better than performance on the imitation or command tasks because the right

hemisphere can still access action semantics information. This may be because deficient

callosal transfer of information is specific to the transfer of spatial and temporal

information or because semantic information is represented in a distributed network that

encompasses both hemispheres.

Additionally, the study findings suggested that the disconnection in individuals

with AD is an interhemispheric callosal motor disconnection rather than an

intrahemispheric verbal motor disconnection meaning that verbal information does not

interfere with the interhemispheric transfer of praxis information. Deficient

interhemispheric transfer is specific to the transfer of information from praxis movement

representations. Future studies will investigate the particular types of information from

praxis movement representation that are unable to cross the corpus callosum by using a

discriminate analysis of the error data.

The purpose of this study was to investigate how praxis information processing is

represented in the brain by examining the transfer of different types of praxis information

across the corpus callosum. The findings of this study support the notion that praxis

movement representations are localized in the left hemisphere of right handed individuals

and suggest the conclusion that action semantics representations are distributed across

both hemispheres. In addition, it can be surmised that only information from praxis

movement representations is transferred across the corpus callosum and that information









from the input modality must access praxis movement representations prior to the

interhemispheric transfer of praxis information. In conclusion, it is evident that the

interaction between left hemisphere praxis movement representations and right

hemisphere motor areas is necessary for left hand movement precision and intact praxis

movement representations and action semantics representations are necessary for

bimanual movement precision. All of these outcomes provide significant contributions to

the study of praxis mechanisms.

The fact that measurements of cortical and callosal volumes were not obtained and

that these measurements were not correlated with the presence of limb apraxia is a

potential weakness of this study. The absence of this data limits the ability to draw

conclusions about the neuroanatomical correlates of the praxis mechanisms described in

this study. However, this does not negate the value of the findings of the current study

because it is possible that individuals with AD can still demonstrate limb apraxia in both

hands without showing radiological evidence of cortical or callosal atrophy. While it is

possible to measure the volume of cortical and callosal structures, the integrity of the

pyramidal cells in the cortical layers that are responsible for transferring information

within and between hemispheres is not measurable while the individual is living

(examination of senile placques and neurofibrillary tangles in the cortical layers requires

post-mortem analysis). Therefore, regardless of the presence or absence of cortical

and/or callosal atrophy in individuals with AD limb apraxia in both hands may still be

present due to underlying neuranatomical processes that cannot be adequately examined.

Additionally, it should be noted that an attempt was made to balance the two

groups for age and education level. While the experimenter succeeded in matching the









two groups for age, there was a significant difference in education level between the two

groups. Unfortunately, low education level is one of the predictive factors for

development of this type of dementia. Finding healthy elderly control subjects that

matched subjects in the AD group for both age and education level was difficult.

Typically, in the AD population older individuals have low levels of education while in

the HC population older individuals have high levels of education. Because low

education level is a predictive factor for developing AD, most individuals within the age

group studied (i.e., 60-90 yrs old) who had low levels of education had developed

symptoms of AD so healthy elderly individuals that were recruited because they matched

the AD subjects for age and gender had higher levels of education.

Lastly, it is important to point out that general cognitive decline in the AD group is

a potential alternative explanation for the findings of this study. It is possible that

individuals with AD demonstrated impaired performance on limb praxis tasks because

their overall cognitive abilities are affected by AD and not because of degraded praxis

representations or deficient interhemispheric transfer of praxis information. In this study,

the design attempted to control for the effects of general cognitive decline on limb praxis

performance by enrolling individuals in the earlier stages of the disease process and by

excluding individuals who exhibited cognitive deficits that would interfere with their

performance on the experimental tasks (like visual object agnosia and severe auditory

comprehension deficits). Therefore, it is not likely that the effect of general cognitive

decline had a significant impact on these results.

Implications

Rothi and Homer (1982) described two theories of physiologic mechanisms of

recovery that can be applied to rehabilitation of individuals with neurologic disease or









injury. Restitution of function "suggests that as the lesion area heals neural pathways

resume activities and the functions subserved by the involved neural systems are

restored" (p. 74). Substitution of function "suggests that the brain is physiologically

capable of spontaneous restoration of function beyond the acute phase of recovery

through substitution and reorganization of neuronal structures" (p. 74). Behavioral

treatment approaches consistent with a restitution-of-function model are based on the

idea that "functions are lost or impaired following brain damage and that lost function

must be retrained and impaired functions must be maximally stimulated in order to be

maintained" (p.77). Behavioral treatment approaches consistent with a substitution-of

function model are based on the idea that "the clinician treats that patient by discouraging

the use of ineffective strategies while encouraging the use of new strategies not

previously available to him" (p. 78).

Studies that have attempted to treat individuals with limb apraxia can be subdivided

into approaches that are consistent with a restitution-of-function model (Butler, 1997;

Maher, Rothi, & Greenwald, 1991; Ochipa, Maher, & Rothi, 1995; Smania, Girardi,

Domencali, Lora, & Aglioti, 2000; Wilson, 1988) and approaches that are consistent with

a substitution-of-function model (Donkervoort, Dekker, Stehmann-Saris, & Deelman,

2001; Goldenberg, Daumuller, & Hagmann, 2001; Goldenberg & Hagmann, 1998; van

Heugten, Dekker, Dellman, van Dijk, Stehlmann-Saris, &Kinebanian, 1998). Those

studies that aimed to restore praxis functions usually demonstrated improvement on

outcome measures but the improvement was typically limited to gestures targeted during

the treatment. Strategy training in individuals with apraxia was designed to teach

strategies to compensate for apraxia rather than rehabilitate the apraxia itself and these









approaches tended to be successful in instituting compensatory strategies that allow the

patient to function more independently despite the persistence of apraxia. For individuals

with AD, who have apraxia that impacts their ability to perform everyday activities

independently, perhaps a combination of these two treatment methods would be useful.

Additionally, because the results of this study have shown the individuals with AD have

apraxia in both the right and left hands, assessment of apraxia should include the

examination of both right hand and left hand performance and apraxia treatment should

comprehensively be aimed at improving function in both hands.

With respect to AD, limb apraxia continues to be an important area of study. This

study has shown that many areas of praxis function are impacted by the cognitive decline

that characterizes AD. Since numerous studies have shown the impact of limb apraxia on

this population and the resultant burdens associated with its presence (Foundas et al.,

1995; Giaquinto et al., 1999; Saeki et al., 1995), research into the nature, assessment and

treatment of this disorder in individuals with AD should continue to be vigorously

pursued.















APPENDIX A
LIST OF STIMULI


stimulus
# Task 1- VC Task 2- PI Task 3- CP
1 Show me how you would hold and imitate using a pictured object:
use a paddle to play ping pong. paddle to play ping ping pong ball and
pong table
2 Show me how you would insert a imitate inserting a pictured object:
plug into an electrical outlet. plug into an electrical outlet
electrical outlet
3 Show me how you would hold and imitate shaving your pictured object:
use a razor to shave your face. face unshaven face


4 Show me how you would hold and imitate lighting a pictured object:
use a match to light a candle. candle with a match unlit candle


5 Show me how you would hold and imitate using a pictured object:
use a screwdriver to turn a screw screwdriver to turn a screw sticking out
into the wall. screw into the wall of a piece of wood
6 Show me how you would thil /,' a imitate throwing a pictured object:
bowling ball. bowling ball upright bowling
pins
7 Show me how you would hold and imitate ironing a pictured object:
use an iron to press a shirt. shirt ironing board with
a shirt on it
8 Show me how you would beat a imitate drumming pictured object:
drum with a drumstick. drum set
9 Show me how you would hold and imitate turning eggs pictured object:
use a spatula to turn eggs in a frving with a spatula skillet with eggs in
pan. it
10 Show me how you would hold and imitate spreading pictured object:
use a knife to spread butter on bread, butter on bread with piece of bread with
a knife butter on it
11 Show me how you would hold and imitate using a paint pictured object:
use a paint roller to paint a wall. roller to paint a wall paint roller pan










stimulus
# Task 1- VC Task 2- PI Task 3- CP
12 Show me how you would hold and imitate using a pictured object:
use a spoon to eat a bowl of soup. spoon to eat soup bowl of soup
13 Show me how you would hold and imitate using a pictured object:
use a paintbrush to paint on an easel. paintbrush to paint painter's easel and
on a canvas palette
14 Show me how you would hold and imitate using a pictured object:
use a paintbrush to paint a wall in paintbrush to paint a open paint can
front of you. wall
15 Show me how you would hold and imitate using a knife pictured object:
use a knife to carve a turkey. to carve a turkey whole turkey


16 Show me how you would / thi ,' a imitate throwing a pictured object:
dart at a dart board, dart dart board
17 Show me how you would hold and imitate stirring pictured object:
use a spoon to stir a cup of coffee. coffee cup of coffee and
open packet of
sugar
18 Show me how you would hold and imitate sawing wood pictured object:
use a saw to cut wood on a wood on a
sawhorse. sawhorse
19 Show me how you would hold and imitate using a pictured object:
use a match to light a fire in a match to light a fire wood in a fireplace
ti, e7,Ltn e
20 Show me how you would hold and imitate using lipstick pictured object:
use lipstick to paint your lips. lips with partial
lipstick
21 Show me how you would hold and imitate cutting and pictured object: cut
use a spatula to cut and serve cake. serving cake bundt cake


22 Show me how you would use a jack imitate using a jack pictured object: car
to lift a car that had a flat tire. to fix a flat tire with a flat tire


23 Show me how you would hold and imitate using a fork pictured object:
use a fork to eat dinner, to eat dinner plate of food on
table











stimulus
# Task 1- VC Task 2- PI Task 3- CP
24 Show me how you would hold and imitate brushing pictured object:
use a toothbrush to brush your teeth. your teeth dirty teeth

25 Show me how you would hold and imitate hammering a pictured object:
use a hammer to pound a nail into nail into a wall nail sticking out of
the wall. a piece of wood
26 Show me how you would hold and imitate unlocking a pictured object:
use a key to unlock a door. door with a key keyhole and
doorknob
27 Show me how you would hold and imitate removing a pictured object:
use a hammer to remove a bent nail bent nail from wood bent nail in a piece
from a piece of wood. with a hammer of wood
28 Show me how you would hold and imitate scooping pictured object:
use a shovel to scoop sand into a sand into a bucket sandbox with sand
bucket. with a shovel and pail
29 Show me how you would hold and imitate combing pictured object:
use a comb to fix your hair. your hair messy hair


30 Show me how you would thread a imitate threading a pictured object:
needle. needle spool of thread and
a button
31 Show me how you would hold and imitate using a pictured object:
use a turnkey to open a can of turnkey to open a partially opened
sardines, can of sardines sardine can
32 Show me how you would hold and imitate sharpening a pictured object:
use a pencil sharpener to sharpen a pencil broken pencil
broken pencil.
33 Show me how you would hold and imitate opening a pictured object:
use a bottle opener to open a soda soda bottle with a soda bottle
bottle. bottle opener
34 Show me how you would hold and imitate using a pictured object:
use a screwdriver to open a can of screwdriver to open closed paint can
paint. a paint can
35 Show me how you would turn offa imitate turning off a pictured object:
dripping faucet. dripping faucet dripping faucet
36 Show me how you would th/l ,i' a imitate throwing a pictured object:
baseball to the catcher. baseball baseball catcher










stimulus
# Task 1- VC Task 2- PI Task 3- CP
37 Show me how you would hold and imitate chopping a pictured object:
use a hatchet to chop a log. log with a hatchet partially chopped
log
38 Show me how you would hold and imitate using tongs pictured object: ice
use tongs to serve ice. to serve ice bucket and glass
39 Show me how you would hold and imitate using an ice pictured object: ice
use an ice cream scoop to serve ice cream scoop to serve cream and cone
cream. ice cream
40 Show me how you would hold and imitate erasing a pictured object:
use an eraser to clean a chalkboard, chalkboard scribbles on chalk
board
41 Show me how you would hold and imitate using a pictured object:
use a wrench to turn a bolt. wrench to turn a bolt hexhead bolt


42 Show me how you would hold and imitate shooting a pictured object:
use a gun to shoot at a target. gun human shaped
target
43 Show me how you would roll up a imitate rolling up a pictured object:
car window, car window partially opened
car window
44 Show me how you would hold and imitate using pictured object:
use clippers to trim a rose stem. clippers to trim a rose and vase
rose
45 Show me how you would hold and imitate cutting paper pictured object:
use scissors to cut a piece ofpaper. with scissors partially cut out
paper doll














APPENDIX B
DESCRIPTION OF ERRORS


Content
P Perseverative- The patient produces all or part of a previously produced
pantomime.
R Related- The pantomime is an accurately produced pantomime associated in
content to the target. For example, the participant might pantomime playing
a trombone for a target of a bugle.
N Nonrelated- The pantomime is an accurately produced pantomime not
associated in content to the target. For example, the participant might
pantomime playing a trombone for a target of shaving.
H Hand- The patient performs the action without benefit of a real or imagined
tool. For example, when asked to cut a piece of paper with scissors, they
pretend to rip the paper. Another example would be turning a screw by
hand rather than with an imagined screwdriver.
Temporal
S Sequencing- Some pantomimes require multiple positioning that are
performed in a characteristic sequence. Sequencing errors involve any
perturbation of this sequence including addition, deletion, or transposition
of movement element as long as the overall movement structure remains
recognizable.
T Timing- This error reflects any alteration from the typical timing or speed
of a pantomime. May include abnormally increased, decreased, or irregular
rate of production.
O Occurrence- Pantomimes may characteristically involve either single (i.e.
unlocking a door with a key) or repetitive (i.e. screwing in a screw with a
screwdriver) movement cycles. This error reflects any multiplication of
characteristically single cycles or reduction of a characteristically repetitive
cycle to a single event.
Spatial
A Amplitude- Any amplification, reduction, or irregularity of the
characteristic amplitude of a target pantomime.
IC Internal Configuration-This error type reflects any abnormality of the
required finger/hand posture and its relationship to the target tool. For
example, when asked to pretend to brush teeth, the participant may close the
hand tightly into a fist with no space allowed for the imagined toothbrush
handle.
BPT Body Part as Tool- The patient uses finger, hand, or arm as the imagined
tool of the pantomime. For example, when asked to pretend to smoke a
cigarette, the participant might puff on the end of an extended index finger.









Spatial
EC External configuration- This error type reflects any abnormality of the
required finger/hand posture and its relationship to the target object. For
example, when asked to pretend to brush teeth, the participant might hold
his hand next to his mouth without reflecting the distance necessary to
accommodate an imagined toothbrush.
M Movement- When acting on an object with a tool, a movement
characteristic of the action and necessary to accomplishing the goal is
required. Any disturbance of the characteristic movement of the action.
For example, when asked to pantomime using a screwdriver, a participant
may orient the imagined screwdriver correctly to the imagined screw but
instead of stabilizing the shoulder and wrist while twisting at the elbow, the
participant stabilizes the elbow and twists at the wrist or shoulder.
Other
C Concretization- The participant performs a transitive pantomime not on an
imagined object but instead on a real object not normally used in the task.
For example, when asked to pretend to saw some wood, they pantomime
sawing on their leg.
NR No Response
UR Unrecognizable Response- A response that is not recognizable and shares
no temporal or spatial features of the target.




Full Text

PAGE 1

INTERHEMISPHERIC TRANSFER OF PRAXIS INFORMATION USING PROBABLE ALZHEIMER'S DISEASE AS A MODEL FOR DISCONNECTION APRAXIA By ANN MARIE KNIGHT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Ann Marie Knight

PAGE 3

This dissertation is dedicated to my lovi ng husband, Travis W. Knight, Ph.D. Without his unwavering support and encouragement, th is research study could not have been completed.

PAGE 4

iv ACKNOWLEDGMENTS I would like to begin by ex tending tremendous gratitude and empathy to the study participants and their caregive rs. Without their willingness to sacrifice their time, this research could not have been completed. I would also like to thank my parents, Marianne a nd Bernard Cimino and Dannis and Frances Knight for their infinite s upport and wisdom throughout this process. Without their love, enthusiasm, encouragem ent and help caring for my son, Connor, I truly would not have been able to complete this academic endeavor. To my friend and mentor, Leslie J. Gon zalez Rothi, Ph.D., I would like to extend immense gratitude and appreciation for he r extensive efforts throughout my academic career. For the past five years, she has spent countless hours with me on the design and implementation of numerous research projec ts, has taught me how to apply academic knowledge to clinical practice, and has provide d me with an example of a compassionate and knowledgeable speech pathologist. Without her extensive knowledge of neuropsychology, communication disorders and research design, the writing of this manuscript could not have come to fruition. Her endless dedication to educating students has impacted my life in numerous ways. I ha ve learned a great deal through her expertise and supervision. Further acknowledgements should be exte nded to Kenneth Heilman, MD, Leilani Doty, Ph.D. and the fellows and staff of th e University of Florida Memory Disorder Clinic not only for their assistance in recr uiting participants for this study but for

PAGE 5

v imparting their knowledge about the diagnosis and treatment of persons with memory disorders. Special thanks are extended to Dr. Heilman for all of his advice on the design and implementation of this experiment. It ha s truly been an honor a nd a privilege to be mentored by someone who has so greatly c ontributed to the fi eld of behavioral neurology. I am extremely honored to have st udied with him and to call him not only my mentor but also my friend. Additionally, I would like to acknowledge th e other members of my committee, Dr. Russell M. Bauer and Dr. Christine M. Sapienza for their support of th is project and their commitment to research excellence. My expe rience at the University of Florida has been enhanced greatly by having had the opportunity to work with these extremely talented individuals. I would like to extend a special thanks to Cristina Posse, MHS and Lauren Meffen, BA, for their willingness to collaborate with me on this project. They were responsible for spending countless hours anal yzing all of the data for this project, and this research truly could not have been completed without their dedication and pe rseverance. I feel privileged to have been given the opport unity to work with two such bright and outstanding students. In additi on, special thanks are extended to Haijing Qin, M.S. of the VA RR&D Rehabilitation Outcomes Research Ce nter for excellent statistical support. Statistical analysis of these data would not have been possible wit hout her expertise and advice. To the support staff at the Univers ity of Florida Departme nt of Neurology (i.e. Doug Perkinson) and the VA RR&D Brain Reha bilitation Research Center (i.e. Susan Nadeau, Joy McCallum, and Lisa Demanuel), I would like to extend si ncere gratitude for providing excellent clerical support. Lastly, to th e health reporters at The Gainesville Sun

PAGE 6

vi and The Ocala Star Banner for helping me to recruit participants by publishing study announcements free of charge and to the Gain esville, FL Alzheimers Association for being proactively involved with this population and for bei ng willing to support research endeavors. This study was funded by 1) the VA Office of Academic Affiliations and Patient Care Services Predoctoral Fellowship in Speech Pathology, 2) the VA Rehabilitation Research and Development Office Centers of Excellence Brain Rehabilitation Research Center, 3) the National Institute of Deaf ness and Communication Disorders, National Institutes of Health, 4) the Fl orida Department of Elder Affa irs, Memory Disorder Clinic and 5) the University of Florida, Departme nt of Communication Sc iences and Disorders and Department of Neurology.

PAGE 7

vii TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 What is Limb Apraxia?.................................................................................................2 Three Types of Disconnection Ap raxia: Literature Review.........................................7 What is Alzheimers Disease?....................................................................................12 What is Known About Limb Apraxia in AD?............................................................15 Why Study Disconnection Apraxia in AD?................................................................19 Summary.....................................................................................................................22 Purpose, Questions, and Hypotheses..........................................................................24 Research Question 1............................................................................................25 Research Question 2............................................................................................26 Research Question 3............................................................................................27 Research Question 4............................................................................................28 2 METHODS.................................................................................................................31 Subjects....................................................................................................................... 31 Inclusion Criteria.................................................................................................31 Subject Demographics.........................................................................................33 Sample Size Estimation..............................................................................................34 Sample Size Estimation-2 Samples Equal Variances (most conservative estimate)...........................................................................................................34 Sample Size Estimation-2 Samples Un equal Variances (least conservative estimate)...........................................................................................................35 Experimental Tasks....................................................................................................35 Data Collection Procedures.................................................................................35 Task 1: Verbal Command Pantomime (VC).......................................................36 Task 2: Pantomim e Imitation (PI).......................................................................37 Task 3: Conceptual Pantomime (CP)..................................................................38 Rater Training.............................................................................................................40

PAGE 8

viii Reliability...................................................................................................................4 1 Statistical Analysis......................................................................................................42 Research Question 1............................................................................................43 Research Question 2............................................................................................43 Research Question 3............................................................................................43 Research Question 4............................................................................................44 3 RESULTS...................................................................................................................49 Subject Demographics................................................................................................49 Neuropsychological Screening...................................................................................49 Reliability...................................................................................................................5 1 Descriptive Statistics..................................................................................................53 Task 1: Verbal Command Pantomime (VC).......................................................53 Task 2: Pantomim e Imitation (PI).......................................................................53 Task 3: Conceptual Pantomime (CP)..................................................................54 Error Types Task 1, 2, and 3...............................................................................55 Statistical Analysis......................................................................................................56 Research Question 1............................................................................................56 Research Question 2............................................................................................56 Research Question 3............................................................................................57 Research Question 4............................................................................................57 Summary.....................................................................................................................58 4 DISCUSSION.............................................................................................................67 Summary and Explanation of Findings......................................................................68 Research Question 1............................................................................................68 Research Question 2............................................................................................70 Research Question 3............................................................................................72 Research Question 4............................................................................................75 Conclusions.................................................................................................................77 Implications................................................................................................................80 APPENDIX A LIST OF STIMULI.....................................................................................................83 B DESCRIPTION OF ERRORS....................................................................................87 LIST OF REFERENCES...................................................................................................89 BIOGRAPHICAL SKETCH.............................................................................................97

PAGE 9

ix LIST OF TABLES Table page 1-1: Diagnostic criteria for AD..........................................................................................29 2-1: Individual subject de mographics for AD group.........................................................45 2-2: Individual subject de mographics for HC group.........................................................46 2-3: Strength of observer agreement fo r ranges of kappa statistic values.........................47 3-1: Scores for screening measures for individual subjects in HC group..........................59 3-2: Scores for screening measures for individual subj ects in AD group..........................60 3-3: Inter-rater reliability using % agreement and the Kappa statistic for task 1, 2, and 3.....61 3-4: Intra-rater reliability using % agreement and the Kappa statistic for task 1, 2, and 3.....62 3-5: Response accuracy (percent) data with difference scores and asymmetry ratios for individual subjects in HC group for Tasks 1, 2, and 3.............................................63 3-6: Response accuracy (percent) data with difference scores and asymmetry ratios for individual subjects in AD group for Tasks 1, 2, and 3.............................................64 3-7: Error analysis descriptive data for Task 1 (VC).........................................................65 3-8: Error analysis descriptive data for Task 2 (PI)...........................................................65 3-9: Error analysis descriptive data for Task 3 (CP)..........................................................66 3-10: Error totals for tasks 1, 2, and 3................................................................................66

PAGE 10

x LIST OF FIGURES Figure page 1-1: Cognitive neuropsychologica l model of limb apraxia................................................30 2-1: Examples of pictures used in the Florida Action R ecall Test (FLART)....................48

PAGE 11

xi Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INTERHEMISPHERIC TRANSFER OF PRAXIS INFORMATION USING PROBABLE ALZHEIMER'S DISEASE AS A MODEL FOR DISCONNECTION APRAXIA By Ann Marie Knight May 2005 Chair: Leslie J. Gonzalez Rothi Major Department: Communica tion Sciences and Disorders Praxis, or the ability to perform skilled movements, is essential to independent living. Most skilled movements require the us e of both hands and the inability to perform skilled movements effectively can significan tly impact quality of life. Despite the importance of being able to perform skilled movements effectively and efficiently, little is known about how the brain pr ocesses praxis movement in formation. What is known about this type of processing has been learne d from an ablation model. However, this model impairs the motor function of the cont ralesional limb and does not allow the study of bimanual praxis mechanisms. The purpose of this study was to investigate how praxis information processing is represented in the brain by examining the interhemispheric transfer of different types of praxis info rmation. This was accomplished by examining bimanual praxis mechanisms in individuals wi th Alzheimers disease because individuals in this population can perform praxis task s with both hands and demonstrate both limb apraxia and neural degeneration. This model allowed us to study how praxis information

PAGE 12

xii is transferred between the two brain hemispheres and differentiate what type of praxis information is being transferre d across the corpus callosum. In order to accomplish the goals of th is study, it was necessary to confirm the presence of limb apraxia in individuals with Alzheimers disease. This study also attempted to determine whether the limb apraxia that is present in this population is due to the degradation of left hemisphere move ment representations or the interruption of interhemispheric transf er of praxis information. Anot her purpose of this study was to differentiate whether the interhemispheric disc onnection of praxis information was due to the inability to transfer verbal or motor in formation across the corpus callosum. Findings indicated that individuals with Alzheimers disease have ideomotor and conceptual apraxia in the nondominant hand and that information from praxis movement representations in the left hemisphere are not transferred across the corpus callosum adequately in individuals with Alzheimers disease.

PAGE 13

1 CHAPTER 1 INTRODUCTION Humans use skilled movement in nearly every aspect of independent functioning from preparing food to getting dressed to usi ng hands and arms to gesture in combination with verbal communication. When the ability to perform skilled movements is disrupted functional independence can be severely co mpromised. Skilled movement is extremely important to everyday life and movement precision in both hands is necessary for effective and efficient completion of acti ons. Despite the im portance of skilled movement, little is known about how the prax is system that governs skilled movement execution is organized in the br ain. To perform skilled moveme nts, motor cortex in each hemisphere must access praxis movement repres entations that are thought to be localized in the left hemisphere. To perform skilled movements with the left hand, information from left hemisphere praxis movement repr esentations must be transferred across the corpus callosum to right hemisphere motor cortex. To perform skilled movements with the right hand, information does not need to cr oss the corpus callosum but rather must be transferred to left hemisphere motor cort ex by intrahemispheric connection fibers. Previous studies have typically relied on an ablation model to study the mechanisms of the praxis system. However, unilatera l stroke patients commonly demonstrate contralesional hemiplegia, which may mask th e presence of apraxia in the weak hand. Therefore, stroke does not provide an ideal model for studying bimanual praxis mechanisms, which rely on the transfer of motoric and conceptual praxis information across brain hemispheres. Because we use bot h hands to perform skilled movements, it is

PAGE 14

2 necessary to study prax is mechanism using a model that allows us to study bimanual performance. The purpose of this study is to investigate how praxis information processing is represented in the brain by ex amining the interhemispheric transfer of different types of praxis information. This chapter defines limb apraxia and provides a rationale for studying praxis mechanisms using Alzheimers disease (AD) as a pathological model. What is Limb Apraxia? Limb apraxia is an acquired disorder of skilled, learned, purposive movements resulting from neurologic disease or injury that cannot be explaine d by language deficits or primary sensorimotor disturbance (Mah er & Ochipa, 1997; Rothi & Heilman, 1997). In order to perform movements, sensory input (auditory, tactile, visual) must interact with stored movement representations that are tran slated into patterns of innervation. Both disconnection (Geschwind, 1965; Liepmann, 1980) and representational (Rothi, Ochipa, & Heilman, 1991) models of apraxia have been proposed. Liepmann (1980) (as described by Rothi, Ochipa, and Heilman, 1997a) proposed that in right handed individuals, the left he misphere guides skilled movements of both the left and right hands and that the acquisiti on of skilled limb movements required the acquisition of movement formulae, innervator y patterns, and kinetic memories for learned movements. Liepmann proposed that movement formulae contain spatial and temporal patterns for the production of move ment sequences. Innervatory patterns are acquired through practice and provide a me thod for transforming movement formulae into muscle innervation patte rns for correct limb positioning. Kinetic memories are associations between innervatory patterns fo r action, which are highly practiced and can be performed without spa tial or visual feedback.

PAGE 15

3 Geschwind (1965a, 1965b) also proposed th at skilled movements of both hands were mediated by the left hemisphere in ri ght handed individuals. He suggested that pantomime to command requires processing by left hemisphere language mechanisms. For right handed movements, information is transferred from left hemisphere language areas to left motor association cortex fo r programming of movements and left primary motor cortex for motor innervation of the right hand. For left handed movements, information is transferred from left hemisphe re language areas to right hemisphere motor cortex via the corpus callosum for motor inne rvation of the left hand. Disconnections can occur which interfere with transfer of inform ation from left hemisphere language areas to left motor cortex for control of the right hand (left hemisphere lesions) or right motor cortex for control of the left hand (corpus callosum lesions). According to a representational model of limb apraxia developed by Rothi et al. (1991) (Figure 1-1), the praxis system can be divided into con ceptual and production subsystems. This model can account for disso ciations in praxis performance including separate systems for receptive and expressive praxis, selective di ssociation of sensory input modalities from praxis movement repr esentations, a direct route for praxis imitation, and the notion that ther e is a separate system for ac tion semantics (Rothi et al., 1991). The production system can be divided into a store of learned spatial-temporal movement representations or praxicons and a mechanism for translating these representations into motor programs or innervatory patterns. state that: In order to perform a skilled learned act, one must place particular body parts in certain spatial positions in a specific order at specific times. The spatial positions assumed by the relevant body parts depend not only on the nature of the act but also on the position and size of an extern al object with which the body parts must

PAGE 16

4 interact. Skilled acts also require orderl y changes in the spatial positions of the body parts over time. These movement formulas command the motor systems to adopt the appropriate spa tial positions of the rele vant body parts over time. (Heilman & Rothi, 1993, p. 146). A disruption of the production system, or ideomotor apraxia, is characterized predominantly by spatiotemporal errors dur ing pantomime to command and imitation of gestures (Poizner, Mack, Verfaille, Rothi, & Heilman, 1990; Rothi, Mack, Verfaellie, Brown, & Heilman, 1988). Performance in ideomotor apraxia may improve with manipulation of the actual tool as a result of in creased tactile and visual cuing as well as contextual information. However, there is some evidence that actu al tool use remains defective (Poizner, Soechting, Bracewell, Roth i, & Heilman, 1989). Sp atial errors are the most characteristic errors of ideomotor apraxi a and there are three forms of spatial errors (Poizner et al., 1990; Rothi et al., 1988). Postural errors re flect an abnormality of the required finger or hand posture a nd its relationship to the target tool (Rothi et al., 1997b). Spatial orientation errors occur when the hand movement that is produced does not appear to direct the tool toward an imagin ed object (Heilman & Ro thi, 1993). Spatial movement errors are disturbances of the characteristic joint movements necessary to produce the correct action (He ilman & Rothi, 1993). Temporal errors in ideomotor apraxia may occur in the form of a delay in the initiation of movement, occasional pauses during the movement, or a failure to coordi nate the speed of th e movement with the spatial components of the movement (Heilman & Rothi, 1993). Thus, ideomotor apraxia may result from deficits of action implemen tation or degradation of praxis movement representations (Heilman, Ro thi, & Valenstein, 1982; Roth i, Heilman, & Watson, 1985). In the first case, the patien t is able to recognize gestur es but gesture production is impaired. This is thought to result from an impaired ability to execute skilled movements

PAGE 17

5 despite intact movement representations. In the second case, the pati ent is not able to recognize gestures and gesture pr oduction is impaired. This is thought to result from a degradation of the movement representati ons (Cimino-Knight, Hollingsworth, Maher, Raymer, Foundas, Heilman, & Rothi, 2002). The conceptual subsystem involves thr ee types of knowledge: knowledge of tool and object functions, knowledge of actions i ndependent of tools, and knowledge about the organization of single actions into goal or iented sequences (i.e., action semantics) (Rothi et al., 1991; Roy & Square, 1985). A tool is used to provide a mechanical advantage in an action and an object is the recipient of an action (Rothi et al., 1997a). Knowledge of the functions of objects and tool s may have internalized linguistic referents and externalized function know ledge (Roy & Square, 1985). The internalized linguistic referents contain semantic desc riptions of objects and actions The externalized function knowledge provides information about the percep tual attributes of the object and action and the environmental context in which tools are used. It has been proposed that the semantic system has specialized subsystems for all modalities and modes of processing which contain specific conceptual represen tations (i.e., action semantics, verbal semantics, visual semantics, auditory semantics). These multiple semantic systems are thought to communicate with each other such that visual or verbal input can result in action output (Rothi et al, 1997a; Raymer & Ochipa, 1997).Conceptual apraxia is a disruption of the conceptual system that interf eres with the knowledge of tool and object functions and their associated actions (Ochipa, Rothi, & Heilman, 1992). Patients with conceptual apraxia cannot recall the type of actions associated with specific tools or objects and thus exhibit conten t errors (DeRenzi & Lucchel li, 1988; Ochipa, Rothi, &

PAGE 18

6 Heilman, 1989; Ochipa et al, 1992 ). Other types of errors th at are possible in patients with conceptual apraxia include the inability to recall which tool is associated with an object, lack of awareness of the mechanical advantage of partic ular tools, or the inability to create novel tools to solve mechanical problems (Heilman & Rothi, 1993). It is hypothesized that the basic defi cit underlying conceptual apraxia is a degradation of action semantics (Schwartz, Adair, Raym er, Williamson, Crosson, Rothi, Nadeau, & Heilman, 2000). Ideational apraxia is a disruption of the c onceptual system that interferes with knowledge about the organization of single act ions into sequences. Patients with ideational apraxia demonstrate an inability to carry out a series of actions and have difficulty sequencing actions in the proper order (Heilman & Rothi, 1993). Essentially, ideational apraxia is a loss of ability to conceptualize, plan, and execute a complex sequence of motor actions involving the use of tools or objects (LeClerc & Wells, 1998). Lesions that disconnect various forms of sensory input from praxis movement representations have also been described in the literature (DeRenzi Faglioni, & Sorgato, 1982; Gazzaniga, Bogen, & Sperry, 1967; Geschwind & Kaplan, 1962; Heilman, 1973). Gazzaniga et al. (1967), Geschwind and Ka plan (1962), Geschwind (1965a, 1965b) and Heilman (1973) described individuals w ho demonstrated a disconnection between language areas necessary for comprehension of commands and movement representations necessary for selecting and programming the appropriate actions (Heilman & Rothi, 1993). Furthermore, DeRenzi et al. (1982) describe d modality-specific apraxias that result from a disconnection of praxis movement representations and specific sensory input

PAGE 19

7 (visual, verbal, or tactile) These types of dissociati on apraxia are known as verbalmotor, visuo-motor, and tactile-motor dissoci ation apraxias (Heilman & Rothi, 1993). Callosal apraxia refers to apraxia that is more severe in the left hand than in the right hand due to a lesion of the corpus callo sum (in some patients apraxia may be absent in the right hand) (Geschwind, 1965; Geschw ind & Kaplan, 1962; Gr aff-Radford, Welsh, & Godersky, 1987; Watson & Heilman, 1983). This type of lesion disconnects the movement representations and action semantic s in the left hemisphere (selection of appropriate actions from a st ore of learned movement patterns) from right hemisphere motor association areas and primary motor co rtex (programming of innervatory patterns for movements of the left hand). This can be explained by an inte raction between left hemisphere localization of praxis movement representations and cont rol of the left hand and arm by contralateral primary motor corte x. Because the left hand is controlled by right primary motor cortex, lesions of the co rpus callosum disconnect the left hand from left hemisphere movement and semantic represen tations. Therefore, if there is damage to or degeneration of the corpus callosum, right primary motor cortex may not be able to access left hemisphere praxis movement representations and action semantics. Three Types of Disconnection Apraxia: Literature Review Thus far an overview of what is known about the types and mechanisms of limb apraxia has been presented. The majority of studies of limb apraxia to date have used unilateral stroke patien ts to investigate intrahemispheric transfer of praxis information. These studies have examined praxis performance in the ip silesional (left) hand only due to the presence of contralesional hemipleg ia. This population does not provide an adequate model for studying interhemispheric tr ansfer of praxis information. Following is a discussion of what is known about the transfer of praxis information across the

PAGE 20

8 corpus callosum from individuals with callosal lesions. Acco rding to the literature, at least three types of apraxia ar e possible as a result of ca llosal disconnection: ideomotor apraxia, conceptual apraxia, and verbal-mot or dissociation apraxia (Degos, Gray, Louarn, Ansquer, Poirier, & Barbizet, 1987; Gazzaniga et al., 1967; Geschwind, 1965; Geschwind & Kaplan, 1962; Goldenberg, Wimmer, Holzner, & Wessely, 1985; GraffRadford et al., 1987; Kazui & Sawada, 1993; Tanaka, Iwasa, & Obayashi, 1990; Watson & Heilman, 1983). Heilman (1973) described three individuals with left hemisphere lesions who could not perform actions to command with either hand but could imitate gestures and use objects flawlessly with both hands. When asked to pantomime to command the participants appeared as if they did not understand the comm and (p.862) but the spared ability to imitate gestures a nd use objects suggests that the engrams for motor sequences are intact (p.863). Heilman (1973) explained th e deficit in these individuals as a deficit in the transfer of inform ation between language comprehension and motor encoding. Similarly, Geschwind and Kaplan (1962) and Gazzaniga et al. (1967) reported individuals with callosal lesi ons who could not perform acti ons to verbal command with the left hand but could imitate gestures a nd use objects. Based on the patient described by Geschwind and Kaplan (1962), Geschwi nd (1965a, 1965b) hypothesized that a lesion of the corpus callosum would result in the disconnection of right hemisphere motor cortex from left hemisphere language processi ng areas. This would re sult in the inability to perform pantomime to command actions with the left hand, while gesture imitation and actual object use are relatively preserved. This has been interpreted as a disconnection between language areas necessary for co mprehension of commands and movement

PAGE 21

9 representations necessary for selecting and programming the appropriate actions (Heilman & Rothi, 1993). The patient described by Gra ff-Radford et al. (1987) was similar to those described by Geschwind and Kaplan (1962) and Gazzaniga et al. (1967) in that she demonstrated impaired pantomime to command but relatively spared gesture imitation with the left hand. Praxis performance in this individua l, however, was worst when she held the object in her hand and attempted to perfor m the action. The authors explained the deficits in this individual as result ing from a verbal-motor disconnection. The patient described by Watson and Heilm an (1982) experienced a lesion of the corpus callosum that was vascular in nature. The anterior extent of the lesion was at the junction between the genu and body while the po sterior one-fourth to one-fifth of the body and splenium as well as the supplementary and cingulate cortex remained intact. In the course of recovery from the lesion, this patient demonstrated conceptual, ideomotor, and verbal-motor dissociation ap raxia. Initially, the patien t was unable to pantomime to command, imitate gestures or use objects with the left hand and could not demonstrate the intent of actions. Because she was una ble to demonstrate that she understood the intent of the required action, this was consider ed evidence of conceptual apraxia. During the course of recovery, the ability to imita te gestures and use objects improved (although spatiotemporal movement errors were presen t), but pantomime to command with the left hand remained impaired. Because the patien t was able to imitate gestures and use objects, this was considered evidence of a verbal-motor dissociation apraxia. Finally, praxis testing showed that pantomime to command, gesture imitation and object use improved with the left hand and the patient demo nstrated the correct in tent of actions but

PAGE 22

10 continued to make spatiotemporal moveme nt errors. Because praxis performance remained impaired for all tasks with the left hand, this was considered evidence of persistent ideomotor apraxia. Furthermore, DeRenzi et al. (1982) also described individual s who demonstrated modality-specific apraxias. These individua ls performed better with certain input modalities (visual, verbal, or tactile). For ex ample, six participants performed better with tactile and visual input than with verbal input and six pa tients performed better with verbal and tactile input than with visual inpu t (this could not be at tributed to receptive aphasia or visual agnosia). Two participan ts who performed more poorly with tactile input than verbal or visual input were al so reported. To explain this disconnection between modality-specific input pathways and the cen ter where movements are programmed, it has been stated that in the majority of patients the lesion w ill result in apraxia appearing in every modality, either because it destroys the pr ogramming center or because it interrupts all the pathways connecting it with ot her sensory or motor areas; however, a discrete injury may well isolate the pr ogramming center from one type of information and render the patient unable to execute the gesture when it is elicited by a given sensory center but capable of performing it under the guidance of other modalities. (DeRenzi et al., 1982, p. 310). As evidenced by patients described by Gazzaniga et al. (1967), Geschwind and Kaplan (1962), and Graff-Radfor d et al. (1987), disruption of the transfer of information from left hemisphere language processing cen ters and praxis move ment representations from right hemisphere motor areas that cont rol the left hand results in verbal-motor dissociation apraxia. Furthe rmore, the patient described by Watson and Heilman (1983) demonstrated that conceptual and ideomotor forms of apraxia are possible from a lesion of the corpus callosum. Finally, DeRenzi et al. (1982) provided evidence that apraxia can be modality-specific.

PAGE 23

11 The literature from individuals with callo sal lesions has provided evidence that several different types of praxis informati on are transferred across the corpus callosum for skilled movements of the left hand. Moveme nt representations that are translated into innervatory patterns and action semantics info rmation that guides the selection of the appropriate action must be transferred from left hemisphere praxis areas to right hemisphere motor areas across the corpus callo sum. In addition, the input modality (such as verbal input), must interact with both praxis movement repr esentations and action semantics for the producti on of skilled movement. As mentioned previously, un ilateral stroke does not provide a favorable model for studying bimanual praxis mechanisms due to th e presence of hemiplegia. The literature described above has utilized individuals with callosal lesions to investigate interhemispheric transfer of praxis informa tion at the single case level of evidence. Because callosal lesions are rare and heteroge neous, this population also does not provide the best model for studying the mechanisms of praxis information transfer at the clinical trial level of evidence. This study propos es to use individuals with AD to study interhemispheric transfer of praxis info rmation because this population can perform praxis tasks with both hands ( unlike individuals with unilatera l strokes), this disease is prevalent among the elderly popu lation (unlike specific callos al lesions), and there is evidence of callosal atrophy in the areas of the corpus callo sum that are suspected to carry praxis information (similar to individua ls with callosal lesions). Following is an overview of the symptoms, diagnosis, and pa thology of AD, a summary of what is known about limb apraxia in AD and a review of th e literature regarding cal losal atrophy in AD.

PAGE 24

12 This will lead to a rationale for studying interhemispheric transfer of praxis information using praxis asymmetries in individua ls with AD as a pathological model. What is Alzheimers Disease? Alzheimers disease (AD) is a degenerative disease of the cent ral nervous system (Boller & Duyckaerts, 1997). It is characterized clinically by progressive dementia and cognitive decline and histological ly by senile placques and ne urofibrillary tangles. There are many factors that are thought to contribute to the develo pment of AD; advanced age, genetic predisposition, the pres ence of the apolipoprotein E4 allele, gender (female/male ratio, 2:1), low education level and previous head trauma have been implicated (Barclay, Zemcov, Blass, & Sanson, 1985; Rocca, Bonaiut o, Lippi, Luciani, Turtu, Cavarzeran, & Amaducci, 1990). A clinical diagnosis of AD requires the pr esence of dementia, c ognitive decline and functional impairment. The Diagnostic and Statistical Manu al of Mental Disorders (DSM-IV) and the National Inst itute of Neurological and Communicative Disorders and Stroke (NINCDS) and the Alzheimers Disease and Related Disorders Association (ADRDA) (NINCDS/ADRDA) have published crit eria for the diagnosis of AD (Table 11) (McKhann, Drachman, Folstein, Katzman, Price, Stadlan, 1984). The DSM-IV defines dementia as the development of multip le cognitive deficits that include memory impairment and at least one of the following: ap hasia, apraxia, agnosia or a disturbance in executive functioning. According to the NINCDS/ADRDA criteria a diagnosis of probable AD requires the following criteria: dementia established by neurologic examination and documented by objective testing, deficits in two or more cognitive areas, progressive worsening of memory and other cognitive functions, no disturbance in consciousness, absence of systemic disorders or other brain diseases that could account

PAGE 25

13 for the progressive deficits in memory a nd cognition, and onset between 40 and 90 years of age. The diagnosis of probable AD is supported by progressive deficits in language (aphasia), perception (agnosia), and motor skil ls (apraxia), impaired activities of daily living and altered patterns of behavior, family history of similar disorders, and consistent laboratory results. Possible AD is diagnosed when the patient has a variation in the typical presentation of dementia or when another potentially dementing disorder is present but is not the primary source of the dementia symptoms. Definite AD is reserved for clinically diagnosed patients with hist opathological confirmation by cerebral biopsy or autopsy. Histopathologic evidence of AD as confir med by cerebral biopsy or postmortem autopsy requires the presence of senile plac ques and neurofibrillar y tangles. Senile placques (SP) are extracellular amyloid de posits (extracellular byproducts of neuronal degeneration) (Afifi & Bergman, 1998; Guilme tte, 1997). Neurofibri llary tangles (NFT) are intracellular aggregates of cytoskeletal filaments (tangles of fine fibers found in cell bodies) (Afifi & Bergman, 1998; Guilme tte, 1997). The NFTs represent the accumulation of abnormal components of the neuronal cytoskeleton that form paired helical filaments (Hof & Mo rrison, 1999). In individuals with AD, SPs and NFTs are morphologically and topographically distinct, have different histological compositions, and are present in specific cortical areas a nd layers. Specifically, pyramidal neurons in Layer II and III of the cortex project to othe r ipsilateral and contralateral cortical areas, respectively, via intraand interhemispheric projection fi bers including the corpus callosum. In the cortex, SPs and NFTs ar e found in all cortical areas and are numerous in layer III of the cortex. So the presence of SPs and NFTs in la yer III coul d potentially

PAGE 26

14 disrupt interhemispheric transfer of neuronal signals. Theref ore, cognitive functions that require interhemispheric transfer of informa tion across the corpus callosum, like praxis, could possibly be impaired in this patient population. Clinically, patients with AD typically progr ess through three stages of the disease process (Boller & Duyckaerts 1997). As the individual with AD progresses through these stages, significant cognitive decline occurs resulting in decreased functional independence. The first stage, amnestic is characterized by semantic and episodic memory impairments and the presence of aphasia. The second stage, dementia, involves a progressive decline in intelle ctual abilities that significantl y impacts the ability to live independently. The third stage, vegetative, is characterized by the inability to perform activities of daily living as well as an in ability to express wants and needs through communication. During the second and third st ages, memory and language become more impaired, significantly impacting the individua ls ability to communicate and remember. The person is not able to understand verbal in structions or communi cate basic needs and may become disoriented in familiar places and unable to recognize familiar people. Additionally, individuals at these stages of AD may demonstrate ideomotor, ideational, conceptual and constructional apraxia, which in terfere with the ability to manipulate tools and objects in the environment. Individua ls with AD experience significant cognitive decline, decreased functional independence, and the need for more supervised care, which ultimately increase th e costs of their care. Cognitive decline has a significant impact on impairment of functional abilities in individuals with AD. Functional impairment in individuals with AD, evidenced by the loss of the ability to perform activities of da ily living (ADLs) and in strumental activities

PAGE 27

15 of daily living (IADLs), has a major impact on the quality of life of patients and caregivers and is an important predictor of inst itutionalization (Canad ian Study of Health and Aging, 1994). The presence of amnesi a, aphasia, apraxia, agnosia, and/or visuospatial impairments cont ribute to functional disability and functional disability contributes to dependence, wh ich in late stages of AD may result in institutionalization (Tekin, Fairbanks, OConnor, Rosenberg, & Cummi ngs, 2001). It is bene ficial to health care providers to understand both the neur obehavioral mechanisms and clinical implications of cognitive deficits, like limb apraxia, in order to understand how these cognitive deficits, especially limb apraxia, interfere with the ability to function independently in individuals with AD. What is Known About Limb Apraxia in AD? Limb apraxia is prevalent in all stages of the disease process in individuals with dementia and the presence of limb apraxia has been demonstrated to have a significant impact on functional abilities In a study by Edwards, De uel, Baum, and Morris (1991), 22% of subjects with suspected dementia, 47.1% of patients with mild dementia, 58.6% of participants with moderate dementia, a nd 98.1% of individuals with severe dementia demonstrated evidence of limb apraxia. Severa l studies have shown that limb apraxia has a significant impact on the ability to pe rform activities of da ily living (Foundas, Macauley, Raymer, Maher, Heilman, & Rothi, 1995; Giaquinto, Buzzelli, DiFrancesco, Lottarini, Montenero, Tonin, & Nolfe, 1999; Saeki, Ogata, Okubo, Takahashi, & Hoshuyama, 1995). In a study of individuals with AD (Cho, Cho, Cho, Choi, Oh, & Bae, 2001), 56.6% of participants were dependent for one or more ADLs including bathing (54.7%), dressing (47.2%), and feeding ( 5.7%), and for IADLs patients with AD demonstrated dependence in cooking ( 66.0%), cleaning (64.2%), housework (79.2%),

PAGE 28

16 and laundry (71.7%), all of whic h require skilled movement and the ability to manipulate tools and objects (praxis). There is also evidence that individuals with AD who develop apraxia may decline more rapidly (Yesavage, Brooks, Taylor, Ti nklenberg, 1993) and that apraxia may be more predictive of early d eath than aphasia or amnesia (Burns, Lewis, Jacoby, & Levy, 1991). Furthermore, studies of limb apraxia in acute stroke have shown that the presence of limb apraxia is a significant predictor of failure to return to work (Saeki et al., 1995), poor functional recovery followi ng stroke (Giaquinto et al ., 1999), and poor performance of IADLs (Foundas et al., 1995). There is significant evidence that the presence of limb apraxia has an impact on functional abilities in indi viduals with AD (Cho et al., 2001; Foundas et al., 1995; Giaquinto et al., 1999; Saeki et al., 1995). Func tional impairment has been shown to be a predictor of institutionalizati on and thus increases the costs of care for individuals with AD. A review of investigations of limb apraxia in individuals with AD indicated the presence of 3 types of apraxia: ideomotor apraxia, ideational apraxia, and conceptual apraxia. Ideomotor apraxia has been reported freque ntly as a cognitive sequela of AD (Della Sala, Lucchelli, & Spinnler, 1987; Deroue sne, Lagha-Pierucci Thibault, BaudouinMadec, Lacomblez, 2000; Foundas et al ., 1999; Giannakopoulos, Duc, Gold, Hof, Michel, & Bouras, 1998; Jacobs, Adair, W illiamson, Na, Gold, Foundas, Shuren, Cibula, & Heilman, 1999; Kato, Meguro, Sato, Shimada, Yamazaki, Saito, Yamaguchi, & Yamadori, 2000; Rapcsak, Croswell, & Rube ns, 1989; Travniczek-Marterer, Danielczyk, Simanyi, & Fischer, 1993; Willis, Behrens, Mack, & Chui, 1998). According to recent

PAGE 29

17 literature, individuals with AD demonstrat e impaired performance with the dominant (right) hand on both gesture to verbal comma nd and imitation tasks (Travniczek-Marterer et al., 1993). The severity of dementia has an impact on praxis performance in individuals with AD (Foundas et al., 1999) and praxis perfor mance degrades with the progression of AD (Della Sala et al., 1987). With regards to e rror types, individuals with AD produce more content (100%) than spatialtemporal (0%) errors with intransitive pantomimes and more spatial-temporal (96%) than content (4%) errors with transitive pantomimes when the dominant (right) hand is tested (Foundas et al., 1999). Patients with AD also produce significantly more body-pa rt-as-tool responses with the right hand when compared with normal controls (Kato et al., 2000). The curre nt literature on limb apraxia in AD has examined ideomotor a nd conceptual apraxia in the dominant hand only. None of the previously published studies have examined the left hand performance of individuals with AD on praxis production or conceptual tasks. This study proposes to investigate the mechanisms of left hand prax is performance in right handed individuals with AD. It should also be noted that the majority of studies examining ideomotor apraxia in AD have not typically utilized a standardiz ed battery for the assessment of praxis. Furthermore, most of the studies scored res ponses as either correct or incorrect and did not analyze error types. Therefore, a clinical ly efficient and standardized praxis measure might be helpful in the assessm ent of individuals with AD. Studies of conceptual apraxia in AD have focused on determining the characteristics of the disorder and attempting to clarify the nature of the semantic system (Dumont, Ska, & Joanette, 2000; Ochipa et al., 1992; Schwartz et al., 2000). Thus far 54

PAGE 30

18 individuals with AD have been tested for c onceptual apraxia by va rious authors (Dumont et al., 2000; Ochipa et al., 1992; Schwartz et al., 2000) and 50 /52 (96%) participants were found to have deficits of the praxis conceptual system. Ochipa et al. (1992) hypothesized that th ere could be three types of conceptual apraxia in individuals with AD due to disrupt ions of different cognitive mechanisms of the praxis conceptual system. First, ther e may be a loss of knowledge of the type of actions associated with tools or objects (tool-object action kn owledge) resulting in content errors in tool use. Second, there may be an inability to a ssociate tools with the appropriate objects (tool-objec t associative knowledge) lead ing to the inappropriate selection of tools. Finally, there may be impairment in the ab ility to understand the mechanical nature of problems and the mechanical advantages of particular tools (mechanical knowledge) leading to an inabi lity to solve mechanical problems and an inability to develop novel tools. In this study (Ochipa et al., 1992), the 32 participants with AD were divided into four experiment al subgroups: good ideomotor praxis without semantic language impairment, poor ideo motor praxis without semantic language impairment, good ideomotor praxis with sema ntic language impairment, poor ideomotor praxis with semantic language impairment. Each element of the praxis conceptual system mentioned above was tested in the pa tients and controls. The results indicated that individuals with AD have an impairment of the praxis conceptual system and that conceptual apraxia can be differentiated from both ideomotor apraxia and semantic language deficits. Additionally, AD participants were significantly impaired in all three proposed domains of the praxis conceptual system (tool-object action knowledge, toolobject associative knowledge, and mechanical knowledge) so it is not known if these

PAGE 31

19 three components of the praxis conceptual system are functionally or neurologically distinct. Rapcsak et al. (1989) examined ideati onal apraxia in individuals with AD by testing serial actions requiring the use of several objects to achieve an intended goal (i.e. prepare a cup of instant coffee with cream and sugar). The se rial actions were scored by counting the number of com ponent actions correctly ex ecuted in the appropriate sequence. When compared to controls, partic ipants with AD were significantly impaired on measures of ideational apraxia. Why Study Disconnection Apraxia in AD? Individuals with AD loose pyramidal neurons from layer III of the cortex that project to the corpus callosum from analogous areas of the contralateral hemisphere. Therefore, in addition to the presence of id eomotor, ideational, and conceptual apraxia with the dominant hand, it is likely th at patients with AD will demonstrate interhemispheric disconnecti on syndromes that include the presence of ideomotor, conceptual, and verbal-motor dissociati on apraxias with the nondominant hand. Several studies have found co rtical atrophy in the tempor al and parietal lobes in individuals with AD (Foundas, Eure, & Selt zer, 1996; Halliday, Double, & Macdonald, 2003; Pantel, Schonknecht, Essig, & Schroder, 2004; Thompson, Hayashi, Zubicaray, Janke, Rose, Semple, Herman, Hong, Dittmer, Doddrell, & Toga, 2003; Thompson, Mega, Woods, Zoumalan, Lindshield, Blant on, Moussai, Holmes, Cummings, & Toga, 2001). Atrophy of these regions co rrelates with the cognitive symptoms that are seen in the early stages of AD (i.e. apraxia and aphasi a). Of interest in this study is atrophy of areas that are critical to spoken lang uage processing and praxis movement representations. Thompson et al. (2003) found highly significan t decreases in gray matter

PAGE 32

20 in bilateral temporal and parietal corti ces and that atrophy of these regions was asymmetric with greater atrophy of left hemisphere as compared to the right hemisphere. Additionally, the precentral a nd postcentral gyri (important for execution of movement and perception of movement) were relative ly spared compared with the parietal association cortex located immediately poste rior. Pantel et al. (2004) also found a significant decrease in tempor al and parietal cortical vol ume bilaterally and showed a correlation between left tempor al and parietal volumes and performance on tests of naming and praxis. Thompson et al. (2001) not ed relative sparing of occipital cortex bilaterally suggesting preserve d processing of visual sensor y input in individuals with dementia. These findings explain the presence of apraxia in the dominant hand of right handed individuals with AD but would not be sufficient to e xplain a right hand to left hand asymmetry in praxis performance. Studies that have measured the corpus callosum in individuals with AD have found atrophy in specific regions (Janowsky, Kaye & Carper, 1996; Lyoo, Satlin, Lee, & Renshaw, 1997; Pantel, Schroder, Jauss, Essig, Minakaran, Schonknecht, Schneider, Schad, Knopp, 1999; Teipel, Hampel, Alexande r, Schapiro, Horwitz, Teichberg, Daley, Hippius, Moller, & Rapoport, 1998; Verm ersch, Roche, Hamon, Daems-Monpeurt, Pruvo, Dewailly, & Petit, 1996; Vermersch, Sche ltens, Barkhof, Steinling, & Leys, 1993; Weis, Jellinger, & Wenger, 1991). However, these reports have yielded conflicting results regarding which areas of the corpus callosum are decreased in AD. Several studies have reported a reduction in the total area of the corpus cal losum in individuals with AD as compared to normal controls (Biegon, Eberling, Richardson, Roos, Wong, Reed, & Jagust, 1994; Black, Moffat, Yu, Pa rker, Stanchev, & Bronskill, 2000; Hampel,

PAGE 33

21 Teipel, Alexander, Horwitz, Teichberg, Sc hapiro, & Rapoport, 1998; Pantel, Schroder, Essig, Minakaran, Schad, Friedlinger, Jauss, & Knopp, 1998; Teipel, Bayer, Alexander, Zebuhr, Teichberg, Kulic, Schapiro, Moller, Rapoport, & Hampel, 2002; Teipel, Hampel, Pietrini, Alexander, Horwitz, Daley, Moller, Schapiro, & Rapoport, 1999). Teipel and colleagues (2002, 1999) reported a significant re duction in the area of the rostrum and splenium with sparing of the body of the co rpus callosum while others have reported significant reductions in the genu (Biegon et al., 1997; Black et al., 2000) and body (Lyoo et al., 1997; Black et al., 2000). Hampel et al. (1998) noted decreased area in the most rostral and most caudal regions of the corpus callosum in patients with AD with no reduction of the posterior body. Weis et al. (1991) attempted to differentiate callosal degeneration patterns in normal aging and AD. Results indicated a significant decrease in the anterior portions (rostrum, genu, anteri or body) of the corpus callosum with no change in the posterior por tions (posterior body, isthmus, and genu) in normal aging. However, in individuals with AD, a signi ficant decrease in th e body of the corpus callosum occurred with no change in th e anterior and poste rior portions. Furthermore, the patients described by Kazui and Sawada (1993) and Watson and Heilman (1983) demonstrated apraxia that wa s more severe when performing gestures with the left hand than the ri ght hand due to a lesion of the anterior portion of the body of the corpus callosum. The case reported by Degos et al. (1987) presented with left apraxia without agraphia following a lesion of the pos terior portion of the body and splenium of the corpus callosum. This dissociation s uggests that callosal fibers for writing are concentrated in the posterior portion of the corpus callo sum while callosal fibers for

PAGE 34

22 praxis are concentrated in the anterior por tion of the corpus callosum (Kazui & Sawada, 1993). Although a systematic investig ation of the interhemisphe ric transfer of praxis information using AD as a pathological mode l has not been completed to date, several studies have reported differences in praxis performance with the right hand (dominant) versus the left hand (nondominant) in this population (Ball, Lant os, Jackson, Marsden, Scadding, & Rossor, 1993; Derouesne et al., 200 0; Willis et al., 1998). Derouesne et al. (2000) found that praxis performance was better with the right hand than with the left hand in patients with AD. Willis et al. (1998) found that while performance accuracy between the right and left hands was not signi ficantly different, gesture response latencies were significantly longer for the AD group when the left hand was used. Rapcsak et al. (1989) found no difference in praxis performa nce between the right and left hands in individuals with AD. Furthermore, due to the presence of contra lesional hemiplegia, stroke does not provide an ideal model for st udying the praxis abilities of the left and right hands independently. Therefore, a dise ase process which affects the fibers of the corpus callosum that transfer praxis inform ation across the hemispheres would provide a superior model for studying apraxia asymmetrie s. Because there is evidence of callosal atrophy in patients with AD, this disease ma y provide a more useful model for studying interhemispheric transfer of praxis information. Summary Alzheimers disease (AD) is a costly and debilitating condition. It causes numerous cognitive and behavioral impairments including limb apraxia. Limb apraxia is a disorder that disrupts skilled movements of the arms and hands and has a negative affect on the performance of activities of daily living. Thus far, the mechanisms of limb

PAGE 35

23 apraxia have primarily been studied in groups of individuals with uni lateral strokes while the mechanisms of interhemispheric transfer of praxis information have primarily been studied in single cases of indivi duals with specific callosal lesions. Unilateral stroke does not provide a good model for the study of bimanual praxis mechanisms because the presence of hemiplegia in these individuals in terferes with the examination of praxis in the contralesional hand. Callosal lesions provide an excellent model for studying bimanual praxis mechanisms but these lesi ons are extremely rare and physiologically heterogeneous and therefore this population is not well suited for a group study. Perhaps AD could provide a comparable model for st udying the mechanisms of interhemispheric transfer of praxis information. It has been shown that individuals wi th AD demonstrate limb apraxia (like unilateral stroke patients) and display callo sal degeneration (like the callosal lesion patients). Individuals with AD are able to us e both hands to perform praxis tasks and AD is a fairly common diagnosis within the el derly population. For these reasons, AD is presented as a potentially superior model for studying interhemispheric transfer of praxis information; specifically production, conceptual and verbal-motor praxis information. The corpus callosum is responsible for tr ansferring information from one cerebral hemisphere to the other. The movement re presentations that govern skilled movement and the semantic representations that relate sensory input to moto r output are thought to be localized in the left hemisphere. In orde r for the left hand to correctly perform skilled movements praxis representati ons and action semantics in the left hemisphere must be transferred to right motor cortex via the corpus callosum. Therefore, if individuals with AD have degeneration of the corpus callosum fibe rs that transfer praxis information from

PAGE 36

24 the right hemisphere to the left hemisphere and if individuals with AD have been shown to have different types of limb apraxia, it can be hypothesized that individuals with AD will demonstrate disconnection apraxia of th e ideomotor, conceptual, and verbal-motor types (i.e. better performance of praxis tasks with the ri ght hand than the left hand). Purpose, Questions, and Hypotheses The purpose of the present study is to investigate how praxis information processing is represented in the brain by exam ining the transfer of different types of praxis information from praxis movement repr esentations in the left hemisphere to motor cortex in the right hemisphere across the co rpus callosum. This will be accomplished by examining bimanual praxis performance in in dividuals with AD because individuals in this population can perform praxis tasks with both hands (i.e. they do not have hemiplegia), they are prevalent within the elderly population (i.e. this is not a rare syndrome), and they demonstrate both limb apraxia and callosal atrophy (i.e. can potentially differentiate what type of information is bei ng transferred via the corpus callosum). First, it will be necessary to confirm th at individuals with AD demonstrate limb apraxia. Second, this study will attempt to determine whether the limb apraxia that is present in individuals with AD is due to th e degradation of left hemisphere movement representations or the interruption of interh emispheric transfer of praxis information across the corpus callosum. Third, examination of the transfer of praxis conceptual and production information will provide information about what types of praxis information are dissociated due to the degradation of callo sal fibers in individua ls with AD. Finally, this study will attempt to differentiate whet her the interhemispheric disconnection of praxis production information is due to th e inability to transfer verbal or motor

PAGE 37

25 information across the corpus callosum in i ndividuals with AD. The following research questions will address each of these issues. Research Question 1 Do individuals with AD have conceptual an d/or ideomotor apraxia in the left hand? Hypothesis. This research question will be examined by comparing the left hand performance of individuals with AD to the left hand performance of healthy elderly individuals on a verbal command pantomime ta sk and a conceptual pantomime task. Previous studies provide evidence that in dividuals with AD have conceptual and ideomotor apraxia in the dominant (right) ha nd but it is also necessary to examine the presence of conceptual and ideomotor apra xia in the nondominant (left) hand. If there were a significant difference between i ndividuals with AD and healthy elderly individuals on the verbal command task (left hand), this would suggest the presence of ideomotor apraxia in the AD group. It is pr edicted that there w ill be a significant difference between the two groups for left hand performance on the verbal command pantomime task (i.e., individuals with AD will demonstrate ideomotor apraxia in the left hand). If there were a significant differen ce between individuals with AD and healthy elderly individuals on the concep tual pantomime task (left hand), this would suggest the presence of conceptual apraxia in the AD gr oup. It is predicted that there will be a significant difference between the two groups for left hand performance on the conceptual pantomime task (i.e., individua ls with AD will demonstrate conceptual apraxia in the left hand). Only left hand pe rformance is being compared to answer this question because left hand performance requir es the recruitment of both left hemisphere praxis movement representations and right hemisphere motor areas which requires the transfer of praxis movement information ac ross the corpus callosum. Further questions

PAGE 38

26 will address the contributions of degraded movement representations and interhemispheric callosal disconnection to the apraxia in indi viduals with AD. Research Question 2 What is the contribution of degraded praxis movement and conceptual representations (due to corti cal atrophy) to the limb apra xia in individuals with AD? Hypothesis. This issue will be examined by co mparing the right hand performance of individuals with AD to the right hand perf ormance of healthy elde rly individuals on a verbal command pantomime task and a con ceptual pantomime task. If there is a significant difference between individuals with AD and healthy elderly individuals on the verbal command task (right hand), it can be assumed that the praxis movement representations are degraded in individuals with AD. The pred iction is that there will be a significant difference between the two groups for right hand performance on the verbal command pantomime task (i.e., there will be evidence of degraded movement representations in individuals with AD). If there is a significant difference between individuals with AD and healt hy elderly individuals on the conceptual pantomime task (right hand), it can be assumed that the praxis conceptual representations are degraded in individuals with AD. The prediction is that there will be a significant difference between the two groups for right hand performance on the conceptual pantomime task (i.e., there will be evidence of degraded praxis conceptual representations in individuals with AD). The comparison of right hand performance answers this question because right hand performance does not require the transfer of praxis information across the corpus callosum but requires within hemisphere ac cess to praxis movement representations. Further questions will address the role of interhemispheric disconnection in the transfer

PAGE 39

27 of different types of praxis information acro ss the corpus callosum in individuals with AD. Research Question 3 What is the contribution of interhemisphe ric disconnection (due to callosal atrophy) to the limb apraxia in individuals with AD? Hypothesis. The disparity or asymmetry betw een right hand and left hand performance of individuals with AD and hea lthy elderly individuals on praxis production and conceptual tasks will be compared to answer this research question. If the performance asymmetry of the two groups on the conceptual pantomime task is significantly different, the conclu sion would be that praxis co nceptual information is not being transferred across the corpus callosum in individuals with AD. The prediction is that performance asymmetry of the two gr oups on the conceptual pantomime task will not be significantly different (i.e. there will not be eviden ce of a callosal disconnection that is specific to praxis conceptual in formation in individuals with AD). If the performance asymmetry of the two groups on the verbal command pantomime task and the pantomime imitation task is significantly different, the conclusion would be that information from praxis movement represen tations is not being transferred across the corpus callosum in individuals with AD. Th e prediction is that performance asymmetry of the two groups on the verbal command pantomime and pantomime imitation tasks will be significantly different (i.e., there will be ev idence of a callosal disconnection that is specific to praxis movement information in individuals with AD). Because the verbal command pantomime task requires transfer of language and motor information, it is necessary to attempt to differentiate whether verbal information or motor information is being interrupted by the proposed callosal disconnection in individuals with AD.

PAGE 40

28 Research Question 4 Is the disruption of praxis information transfer a result of an intrahemispheric verbal-motor disconnection or an interhem ispheric corpus callosum disconnection? Hypothesis. Answering this question will involve two comparisons. First, right hand performance of both groups on a verb al command pantomime task and a pantomime imitation task will be compared. If there were a significant difference between the two groups for right hand performance on these two ta sks, this would suggest that impaired performance of individuals with AD results from an intrahemispheric verbal motor disconnection. If there were not a significant difference be tween right hand performance of the two groups on these two tasks, this would suggest that impaired performance of individuals with AD results from an interh emispheric callosal disc onnection. Second, the asymmetry between the right and left hand pe rformance of the experimental group will be compared. If verbal command pantomime performance were more asymmetric (right hand performance greater than left hand perfor mance), this would provide evidence that verbal input interferes with the transfer of praxis moveme nt representations across the corpus callosum in individuals with AD. If pantomime imitation performance were more asymmetric (right hand performance gr eater than left hand performance), this would provide evidence that de ficient transfer of praxis information is specific to the transfer of movement information across the co rpus callosum in individuals with AD. It is predicted that there will be evidence of an interhemispheric callosal disconnection in individuals with AD that is specific to the tr ansfer of information from praxis movement representations.

PAGE 41

29 Table 1-1: Diagnostic criteria for AD. DSM-IV: dementia Alzheimer type Development of multiple cognitive deficits: Memory impairment At least one of the following: Aphasia Apraxia Agnosia Disturbed executive functioning (pla nning, organizing, sequen cing, abstracting) Course characterized by continued gra dual cognitive and functional decline Deficits sufficient to interfere significantly in social and occu pational functioning and representing a decline from past functioning Other causes of dementia excluded (m edical, neurologic, psychiatric) NINCDS-ADRDA: probable Alzheimer disease Dementia established by examinati on and documented by objective testing Deficits in two or more cognitive areas Progressive worsening of memory and other cognitive functions No disturbance in consciousness Onset between 40 and 90 years of age Absence of systemic disorders or other brain disease that could account for the progressive deficits in memory and cognition Diagnosis supported by: Progressive deficits in language (aphasia), perception (agnosia), and motor skills (apraxia) Impaired activities of daily living and altered patterns of behavior Family history of similar disorders Consistent lab results Morris, J.C. (1999). Clinical presentation a nd course of Alzheimer disease. In R.D. Terry, R. Katzman, K.L. Bick, & S.S. Sisodia (Eds.), Alzheimer disease (2nd ed) (pp. 11-24). Philadelphia: Lippincott, Williams, & Wilkins.

PAGE 42

30 Figure 1-1: Cognitive neuropsychological model of limb apraxia. Rothi, L.J.G., Ochipa, C., & Heilman, K.M. (1997a). A cognitive neuropsychological model of limb praxis and ap raxia. In L.J.G. Rothi & K.M. Heilman (Eds.), Apraxia: the neuropsychology of action (pp.29-49). East Sussex, UK: Psychology Press.

PAGE 43

31 CHAPTER 2 METHODS The goals of this study are to determine wh ether praxis information is transferred from left hemisphere movement representati ons to right hemisphere motor areas via the corpus callosum and to examine what types of praxis information are transferred using this neural pathway. This study proposed to use Alzheimers disease as a model for investigating the interhemispheri c transfer of praxis information. The following sections describe the methods for answeri ng the proposed research questions. Subjects Two groups of participants were recruited for particip ation in this study. A group of healthy elderly control subjects (HC) a nd a group of individuals with Alzheimers disease (AD) participated in the study. Inclusion Criteria Inclusion criteria consisted of: 1) for the AD group, a medical diagnosis of AD with no history of other neurologic di sease (i.e. stroke, tumors, TBI, seizures, etc.) and for the HC group, no history of neurologic disease, 2) no history of upper extremity mobility problems, severe hearing loss or severe vi sual impairment, 3) no history of drug or alcohol abuse by self-report, ca regiver report and/or medical r ecord (exclude participants who have experienced alcohol or drug abus e related disease or social or vocational interference as a result of alcohol or drug us e), 4) no history of psychiatric problems by self-report, caregiver report and/or medical r ecords (exclude particip ants who have been hospitalized for psychiatric i llness) 5) because the experi mental stimuli involve object

PAGE 44

32 recognition to perform pantomimes, absence of visual object agnosia as measured using the Associative Match subtest of the Birm ingham Object Recognition Battery (Riddoch & Humphreys, 1993) (i.e., for the AD group, at least a score of 21/ 30 or 70% accuracy and for the HC group, at least a score of 27/30 or 90% accuracy), 6) because the experimental stimuli require processing of verbal commands to perform pantomimes, absence of severe auditory comprehension deficits, as measured using the Sequential Commands subtest of the Western Aphasi a Battery (Kertesz, 1982) (i.e., for the AD group, at least a score of 40/80 or 50% accuracy and for the HC group at least a score of 72/80 or 90% accuracy), 7) English as native language per self-report or caregiver report, 8) right handed (determined by the Waterloo Handedness Questionnaire). The participants with AD were required to provide documentation of a medical diagnosis of AD. All participants with AD also met the DSM-IV / NIDCD/ADRDA criteria for probable AD. Scores from the Mini Mental State Ex am (MMSE) (Folstein, Folstein, & McHugh, 1975; Tombaugh & McInty re, 1992) were used to verify the presence of dementia in the AD group and to group the participants with AD by severity level (a score of less than 27/30 was consider ed impaired). A shortened version of the Boston Naming Test (Fastenau, Denburg, & Mauer, 1998; Kaplan, Goodglass, & Weintraub, 1983) was administer ed to verify cognitive deficits in the AD group and to verify normal naming function in the HC group. Short form 3 from Fastenau et al. (1998) was used in this study. The correlation of this form with the origin al version of the BNT was r = 0.69 which was significant at the p < 0.005 level. In the Fastenau et al. (1998) study, the total sample scored a mean of 13.6 (SD=1.3) on this version. When the performance of the total samp le of healthy older adults was broken down by age, the

PAGE 45

33 following results were reported: 57-68 y ears (n=35) the mean was 14.3 (SD=0.8), 69-76 years (n=38) the mean was 13.3 (SD=1.3), and 77-85 years (n=35) the mean was 13.3 (SD=1.5). As this study has received Inst itutional Review Board approval (IRB # 16602), each participant signed an Informed Consent Form. Subject Demographics Twenty-two (see sample size estimati on below) individuals who had been diagnosed with AD, were recruited for this st udy from the University of Florida Memory Disorder Clinic (UFMDC) and the surroundi ng community. The individuals with AD who were recruited from the UFMDC particip ated in a neuropsychological assessment, a neurologic exam, and a physical exam pr ior to being enrolled in the study. The participants who were recr uited from the community were required to provide documentation of a medical diagnosis of AD from a physician. Only patients who met the DSM-IV/NINCDS-ADRDA criteria for a di agnosis of probable AD were enrolled in the study as experimental subjects. The AD group consisted of 14 women and 8 men with an age range of 61-90 years, mean age of 79.23 years (SD = 6.4 years), and a mean education level of 13.59 years (SD = 2.6 years). Of the individuals with AD en rolled in the study, 17 completed all three experimental tasks and 5 completed 2 out of 3 tasks due to the inability to comprehend the instructions for task 3 (Conceptual Pantomime). In addition to the AD group, a group of 24 healthy elderly control subjects (HC) was recruited to serve as a comparison group for the performance of the AD participants on the experimental tasks described below. The HC group was matched with the AD group for age and gender (see results, Chapter 3) The experimenter attempted to match the HC and AD groups for education level, however this was not accomplished (see

PAGE 46

34 results, Chapter 3). The HC group consisted of 15 women and 9 men with an age range of 63-85 years, mean age of 76.10 years (SD = 6.8 years), and mean education level of 15.52 years (SD = 2.4 years). Sample Size Estimation Data from normal controls and individuals with AD for performance on the Florida Action Recall Test (FLART) (Schwartz et al ., 2000) were used to estimate the group means ( ) for right hand and left hand performan ce of individuals with AD. These data were chosen because the FLART stimuli were us ed for all of the apraxia measures in this study and because there are no published data fo r group means to use as estimates for the performance of individuals with AD on the pr oposed measures. The performance of the normal controls on the FLART in the Schwar tz et al. (2000) study was used as an estimate for right hand performance of the i ndividuals with AD in this study. The performance of the experimental group on th e FLART in the Schwartz et al. (2000) study was used as an estimate for left hand perf ormance of the individuals with AD in this study. Because, left hand performance was expected to be significantly more apraxic than right hand performance, the contro l group was used to estimate right hand performance of the AD group. UCLA Departme nt of Statistics Power Calculator (http://calculators.stat.ucla .edu/powercalc/) was used to perform the following sample size estimates. Sample Size Estimation-2 Samples Equal Variances (most conservative estimate) The mean for left hand performance was es timated to be 56.9% and the mean for right hand performance was estimated to be 86.9% (Schwartz et al ., 2000). Standard deviation for both hands was estimated to be 25%, which is the most conservative estimate of variance (Marks, 1999). A two sided hypothesis was proposed, Ho: A = N

PAGE 47

35 and Ha: A N. Significance level ( ) was set at 0.05 and Power was set at 0.80. With the aforementioned parameter estimates, sample size was estimated to be n = 14 for left hand performance and n = 14 for right hand performance, resulting in a total n=28 for the AD group sample size estimate. Sample Size Estimation-2 Samples Unequal Variances (least conservative estimate) The mean for left hand performance was es timated to be 56.9% and the mean for right hand performance was estimated to be 86.9% (Schwartz et al ., 2000). Standard deviation for left hand performance was estimat ed to be 17.8% and standard deviation for right hand performance was estimated to be 7.6% (Schwartz et al., 2000), which is the least conservative estimate of variance (Marks, 1999). A two sided hypothesis was proposed, Ho: A = N and Ha: A N. Significance level ( ) was set at 0.05 and Power was set at 0.80. With the aforementioned parameter estimates, sample size was estimated to be n = 7 for left hand performance and n = 3 for right hand performance, resulting in a total n=10 for th e AD group sample size estimate. It was decided that a reasonable sa mple size for the AD group would be approximately 19 subjects as this is a compromise between the most conservative estimate and the least conservative estimate. However, a total of 22 participants were enrolled in the study. Experimental Tasks Data Collection Procedures All subjects with AD produced each stimul us item of the following tasks with the right hand and left hand. Right hand and left hand performance within a task was randomized across subjects in both groups. To balance for order effects tasks were presented in random order and task order was counterbalanced for all of the subjects.

PAGE 48

36 Each session was videotaped and analyzed o ffline for correct or incorrect performance and error types as described by Rothi et al (1997b). The conceptual apraxia measure (task 3) was scored according to the criteria set forth by Schwartz et al. (2000) (i.e. each item will be scored based on the concept conveyed by each pantomime regardless of the quality of the movement itself). All of the apraxia task s consisted of 45 items. These were the stimulus items that are included in the stimuli from the Florida Action Recall Test (FLART) (see below, conceptual pantomime task). These stimuli we re chosen in order to provide consistency across tasks for statistical comparison. For ex ample, if the FLART shows a picture of a lock on a door knob and the participant is re quired to pantomime key, the patient will also be required to imitate the pantomime for key produced by the examiner (pantomime imitation) and to pantomime to verbal command Show me how you hold and use a key to open a door (verbal comma nd pantomime). Within each task, each stimulus item was performed once with the right hand and once with the left hand in random order. Task 1: Verbal Command Pantomime (VC) The verbal command task was used to investigate the role of production information in praxis processi ng and the contribution of verbal input to the transfer of praxis information across the corpus callosum. Task 1 Procedures. The examiner provided the part icipant with the following instructions: I am going to ask you to pretend to use different tools. I want you to show me how you would use each t ool if you were actually holdi ng the tool in your hand and using it. I am going to ask you to use either your left hand or your ri ght hand. Listen for this cue and use only the hand I ask you to use. The examiner presented the subject with

PAGE 49

37 a verbal command for each of the stimuli and the subject performed pantomimes to verbal command with each hand. See Appendix A for verbal command stimuli. Task 1 Scoring. Two independent raters were tr ained (see Rater Training, next section) to score each of the experimental tasks. Step 1 of the scoring process was to judge the accuracy of each individual response according to the target stimulus and experimental task being scor ed. A correct production rece ived a score of 1 while an incorrect production received a score of 0. For task 1 (VC), each production was scored as correct (1) or inco rrect (0) according to the semantic content of the production and the spatial and temporal aspects of the movement. For example, if the target stimulus was scissors, the participant was required to produce a pantomime for scissors that correctly represented the semantic content and spatiote moral specifications of the movement for using scissors. A production wa s considered correct if it did not contain any error types. Step 2 of the scoring process was to de termine whether an incorrect response was recognizable for the target st imulus. If the production was deemed unrecognizable for the target stimulus, no furthe r categorization of error types was conducted. If the production was deemed recognizable for the targ et stimulus but contained praxis errors, each error was categorized into one or more of the error types described in Appendix B. Task 2: Pantomime Imitation (PI) The pantomime imitation task was used to investigate the role of production information in praxis processi ng and the contribution of verbal input to the transfer of praxis information across the corpus callosum. Task 2 Procedures. The examiner provided the part icipant with the following instructions: I am going to make a movement with my hand and you are going to try to copy my movement. I want you to watch me and wait until my movement is completely

PAGE 50

38 finished before you move your hand. I am going to ask you to use either your right hand or your left hand. Listen for this cues a nd only use the hand I ask you to use. The examiner pantomimed each of the stimuli in random order and with each hand for the subject to imitate and the subject imitated the gestures produced by the clinician with each hand. See Appendix A for pantomime imitation stimuli. Task 2 Scoring. Two independent raters were tr ained (see Rater Training, next section) to score each of the experimental tasks. Step 1 of the scoring process was to judge the accuracy of each individual response according to the target stimulus and experimental task being scor ed. A correct production rece ived a score of 1 while an incorrect production received a score of 0. Fo r task 2 (PI), each pr oduction was scored as correct (1) or incorrect (0) according to th e semantic content of the production and the spatial and temporal aspects of the movement. For example, if the target stimulus was scissors, the participant was required to im itate exactly both the semantic content and spatiotemporal aspects of the movement for sc issors that was produced by the examiner. A production was considered correct if it did not contain any error type s. Step 2 of the scoring process was to determine whether an incorrect response was recognizable for the target stimulus. If the production was deemed unrecognizable for th e target stimulus, no further categorization of error types wa s conducted. If the production was deemed recognizable for the target stimulus but contained praxis errors, each error was categorized into one or more of the error types described in Appendix B. Task 3: Conceptual Pantomime (CP) The verbal command task was used to inve stigate the transfer of action semantics information across the corpus callosum.

PAGE 51

39 Task 3 Procedures. The Florida Action Recall Test (FLART) consists of 45 black and white line drawings of objects placed in scenes implying an action. The subject is instructed to imagine what tool is need ed to act upon each object or scene and to pantomime the action associated with that tool in relation to the dr awing. For example, a drawing of an unshaven face requires a shav ing action and a drawing of a cooked turkey requires a carving action. The targeted tool is not shown in the drawing. For this study, conceptual praxis was tested using the stimu li from the FLART (for examples of stimuli see Figure 2-1). The examiner provided the participant with the following instru ctions: I am going to show you some drawings of objects in scen es that imply an action. You must imagine what tool is needed to act upon object in th e picture. Then pretend to do the action associated with the tool that would be used to act on the object shown. A tool is any item that can be held in one hand and can be us ed to act on a pictured object. Tools may include personal care items, kitchen utensi ls, household items, garage tools, sports equipment, or musical instruments. The tool is not shown in the drawing. I will tell you which hand to use to perform the action. Do not name the tool and do not name the object. Using your hand to complete the ac tion without the assistance of a tool is incorrect. See Appendix A for th e stimuli used in this task. Task 3 Scoring. Two independent raters were tr ained (see Rater Training, next section) to score each of the experimental tasks described ab ove. Step 1 of the scoring process was to judge the accura cy of each individual response according to the target stimulus and experimental task being scored. A correct production r eceived a score of 1 while an incorrect production r eceived a score of 0. For task 3 (CP), each stimulus item

PAGE 52

40 was classified as correct (1) or incorrect (0) based on the semantic content of the production only (i.e. the presence of a spat ial and/or temporal error(s) without the presence of a content error, was not consid ered an incorrect production). For example, for stimulus item #45, the pictured object is a paper doll and the target pantomime is scissors. If the subject pantomimes scissors using their fingers as the blades of the scissors (BPT error), the semantic content of the production is corre ct so the production would be scored as correct. If the subject pantomimes coloring the paper doll with a crayon (R error), the semantic content of the pr oduction is incorrect, so the production would be scored incorrect. A production was considered correct if it did not contain any conceptual errors. Step 2 of the scoring pr ocess was to determine whether an incorrect response was recognizable for the target st imulus. If a response was considered recognizable, it was considered incorrect only if it contai ned content errors (i.e., hand error (H), related erro r (R), nonrelated error (N), or c oncretization error (C)). Step 3of the scoring process involved ca tegorizing each conten t error into one of the content error types described in Appendix B. The presence of temporal, spatial, or other errors (see Appendix B) was also noted. Rater Training Two independent raters were trained to sc ore the responses of each participant for each task. Rater 1, the primary rater, scored all of the data for statistical analysis and scored a percentage of the data again for reliability purposes. Rater 1 was a graduate student in Occupational Therapy at the University of Florid a. Before training she had extensive experience viewing and scoring videotapes of apraxi c research participants but had little experience in the clinical assessment and treatment of limb apraxia. Rater 2, the reliability rater, scored a percentage of the data for reliability purposes. Rater 2 was an

PAGE 53

41 undergraduate student in Speech Pathology at the University of Florida. Before training she was inexperienced in viewi ng and scoring videotapes of ap raxic research participants but had received some instruction in the clinical assessment and treatment of limb apraxia. Rater 1 and Rater 2 received extensiv e training by the experimenter with regards to judging the accuracy of responses and cl assification of error types. They both participated in several sessions (approximate ly 4 hours) of focused instruction on the judgment of correct and incorre ct response and identificati on of error types. These sessions involved viewing several practice ta pes (individuals who were not included in the study producing pantomimes) and attempting to judge accuracy and identify errors with discussion following each production be tween the raters and the experimenter. Following these sessions, the two raters wa tched several videotapes of pantomime productions. They were requi red to score each production independently then discuss their scoring until they were able to reach 90% agreement on10 consecutive productions. For reliability, the two raters judged th e praxis productions of each participant independently and were not permitted to confer regarding their judgements. Reliability Two independent raters (Rat er 1 and Rater 2) analy zed 20% of the data to determine interand intrarater reliability. Intra-rater reliability: In order to determine whether scoring of the apraxia tasks was reli able when scored multiple times by the same judge, 20% of the data were re-scored by Rate r 1 who scored the entire data sample for statistical analysis. Inter-rater reliability: In order to determine whether scoring of the apraxia tasks was reliable when scored by i ndependent judges, 20% of the data were rescored by Rater 2 who scored only a small samp le of the data for reliability purposes. Reliability scoring was completed from a videotap e of the original test administration.

PAGE 54

42 In order to describe the reliability of the two raters, percent reliability was calculated as the number of agreements minus the number of disagreements divided by the total number of stimuli multiplied by 100 (# of agreements # of disagreements / total number of stimuli) x 100 = % correct. Although percent agreement reflects the proportion of agreements among the total numbe r of judgments, it does not take into account the amount of agreement expected by chance (Kramer & Feinstein, 1981). Therefore, statistical analysis of the reliability data was completed using the k (kappa) statistic because this is considered the index of choice for measurement of observer agreement and corrects for agreement expect ed by chance (Kramer & Feinstein, 1981). Kappa is ordinarily used to measure the concordance between two observers. According to Kramer and Feinstein (1981) the magnitude or value of kappa is more descriptive than the associated p value and they state that p<.05 is a necessary but not sufficient criterion for meaningful obser ver agreement. Therefore, the following guidelines were suggested (see table 2-4) for the strength of observer agreement. Statistical Analysis Before statistical analysis was completed, the data were collapsed within subject. For the current study, statistical analysis was performed on the response accuracy variable only. Descriptive data for each error type is provided but st atistical analysis of this data will be reserved for future studies. For the dependent variable response accura cy, a percentage was calculated for each participant. For the error types a percentage was calculated and error total represents the total number of errors. The percentages and averages for the dependent variables were calculated as follows: percent response accur acy = number of correct responses / total number of stimuli x 100; percentage of each e rror type = number of errors present / total

PAGE 55

43 number of errors x 100. In addition, an asymmetry ratio was calculated for response accuracy for each subject in both groups. Th e asymmetry ratio was calculated as right hand performance minus left hand performa nce divided by right hand performance plus left hand performance multiplied by 100. The proba bility level for significance for all statistical analyses was set a p < 0.05 Research Question 1 Do individuals with AD have conceptual and/ or ideomotor apraxia in the left hand? Analysis. Separate nonparametric Mann-Whitney U tests were used to compare left hand performance of the AD and HC gr oups on the verbal command pantomime (task 1) and the conceptual pantomime (task 3) task s. The test variable was response accuracy and the grouping variable was group (AD and HC). Research Question 2 What is the contribution of degraded praxis movement and conceptual representations (due to corti cal atrophy) to the limb apra xia in individuals with AD? Analysis. Separate nonparametric Mann-Whitney U tests were used to compare right hand performance of the AD and HC groups on the verbal command pantomime (task 1) and conceptual pantomime (task 3) tasks. The test variable was response accuracy and the grouping variable was group (AD and HC). Research Question 3 What is the contribution of interhemisphe ric disconnection (due to callosal atrophy) to the limb apraxia in individuals with AD? Analysis. Separate nonparametric Mann-Whitney U tests were used to compare the asymmetry ratios of the AD and HC gr oups on the verbal command pantomime (task 1), pantomime imitation (task 2), and conceptu al pantomime (task 3) tasks. The test

PAGE 56

44 variable for each analysis was response accuracy and the grouping variable was group (AD and HC) Research Question 4 Is the disruption of praxis information transfer a result of an intrahemispheric verbal-motor disconnection or an interhem ispheric corpus callosum disconnection? Analysis. A 2x2 ANOVA procedure was used to compare right hand performance of the AD and HC groups on the verbal command pantomime (task 1) and pantomime imitation (task 2) tasks. For the 2x2 ANOVA, fact or one was task with two levels (verbal command pantomime and pantomime imitation) and factor two was group with two levels (AD group and HC group). A MannWhitney U test was used to compare asymmetry ratios of the AD group for the verbal command pantomime (task 1) and pantomime imitation (task 2) tasks. For this test, the test variable was response accuracy and the grouping variable was group (AD and HC).

PAGE 57

45 Table 2-1: Individual subject demographics for AD group. subject # gender age education (# of years) 01-001 F 85 10 01-003 F 78 14 01-004 F 61 12 01-005 F 80 18 01-006 F 82 15 01-007 F 73 12 01-008 M 80 10 01-009 M 77 12 01-010 M 80 15 01-011 M 90 20 01-012 M 80 13 01-013 F 76 12 01-014 M 85 12 01-015 M 88 14 01-016 F 81 12 01-017 M 83 18 01-018 F 71 16 01-019 F 80 12 01-020 F 85 12 01-021 F 81 13 01-022 F 76 15 01-024 F 71 12 mean 79.23 13.59 SD 6.4 2.6 F = female, M = male, SD = standard deviation

PAGE 58

46 Table 2-2: Individual subject demographics for HC group. subject # gender age education (# of years) 02-001 F 84 15 02-002 F 78 16 02-003 F 65 18 02-004 F 73 16 02-005 F 67 12 02-006 F 63 19 02-007 F 68 12 02-009 F 76 16 02-010 M 76 18 02-011 M 70 18 02-012 M 81 12 02-013 M 80 18 02-014 F 83 12 02-017 M 85 18 02-018 M 83 16 02-019 F 82 14 02-020 F 79 14 02-021 M 76 12 02-022 F 74 16 02-023 F 70 16 02-025 M 85 18 02-026 M 77 12 02-028 F 79 12 02-029 F 85 12 mean 76.1015.52 SD 6.82.4 F = female, M= male, SD = standard deviation

PAGE 59

47 Table 2-3: Strength of observer agreemen t for ranges of kappa statistic values. Value of k Strength of agreement < 0 Poor 0 .20 Slight .21 .40 Fair .41 .60 Moderate .61 .80 Substantial .81 1.00 almost perfect Kramer, M.S. & Feinstein, A.R. (1981). Clin ical biostatistics: the biostatistics of concordance. Clinical pharmacology and therapeutics, 29, 111-117.

PAGE 60

48 Figure 2-1: Examples of pictures used in the Florida Action Recall Test (FLART) Target gesture (tool): A. car ving (knife), B. chopping (hatchet), C. sharpening (pencil sharpener), D. spreading (knife), E. openi ng (bottle opener), F. painting (paint brush). Schwartz, R.L., Adair, J.C., Raymer, A. M., Williamson, D.J.G., Crosson, B., Rothi, L.J.G., Nadeau, S.E., & Heilman, K.M. ( 2000). Conceptual apraxia in probable Alzheimers disease as demonstrated by the Florida Action Recall Test. Journal of the International Neur opsychological Society, 6, 265-270.

PAGE 61

49 CHAPTER 3 RESULTS This study examined whether praxis info rmation crosses the corpus callosum to inform right hemisphere motor pathways by comparing right hand and left hand performance on three praxis tasks in indivi duals with AD and hea lthy elderly control subjects. Descriptive statistics and statistical analyses are presented in an attempt to answer the proposed research questions. Subject Demographics As stated in the methods section, an a ttempt was made to match the HC and AD groups for age and education level. Mann-Whitn ey U tests were used to compare the two groups for age and education level. There was not a significant difference between the AD and HC groups for age (U = 204.000, p = 0.186) but there was a significant difference between the two groups for educ ation level (U = 179.500, p = 0.055). Reasons for this difference in education level will be addressed later (see discussion, Chapter 4). Neuropsychological Screening As described in the previous chapter, each participant was evaluated using several cognitive screening measures prior to particip ating in the experimental protocol described above. The Mini-Mental State Exam (MMSE) the Associative Ma tch subtest of the Birmingham Object Recognition Battery (BO RB), the Sequential Commands subtest of the Western Aphasia Battery (WAB), and a 15-item short form of the Boston Naming Test (BNT) were administered to each participant. The purpose of this neuropsychological screening was threefol d. 1) The performance of the AD group on

PAGE 62

50 these measures was used to support the medical diagnosis and to verify the presence of memory and cognitive deficits. Therefore, in order to be included in this study, all of the subjects in the AD group were required to score below a certain level in order to be considered impaired on a particular measure (s ee Chapter 2 for cut-off scores). 2) The performance of the HC group on these measures was used to determine the current level of cognitive functioning for each participant and to verify that each participant was performing at age appropriate levels for meas ures of cognition and memory. Therefore, in order to be included in this study, all of the subjects in the HC group were required to score above a certain level in order to be considered within normal limits for a particular cognitive domain (see Chapter 2 for cut-o ff scores). 3) For the AD group, it was hoped that performance on these neuropsychological m easures could be used to subdivide the group for further data analysis. However, due to the small sample size this type of posthoc analysis was not feasible. Followi ng are the results of the neuropsychological screening. For the MMSE, a score of 27 or higher (out of 30) was considered within normal limits while a score of 25 or lower (out of 30) was considered consistent with dementia (Lezak, 1995). The subjects in the HC group scored a range of 27-30 (mean=28.25, SD=1.1) while the AD group scored a range of 12-25 (mean=20.82, SD=3.9). In the AD group, 0 participants had severe (MMSE < 10) 7 participants had moderate (MMSE > 10 but <20) and 15 participants had mild (MMSE > 20 but <25) dementia. On the Sequential Commands subtest of th e WAB, inclusion into the study required a score of at least 40/80 or 50% accuracy and for the AD group and a score of at least 72/80 or 90% accuracy for the HC group On this measure, the score range for the HC

PAGE 63

51 group was 75-80 (mean=79.58, SD=1.3) and the score range for the AD group was 39-80 (mean=73.91, SD=9.5). One of the participants in the AD group scored below the 40/80 cut-off for inclusion (subject #01-004 scored 39/80 on this measure). However, due to difficulty recruiting subjects for the st udy, the data was nevertheless included. On the Associative Match s ubtest of the BORB, inclusi on into the study required a score of at least 21/30 or 70% accuracy for the AD group and a score of at least 27/30 or 90% accuracy for the HC group. On this measure, the score range for the HC group was 28-30 (mean=29.75, SD=0.5) and the score range for the AD group was 19-30 (mean=27.41, SD=2.6). One of the participants in the AD group scored below the 21/30 cut-off for inclusion (subject #01-004 scored 19/30 on this measure). However, due to difficulty recruiting subjects for the study, the data was nevertheless included. A 15-item version of the BNT (Fastenau et al ., 1998) was also administered to each participant. A score of 12 or below was consid ered impaired for this measure. Subjects in the HC group scored a range of 13-15 it ems correct (mean=13.96, SD=0.8). Subjects in the AD group scored a range of 1-14 ite ms correct (mean=8.77, SD=3.1). Although subject #01-009 scored within normal limits on the 15-item BNT, he was included in the study because he had been previously di agnosed with AD by a neurologist and his MMSE score (23/30) was below that considered to be consistent with dementia. See Table 3-1 and 3-2 for scores on these screeni ng measures for each individual subject. Reliability Because of the subjective nature of the scor ing method used for this study (Rothi et al., 1988), it was important establish the reliab ility of the accuracy judgments and error categorization made by the primary rater (R ater 1). This was accomplished by requiring Rater 1 to score 20% of the data on two sepa rate occasions (intra-rater reliability) and

PAGE 64

52 requiring Rater 2 to score 20% of the data in dependent of the prim ary rater (inter-rater reliability). The following results suggest that high interand intrarater reliability was established thereby lending credibility to the data. For interand intrarater reliability, % agreement was greater than 80% for all response variables and categories with the excepti on of inter-rater reliability of IC in task 1 (VC) (75.9%), IC in task 2 (PI) (77.1%), and IC and M in task 3 (CP) (78.2% and 79.6%, respectively). Overall, in tra-rater reliability was slig htly better than inter-rater reliability in that there were no instances in which the percentage agreement of Rater 1 as compared to Rater 2 was less than 80%. The kappa (k) statistic was significant at the 0.05 level for all response variables, where applicable. Kappa was greater than 0.40 for all reliability comparisons with the exception of inter-rater reliability of C (k=0, poo r) in Task 3 (CP), intra-rater reliability of P (k=0, poor), R (k=0.282, fair), H (k=0.284, fair ), and UR (k=0.402, fair) in task 1 (VC), intra-rater reliability of R (k=0, poor) in ta sk 2 (PI), and intra-ra ter reliability of P (k=0.328, fair), N (k=0.332, fair), and C (k=0, poo r) in task 3 (CP). In the instances in which the % agreement was relatively high (i.e. > 90%) but the value of kappa was relatively low (i.e. < .40) (see bo lded variables in Tables 3-3 and 3-4), k is perhaps not a valid measure of concordance. The reason fo r this disparity between percent agreement and the kappa statistic, is that for the vari ables in question, the two observers did not disagree enough to account for the possibility that they were agreeing by chance. See Tables 3-3 and 3-4 for intera nd intrareliability data for each independent variable in tasks 1, 2, and 3.

PAGE 65

53 Descriptive Statistics Task 1: Verbal Command Pantomime (VC) HC group. For the HC group, mean percent accur acy on this task was 50.0% (SD = 7.2) with the right hand and 47.0% (SD = 7.7) with the left hand with a mean difference between the two hands of 3.0% (SD = 7.2%) and a mean asymmetry ratio of 3.26 (SD = 8.12). In the HC group (see table 3-5), 13/24 (54.2%) participants performed better with the right hand than the left hand (a positive diffe rence) while 8/24 (33.3%) participants performed better with the left hand than the right hand (a negative difference) and 3/24 (12.5%) participants showed no difference between the hands (see table 3-5). AD group. Mean percent accuracy for the AD group on task 1 (VC) was 28.7% (SD = 11.7) with the right hand and 24.7% (S D = 10.7) with the left hand with a mean difference between the two hands of 4.0% (S D = 6.5%) and a mean asymmetry ratio of 7.97 (SD = 13.55). For task 1 (VC) in th e AD group (see table 3-6), 15/22 (68.1%) participants performed better with the right hand than the left hand (a positive difference) while 6/22 (27.3%) participants performed better with the left hand than the right hand (a negative difference) and 1/22 (4.5%) particip ant showed no difference between the hands (see table 3-6). On task 1 (VC), the difference in performance of the HC and AD groups was 21.3% with the right hand and 23.0% with the left hand. In summary, the HC group performed pantomimes more accurately and demonstrated less performance variability than the AD group on this task. Task 2: Pantomime Imitation (PI) HC group. Mean percent accuracy for the HC group, on this task was 49.8% (SD = 11.7) with the right hand and 42.4% (SD = 13.1) with the left hand with a mean

PAGE 66

54 difference between the two hands of 7.3% (S D = 7.2%) and a mean asymmetry ratio of 8.85 (SD = 8.68). In the HC group (see table 3-5), 19/24 (79.2%) participants performed better with the right hand than the left hand (a positiv e difference) while 4/24 (16.7%) participants performed better with the left hand than the right hand (a negative difference) and 1/24 (4.2%) participant showed no differe nce between the hands (see table 3-5). AD group. Data from the AD group for Task 2 showed a mean percent accuracy of 31.1% (SD = 14.1) with the right hand and 20.4% (SD = 9.4) with the left hand with a mean difference between the two hands of 10.7% (SD = 8.7%) and a mean asymmetry ratio of 20.57 (SD = 21.96). For task 2 (PI) in the AD group (see table 3-6), 20/21 (95.2%) participants perf ormed better with the right hand than the left hand (a positive difference) while 1/21 (4.8%) partic ipants performed better with the left hand than the right hand (a negative difference) and 0/ 22 (0%) participant showed no difference between the hands (see table 3-6). On task 2, the difference in performance of the HC and AD groups was 18.7% with the right hand and 22.0% with the left hand. Overall, the HC group performed pantomimes more accurately but with simila r variability in comparison to the AD group on this task Task 3: Conceptual Pantomime (CP) HC group. For the HC group, mean percent accur acy on this task was 86.2% (SD = 7.2) with the right hand and 84.9% (SD = 7.5) with a mean difference between the two hands of 1.3% (SD = 4.6%) and a mean asymmetry ratio of 0.79 (SD = 2.78). In the HC group (see table 3-5), 11/24 (45.8%) part icipants performe d better with the right hand than the left hand (a positive difference) wh ile 7/24 (29.1%) particip ants performed better

PAGE 67

55 with the left hand than the right hand (a negative difference) and 5/24 (20.8%) participants showed no difference be tween the hands (see table 3-5). AD group. Mean percent accuracy for the AD group on task 3 (CP) was 62.8% (SD = 13.9) with the right hand and 61.1% (S D = 13.4) for the left hand with a mean difference between the two hands of 1.7% (S D = 4.5%) and a mean asymmetry ratio of 1.38 (SD = (3.87). For task 3 (CP) in the AD group (see table 3-6), 11/18 (61.1%) participants performed better with the right hand than the left hand (a positive difference) while 6/18 (33.3%) participants performed better with the left hand than the right hand (a negative difference) and 1/18 (5.5%) particip ant showed no difference between the hands (see table 3-6). On Task 3, the difference in performa nce of the HC and AD groups was 23.2% with the right hand and 24.0% with the left hand. In summary, the HC group performed pantomimes more accurately and demonstrated less performance variability than the AD group on this task. Error Types Task 1, 2, and 3 Descriptive data for error types can be f ound in table 3-7 (task 1), 3-8 (task 2), 3-9 (task 3) and 3-10 (error tota ls for tasks 1, 2, and 3). For Tasks 1 and 2, both groups showed a high percentage of sequencing (S), internal configurat ion (IC), external configuration (EC), and movement (M) errors with both hands rela tive to other error types. The AD group also demonstrated a hi gher percentage of body part as tool (BPT) errors than the HC group with both hands on task 1 and a higher percentage of unrecognizable errors (UR) with both hands on task 1 and 2. For task 3 (CP), only content errors are re ported since this is a conceptual task. The HC group showed a higher percentage of related (R) than hand (H) errors while the

PAGE 68

56 AD group showed a higher percentage of hand (H) than related (R) errors on task 3 (CP). The percentage of perseverative (P) and nonrel ated (N) errors on this task was relatively low for both groups with both hands. When the total number of errors for task 1 (VC) and task 2 (PI) was calculated, the HC group produced fewer errors than th e AD group and both groups produced fewer errors with the right hand than the left hand. For task 3 (CP), the AD group made more errors than the HC group but there was little difference between the right hand and the left hand in the total number of errors for both groups. Statistical Analysis Research Question 1 Do individuals with AD have conceptual an d/or ideomotor apraxia in the left hand? Results. To answer this question, Mann-Whitn ey U tests were performed using data from left hand performance of the AD and the HC groups on the verbal command pantomime (task 1) and conceptual pantomime (task 3) tasks. The Mann-Whitney U test for left hand performance on the verbal co mmand pantomime task (ideomotor apraxia) was significant at the p < 0.01 level (U = 25.500, p < 0.001). The Mann-Whitney U test for left hand performance on the conceptual pantomime task (conceptual apraxia) was significant at the p < 0.01 level (U = 17.500, p < 0.001)]. Research Question 2 What is the contribution of degraded praxis movement and conceptual representations (due to corti cal atrophy) to the limb apraxi a in individuals with AD? Results. To answer this question, Mann-Whitn ey U tests were performed using data from right hand performance of the AD and the HC groups on the verbal command pantomime (task 1) and conceptual pantomime (task 2) tasks. The Mann-Whitney U test

PAGE 69

57 for right hand performance on the verbal comm and pantomime task was significant at the p < 0.01 level (U = 50.000, p < 0.001). The Mann-Whitney U test for right hand performance on the conceptual pantom ime task was significant at the p < 0.01 level (U = 25.000, p < 0.001). Research Question 3 What is the contribution of interhemisphe ric disconnection (due to callosal atrophy) to the limb apraxia in individuals with AD? Results. To answer this question, Mann-Whitn ey U tests were performed using asymmetry ratio data from the performan ce of the AD and HC groups on the praxis production and praxis conceptual tasks. Th e Mann-Whitney U test for asymmetry ratios on the verbal command pantomime task (task 1) was signifi cant at the p < 0.05 level (U = 171.000, p = 0.041). The Mann-Whitney U test for asymmetry ratios on the pantomime imitation task (task 2) were significant at the p < 0.01 level (U = 104.500, p = 0.001). The Mann-Whitney U test for asymmetry ratio s on the conceptual pantomime task (task 3) were not significant (U = 181.000, p = 0.373). Research Question 4 Is the disruption of praxis information transfer a result of an intrahemispheric verbal-motor disconnection or an interh emispheric corpus callosum disconnection? Results. To answer this question, a 2x2 ANO VA procedure was used to compare right hand performance of the AD and HC groups on the verbal command pantomime (task 1) and pantomime imitation (task 2) ta sks and a Mann-Whitney U test was used to compare asymmetry ratios of the AD group fo r the verbal command pantomime (task 1) and pantomime imitation (task 2) tasks. The 2x2 ANOVA was significant for the main effect of group [F(1) = 68.498, p < 0.001, effect size = .441] but was not significant for

PAGE 70

58 the main effect of task [F(1) = 0.250, p = 0.619] and there was not a significant task*group interaction [F(1) = 0.237, p = 0.627] The Mann-Whitney U test was also significant (U = 100.000, p = 0.001) Summary With regards to the neuropsychological screening, the AD group demonstrated memory and cognitive deficits consistent with a diagnosis of dementia while the HC group demonstrated normal performance on memory and cognitive tests. Both intra and inter rater reliability were determined to be relatively high lendi ng credibility to the scoring system utilized to analyze the data. With regards to descriptive statistics, overall, the HC group demonstrated greater response accuracy, less performance variabilit y, and fewer errors than the AD group. For both groups with both hands, spatial and tempor al errors were the most common types of errors produced during task 1 (VC) and task 2 (PI) while content errors were the most common type of error in task 3 (CP) fo r both groups with both hands. The AD group also produced more unrecognizable responses than the HC group. The statistical analyses that were con ducted in order to answer the research questions showed that indi viduals with AD demonstrated ideomotor and conceptual apraxia in both the right and left hands. Additionally, the results suggested that callosal degeneration in individuals w ith AD interrupts the interhemispheric transfer of praxis production information but not praxis con ceptual information. Finally, it can be concluded that the interrupti on of interhemispheric transfer of praxis information in individuals with AD is specific to the transfer of motor information from left hemisphere praxis movement representations to right he misphere motor areas. A discussion of the clinical and empirical implications of these results follows.

PAGE 71

59 Table 3-1: Scores for screening measur es for individual subjects in HC group. subject # MMSE WAB BORB BNT 02-001 27 80 30 13 02-002 30 79 30 15 02-003 30 80 30 15 02-004 29 80 30 14 02-005 28 80 30 14 02-006 30 80 30 14 02-007 27 80 30 15 02-009 28 80 29 15 02-010 28 75 29 15 02-011 27 80 29 14 02-012 28 80 30 15 02-013 27 80 30 13 02-014 30 80 30 14 02-017 28 80 30 13 02-018 30 80 30 13 02-019 27 80 30 14 02-020 28 80 28 13 02-021 27 80 30 14 02-022 29 76 30 14 02-023 29 80 30 15 02-025 27 80 30 13 02-026 28 80 30 14 02-028 28 80 30 13 02-029 28 80 29 13 Mean 28.25 79.58 29.75 13.96 SD 1.1 1.3 0.5 0.8 MMSE = Mini Mental State Exam, WAB = Western Aphasia Battery, Sequential Commands Subtest, BORB = Birmingham Object Recognition Battery, Semantic Matching Subtest, BNT = 15-item short form of Boston Naming Test SD = standard deviation

PAGE 72

60 Table 3-2: Scores for screening measur es for individual subjects in AD group. subject # MMSE WAB BORB BNT 01-001 20 76 29 13 01-003 19 80 29 11 01-004 16 39 19 6 01-005 23 80 29 9 01-006 23 80 27 6 01-007 23 80 25 11 01-008 24 80 29 9 01-009 23 78 27 14 01-010 12 71 30 10 01-011 25 80 29 10 01-012 24 68 27 11 01-013 18 80 30 9 01-014 16 62 24 1 01-015 14 67 27 4 01-016 22 72 24 8 01-017 16 70 26 6 01-018 23 80 29 5 01-019 23 80 29 9 01-020 21 80 28 11 01-021 24 70 29 9 01-022 24 75 29 9 01-024 25 78 28 12 mean 20.82 73.91 27.41 8.77 SD 3.9 9.5 2.6 3.1 MMSE = Mini Mental State Exam, WAB = Western Aphasia Battery, Sequential Commands Subtest, BORB = Birmingham Object Recognition Battery, Semantic Matching Subtest, BNT = 15-item short form of Boston Naming Test SD = standard deviation

PAGE 73

61 Table 3-3: Inter-rater reliability using % agre ement and the Kappa statistic for task 1, 2, and 3 Task 1-VC Task 2-PI Task 3-CP response variable % k k strength % K k strength % k k strength accuracy 92.1 0.829 almost perfect92.7 0.833almost perfect96.8 0.912 almost perfect P 100.0 1.000 almost perfect100.01.000almost perfect99.9 0.767 substantial R 99.9 0.856 almost perfect100.01.000almost perfect99.6 0.943 almost perfect N 100.0 1.000 almost perfect100.01.000almost perfect99.9 0.908 almost perfect H 99.7 0.856 almost perfect100.01.000almost perfect98.2 0.906 almost perfect content 99.9 N/A 100.0N/A 99.3 N/A S 99.7 0.821 almost perfect95.5 0.815almost perfect97.1 0.781 substantial T 96.8 0.690 substantial 97.0 0.603moderate 96.3 0.575 moderate O 96.6 0.795 substantial 98.4 0.585moderate 98.9 0.812 almost perfect temporal 97.1 N/A 97.0 N/A 97.4 N/A A 98.2 0.688 substantial 97.6 0.626substantial 98.6 0.699 substantial IC 84.2 0.708 substantial 86.6 0.756substantial 86.8 0.799 substantial EC 87.9 0.565 moderate 91.1 0.736substantial 94.0 0.656 substantial BPT 97.4 0.538 moderate 99.0 0.627substantial 98.9 0.846 almost perfect M 88.2 0.732 substantial 91.2 0.813almost perfect89.1 0.696 substantial spatial 91.5 N/A 93.3 N/A 93.7 N/A C 100.0 1.000 almost perfect100.01.000almost perfect99.9 0.000 poor NR 99.9 0.799 substantial 99.9 0.888almost perfect99.7 0.908 almost perfect UR 99.6 0.867 almost perfect99.6 0.912almost perfect99.3 0.926 almost perfect other 99.8 N/A 99.8 N/A 99.6 N/A % agreement = (# of agreements # of disagreements)/total # of stimuli x 100 VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime % = percent of agreemen t, k = kappa statistic P = Perseverative error, R = Related error, N = Non-related error, H = Hand error, S = Spatial error, T = Timing error, O = Occurrence error, A = Amplitude error, IC = Internal Configuration error, EC = External Configuration error, BPT = Body-part-as-tool error, M = Move ment error, C = Concretisation error, NR = No Response, UR = Unrecognizable response accuracy = percentage of correct responses, content = sum of P, R, N, and H errors, temporal = sum of S, T, and O errors, spat ial = sum of A, IC, EC BPT, and M errors, other = sum of C, NR, and UR errors

PAGE 74

62 Table 3-4: Intra-rater reliability using % agre ement and the Kappa statistic for task 1, 2, and 3. Task 1VCP Task 2PI Task 3CP response variable % k k strength % k k strength % k k strength accuracy 82.7 0.661 substantial 85.4 0.685substantial 92.7 0.810 almost perfect P 99.9 0.000 poor 100.01.000almost perfect98.9 0.328 fair R 99.3 0.282 fair 99.9 0.000poor 97.7 0.713 substantial N 100.0 1.000 almost perfect100.01.000almost perfect99.4 0.332 fair H 99.3 0.284 fair 100.01.000almost perfect95.0 0.769 substantial Content 99.6 N/A 100.0N/A 97.8 N/A S 94.0 0.704 substantial 93.3 0.733substantial 95.3 0.662 substantial T 95.8 0.653 substantial 96.1 0.523moderate 94.6 0.539 moderate O 97.6 0.755 substantial 98.1 0.543moderate 97.7 0.608 substantial temporal 95.8 N/A 95.9 N/A 95.9 N/A A 96.8 0.461 moderate 97.3 0.445moderate 97.9 0.535 moderate IC 75.9 0.580 moderate 77.1 0.621substantial 78.2 0.553 moderate EC 85.0 0.484 moderate 82.5 0.564moderate 88.4 0.452 moderate BPT 98.0 0.622 substantial 99.3 0.702substantial 97.7 0.626 substantial M 80.0 0.593 moderate 84.2 0.690substantial 79.6 0.498 moderate Spatial 87.9 N/A 88.8 N/A 89.0 N/A C 100.0 1.000 almost perfect100.01.000almost perfect99.9 0.000 poor NR 99.7 0.749 substantial 99.7 0.832almost perfect99.2 0.696 substantial UR 97.6 0.402 fair 97.7 0.489moderate 96.0 0.630 substantial Other 99.1 N/A 99.2 N/A 98.4 N/A % agreement = (# of agreements # of disagreements)/total # of stimuli x 100 VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime % = percent of agreemen t, k = kappa statistic P = Perseverative error, R = Related error, N = Non-related error, H = Hand error, S = Spatial error, T = Timing error, O = Occurrence error, A = Amplitude error, IC = Internal Configuration error, EC = External Configuration error, BPT = Body-part-as-tool error, M = Move ment error, C = Concretisation error, NR = No Response, UR = Unrecognizable response accuracy = percentage of correct responses, content = sum of P, R, N, and H errors, temporal = sum of S, T, and O errors, spat ial = sum of A, IC, EC BPT, and M errors, other = sum of C, NR, and UR errors

PAGE 75

63 Table 3-5: Response accuracy (percent) data with difference scores and asymmetry ratios for individual subjects in HC group for Tasks 1, 2, and 3. Task 1-VCP Task 2-PI Task 3-CP subj# % acc RH % acc LH diff ratio % acc RH % acc LH diff ratio % acc RH % acc LH diff ratio 02-001 51.1 46.7 4.4 4.50 44.4 28.9 15.521.15 93.3 86.7 6.6 3.67 02-002 44.4 46.7 -2.3 -2.52 40.0 40.0 0.0 0.00 91.1 91.1 0.0 0.00 02-003 51.1 44.4 6.7 7.02 51.1 40.0 11.112.18 88.9 84.4 4.5 2.60 02-004 42.2 44.4 -2.2 -2.54 33.3 26.7 6.6 11.00 80.0 82.2 -2.2 -1.36 02-005 62.2 48.9 13.3 11.97 53.3 44.4 8.9 9.11 88.9 88.9 0.0 0.00 02-006 46.7 46.7 0.0 0.00 46.7 53.3 -6.6 -6.60 93.3 91.1 2.2 1.19 02-007 44.4 35.6 8.8 11.00 33.3 26.7 6.6 11.00 91.1 93.3 -2.2 -1.19 02-009 55.6 55.6 0.0 0.00 44.4 37.8 6.6 8.03 93.3 93.3 0.0 0.00 02-010 53.3 44.4 8.9 9.11 60.0 64.4 -4.4 -3.54 86.7 84.4 2.3 1.34 02-011 44.4 48.9 -4.5 -4.82 42.2 31.1 11.115.14 80.0 73.3 6.7 4.37 02-012 48.9 51.1 -2.2 -2.20 57.8 35.6 22.223.77 86.7 88.9 -2.2 -1.25 02-013 48.9 48.9 0.0 0.00 68.9 55.6 13.310.68 75.6 66.7 8.9 6.25 02-014 46.7 48.9 -2.2 -2.30 51.1 37.8 13.314.96 84.4 88.9 -4.5 -2.60 02-017 46.7 44.4 2.3 2.52 46.7 37.8 8.9 10.53 88.9 91.1 -2.2 -1.22 02-018 46.7 51.1 -4.4 -4.50 64.4 55.6 8.8 7.33 77.8 82.2 -4.4 -2.75 02-019 44.4 26.7 17.7 24.89 31.1 33.3 -2.2 -3.42 84.4 91.1 -6.7 -3.82 02-020 48.9 35.6 13.3 15.74 35.6 20.0 15.628.06 71.1 68.9 2.2 1.57 02-021 51.1 42.2 8.9 9.54 51.1 46.7 4.4 4.50 88.9 84.4 4.5 2.60 02-022 55.6 51.1 4.5 4.22 46.7 44.4 2.3 2.52 75.6 80.0 -4.4 -2.83 02-023 66.7 55.6 11.1 9.08 67.4 57.8 9.6 7.67 97.9 93.3 4.6 2.41 02-025 66.7 64.4 2.3 1.75 68.9 73.3 -4.4 -3.09 88.9 88.9 0.0 0.00 02-026 44.4 57.8 -13.4 -13.11 46.7 42.2 4.5 5.06 95.6 84.4 11.2 6.22 02-028 51.1 46.7 4.4 4.50 66.7 53.3 13.411.17 91.1 84.4 6.7 3.82 02-029 37.8 42.2 -4.4 -5.50 42.2 31.1 11.115.14 75.6 75.6 0.0 0.00 mean 50.0 47.0 3.0 3.26 49.8 42.4 7.3 8.85 86.2 84.9 1.3 0.79 stdev 7.2 7.7 7.2 8.12 11.7 13.1 7.2 8.68 7.2 7.5 4.6 2.78 VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime % acc RH = percent response accu racy with the right hand calculated as number of co rrect responses / total number of stimuli x 100 % acc LH = percent response accu racy with the left hand calculated as number of correct resp onses / total number of stimuli x 100 diff = difference between % acc RH and % acc LH (i.e. RH minus LH) ratio (asymmetry ratio) = (% acc RH % acc LH) / (% acc RH + % acc LH) DNT = did not test, SD = standard deviation

PAGE 76

64 Table 3-6: Response accuracy (percent) data with difference scores and asymmetry ratios for individual subjects in AD group for Tasks 1, 2, and 3. Task 1-VCP Task 2-PI Task 3-CP subj# % acc RH % acc LH diff ratio % acc RH % acc LH diff ratio % acc RH LH diff ratio 01-001 13.3 8.9 4.4 19.82 15.6 6.7 8.9 39.91 51.1 42.2 8.9 9.54 01-003 11.1 15.6 -4.5 -16.85 4.4 15.6 -11.2-56.00 60.0 64.4 -4.4 -3.54 01-004 24.4 11.1 13.3 37.46 DNTDNT N/AN/A 26.7 26.7 0.0 0.00 01-005 22.2 15.6 6.6 17.46 37.8 20.0 17.830.80 68.9 66.7 2.2 1.62 01-006 20.0 20.0 0.0 0.00 15.6 8.9 6.7 27.35 51.1 46.7 4.4 4.50 01-007 26.7 28.9 -2.2 -3.96 8.9 2.2 6.7 60.36 57.8 53.3 4.5 4.05 01-008 24.4 17.8 6.6 15.64 37.8 24.4 13.421.54 71.1 66.7 4.4 3.19 01-009 22.2 17.8 4.4 11.00 48.9 28.9 20.025.71 82.2 77.8 4.4 2.75 01-010 20.0 24.4 -4.4 -9.91 17.8 15.6 2.2 6.59 62.2 66.7 -4.5 -3.49 01-011 44.4 26.7 17.7 24.89 51.1 22.2 28.939.43 64.4 60.0 4.4 3.54 01-012 28.9 24.4 4.5 8.44 46.7 24.4 22.331.36 62.2 57.8 4.4 3.67 01-013 35.6 37.8 -2.2 -3.00 37.8 33.3 4.5 6.33 62.2 60.0 2.2 1.80 01-014 31.1 24.4 6.7 12.07 37.8 24.4 13.421.54 DNTDNT N/A N/A 01-015 31.1 22.2 8.9 16.70 26.7 22.2 4.5 9.20 DNTDNT N/A N/A 01-016 17.8 22.2 -4.4 -11.00 17.8 11.1 6.7 23.18 DNTDNT N/A N/A 01-017 33.3 40.0 -6.7 -9.14 44.4 28.9 15.521.15 DNTDNT N/A N/A 01-018 53.3 40.0 13.3 14.26 33.3 17.8 15.530.33 66.7 68.9 -2.2 -1.62 01-019 31.1 26.7 4.4 7.61 28.9 22.2 6.7 13.11 48.9 55.6 -6.7 -6.41 01-020 15.6 13.3 2.3 7.96 22.2 11.1 11.133.33 57.8 60.0 -2.2 -1.87 01-021 51.1 46.7 4.4 4.50 46.7 42.2 4.5 5.06 88.9 82.2 6.7 3.92 01-022 46.7 42.2 4.5 5.06 26.7 20.0 6.7 14.35 75.6 77.8 -2.2 -1.43 01-024 26.7 15.6 11.1 26.24 46.7 26.7 20.027.25 73.3 66.7 6.6 4.71 Mean 28.7 24.7 4.0 7.97 31.1 20.4 10.720.57 62.8 61.1 1.7 1.38 SD 11.7 10.7 6.5 13.55 14.1 9.4 8.7 21.96 13.9 13.4 4.5 3.87 VC = verbal command, PI = pantomime imitation, CP = conceptual pantomime % acc RH = percent response accu racy with the right hand calculated as number of co rrect responses / total number of stimuli x 100 % acc LH = percent response accu racy with the left hand calculated as number of correct resp onses / total number of stimuli x 100 diff = difference between % acc RH and % acc LH (i.e. RH minus LH) ratio (asymmetry ratio) = (% acc RH % acc LH) / (% acc RH + % acc LH) DNT = did not test, SD = standard deviation

PAGE 77

Table 3-7: Error analysis desc riptive data for Task 1 (VC) group hand %P %R %N %H %S %T %O %A %IC %EC%BPT%M %C %NR%UR LH 0.00 0.23 0.00 0.00 9.74 5.80 3.48 1.62 27.4914.390.93 32.480.12 0.93 2.78 HC RH 0.12 0.00 0.12 0.00 10.645.57 4.21 2.35 30.6913.371.61 29.460.00 0.25 1.61 LH 0.13 0.91 0.21 0.34 8.92 4.68 3.60 2.72 27.3217.014.27 26.020.86 1.19 5.89 AD RH 0.17 1.02 0.08 0.34 8.48 4.58 3.39 2.97 26.0416.883.73 25.610.85 0.93 4.92 VC = verbal command, HC = healthy control, AD = Alzheimers disease, RH = right hand, LH = left hand % = percentage of, P=Perseverative e rrors, R=Related, N=Non-re lated, H=Hand, S=Spatial, T=Timing, O=Occurrence, A=Amplitude, IC=Internal Configuration, EC=External Configuration, BPT=Body-part -as-tool, M=Movement, C=Concretisation, NR=No response, UR=Unr ecognizable response Percentages for error types were calculated as numb er of error present / tota l number of errors x 100 Table 3-8: Error analysis desc riptive data for Task 2 (PI) group hand %P %R %N %H %S %T %O %A %IC %EC%BPT%M %C %NR %UR LH 0.21 0.21 0.21 0.00 10.102.23 0.53 0.74 33.4818.490.11 31.030.00 0.21 2.44 HC RH 0.13 0.00 0.00 0.00 11.412.41 0.63 0.76 34.9814.830.38 31.050.00 0.76 2.66 LH 0.00 0.07 0.00 0.00 11.013.20 1.04 1.26 33.1119.200.67 25.520.00 0.67 4.24 AD RH 0.09 0.00 0.09 0.00 13.492.66 0.83 1.19 33.1217.250.64 26.510.00 0.09 4.04 PI = pantomime imitation, HC = healthy control, AD = Al zheimers disease, RH = right hand, LH = left hand % = percentage of, P=Perseverative e rrors, R=Related, N=Non-re lated, H=Hand, S=Spatial, T=Timing, O=Occurrence, A=Amplitude, IC=Internal Configuration, EC=External Configuration, BPT=Body-part -as-tool, M=Movement, C=Concretisation, NR=No response, UR=Unr ecognizable response Percentages for error types were calculated as numb er of error present / tota l number of errors x 100 65

PAGE 78

66 Table 3-9: Error analysis desc riptive data for Task 3 (CP). Group hand %P %R %N %H LH 1.98 56.44 4.95 36.63 HC RH 0.98 53.92 5.88 39.22 LH 1.96 31.37 3.27 63.40 AD RH 1.27 29.30 1.91 67.52 CP = conceptual pantomime, HC = hea lthy control, AD = Alzheimers disease RH = right hand, LH = left hand % = percentage of, P=Perseverative e rrors, R=Related, N=N on-related, H=Hand, Percentages for error types were calculated as number of error pres ent / total number of errors x 100 Table 3-10: Error totals for tasks 1, 2, and 3 group task hand error total LH 862 HC RH 808 LH 3605 AD 1 RH 1179 LH 941 HC RH 789 LH 1344 AD 2 RH 1090 LH 101 HC RH 102 LH 153 AD 3 RH 157 HC = healthy control, AD = Alzheimers di sease, RH = right hand, LH = left hand task 1 = verbal command pantomime, task 2 = pantomime imitation, task 3 = conceptual pantomime % = percentage of, P=Perseverative e rrors, R=Related, N=N on-related, H=Hand Error total is the sum of all errors produced by each group for each task. For task 3, only content errors are included in the sum of errors.

PAGE 79

67 CHAPTER 4 DISCUSSION Apraxia is a movement disorder in which voluntary movement is impaired without muscle weakness. This impairment affects th e ability to select a nd sequence previously learned skilled movements. Limb apraxia speci fically refers to an acquired disorder of skilled movement that affects hand and ar m function. In order to perform skilled movements, sensory input must interact with stored movement representations that are translated into patterns of innervation. Empirical evidence has shown that the neural representations for skilled m ovement are located in the pa rietal lobe of the left hemisphere. In order to perform skille d movements with the right hand, praxis movement representations and innervatory patte rns in the left hemisphere must transfer motor program information to left primary motor cortex via intrahemispheric white matter projections. In order to perform sk illed movements with the left hand, praxis movement representations and innervatory patte rns in the left hemisphere must transfer motor program information to right primar y motor cortex via interhemispheric white matter fibers. This study proposed to invest igate the neural mechanisms of limb apraxia by examining the transfer of different type s of praxis information from the left hemisphere to the right hemisphere via th e corpus callosum. AD was proposed as a model for studying this process because individu als in this populati on can perform praxis tasks with both hands (i.e. they do not have hemiplegia), this di agnosis is prevalent among the elderly population (i.e. this is not a rare syndrome), and individuals in this

PAGE 80

68 population demonstrate both limb apraxia and callosal atrophy (i.e. can potentially differentiate what type of information is bei ng transferred across the corpus callosum). Previous studies have provided evidence of neuronal loss in the areas of the brain that govern skilled movement systems (i.e. left parietal lobe) and this likely contributes to the presence of apraxia in the right hand of right-handed individuals with AD. Other studies have suggested neuronal loss in the cort ical layers that proj ect to contralateral motor areas (i.e. corpus callosu m) and this could explain th e presence of apraxia in the left hand of right-handed indivi duals with AD. Therefore, the goal of this study was to examine whether praxis information is tran sferred across the cor pus callosum and what type of praxis information is transferred acr oss the corpus callosum Investigations of callosal apraxia use asymmetries in right and left hand performance on praxis tasks (pantomime to command, pantomime imitation and conceptual pantomime) to examine the mechanisms of transfer of praxis inform ation in terms of white matter disconnections. It has been shown previously that praxis performance in individuals with AD is significantly different than prax is performance of healthy elde rly individuals. Therefore, this study attempted to investigate if there was a greater disparity between right hand and left hand performance on praxis tasks in indi viduals with AD as compared to healthy elderly individuals. Based on descriptions of individuals with ca llosal disconnection (De Renzi et al., 1982; Gazzaniga et al., 1967; Geschwind & Ka plan, 1962; Graff-Radford et al., 1987; Watson & Heilman, 1983), investigation of the transf er of praxis production and conceptual information was proposed. Summary and Explanation of Findings Research Question 1 Do individuals with AD have ideomotor and/ or conceptual apraxi a in the left hand?

PAGE 81

69 Summary. In order to answer this questi on, left hand performance of each group on the verbal command pantomime and concep tual pantomime tasks was compared. A significant difference in performance (with th e left hand) between the two groups on the verbal command pantomime task would indicat e the presence of ideomotor apraxia in the left hand of individuals with AD. Statistical analysis of the data revealed that there was a significant difference (with the left hand) be tween the two groups on the verbal command pantomime task indicating that individuals w ith AD have ideomotor apraxia in the left hand. A significant difference in performan ce (with the left hand) between the two groups on the conceptual pantomime task w ould indicate the presence of conceptual apraxia in the left hand of individuals with AD. Statistical analysis of the data revealed that there was a significant difference (with the left hand) between the two groups on the conceptual pantomime task indicating that i ndividuals with AD have conceptual apraxia in the left hand. Based on these findings, it can be concluded that individuals with AD demonstrated both ideomotor and con ceptual apraxia with the left hand. Explanation. Left hand performance was examined to answer this research question for two reasons. First, previous studi es have reported ideomotor and conceptual apraxia in the right ha nd of individuals with AD but there are no reports in the literature that address left hand performance. Since the goal of this study was to examine bimanual praxis mechanisms, the first step was to establ ish patterns of apraxia in the left hand that were similar to previously reported patterns of apraxia in the right hand. Second, left hand performance requires recru itment of both left hemisphere praxis representations and right hemisphere motor areas and requires the transfer of praxis movement and conceptual information across the corpus callosum. If there is ideomotor and conceptual

PAGE 82

70 apraxia in the left hand, it is unknown whethe r this results from degradation of left hemisphere praxis movement representations or deficient transfer of information from praxis movement representations across the corpus callosum. The next two research questions were aimed at illuminating which of these two processes contributes to the ideomotor and conceptual apraxi a in individuals with AD. The individuals with AD in this study dem onstrated limb apraxia with the left hand that was similar to the limb apraxia in the ri ght hand described in previous studies. Like the participants with AD in previous studi es, the individuals with AD in this study demonstrated impaired performance on ve rbal command pantomime and conceptual pantomime tasks (Ochipa et al., 1992; Schwar tz et al., 2000; Travni czek-Marterer et al., 1993). On verbal command pantomime tasks, previous studies have reported that individuals with AD produce more body part as tool errors th an healthy elderly individuals (Kato et al., 2000) and more spatial and temporal than content errors for transitive gestures (Foundas et al., 1999) (all of the s timuli in this study required transitive gestures, i.e., requi red using a tool to act on an object) and the individuals with AD in this study showed these same characterist ics with both the right and the left hands. With regards to the conceptual pantomime task, the AD group made more total errors, more conceptual errors, and more unrecogni zable errors than the HC group with both hands. Research Question 2 What is the contribution of degraded praxis movement and conceptual representations (due to corti cal atrophy) to the limb apra xia in individuals with AD? Summary. This issue was examined by comparing right hand performance of each group on the verbal command pantomime and c onceptual pantomime tasks. A significant

PAGE 83

71 difference in performance (with the right hand) between the two groups on the verbal command pantomime task would provide eviden ce that praxis movement representations are degraded in individuals with AD. Statisti cal analysis of the data revealed that there was a significant difference (with the right hand) between the two groups on the verbal command pantomime task indicating that praxis movement representations are degraded in individuals with AD. A significant diffe rence in performance (with the right hand) between the two groups on the conceptual pa ntomime task would provide evidence that action semantics representations are degraded in individuals with AD. Statistical analysis of the data revealed that there was a signi ficant difference (with th e right hand) between the two groups on the conceptual pantom ime task providing evidence that action semantics representations are degraded in in dividuals with AD. Based on these findings, there is evidence to suggest that both pr axis movement representations and action semantic representations are degr aded in individuals with AD. Explanation. Right hand performance was examined to answer this question because right hand performance does not require the transfer of praxis information across the corpus callosum but requires within hemi sphere access to praxis information. The results of this study have supported the notion that praxis movement representations and action semantics representations in the left he misphere are degraded such that individuals with AD demonstrate ideomotor and conceptual apraxia in both hands. Several studies have found cortical atrophy in the temporal and parietal l obes in individuals with AD (Foundas et al., 1996; Halliday et al., 2003; Pantel et al., 2004; Thompson et al.,2001; Thompson et al., 2003). Because these area s are important for praxis information processing (production and conceptual), it is lik ely that the apraxia in the right hand of

PAGE 84

72 individuals with AD can be attributed to co rtical atrophy in the re gions that subserve praxis production and conceptual informati on processing. However, the question remains whether the disruption of interh emispheric transfer of praxis information also contributes to the ideomotor and conceptual apraxia in the left hand of individuals with AD. Therefore, it was necessary to examine the role of interhemispheric tr ansfer of different types of praxis information and these anal yses could potentially refute the proposed localization of these functions within the le ft hemisphere of indi viduals with AD. Research Question 3 What is the contribution of interhemisphe ric disconnection (due to callosal atrophy) to apraxia in individuals with AD? Summary. This issue was examined by comparing the disparity or asymmetry between right hand and left hand performance of the two groups on praxis production and conceptual tasks. A significant difference in praxis asymmetry between the two groups on the verbal command pantomime task and th e pantomime imitation task would indicate that information from praxis movement repres entations is not being transferred across the corpus callosum in individuals with AD. Sta tistical analysis of th e data revealed that there was a significant difference in praxis asymmetry between the two groups on the verbal command pantomime task and the pant omime imitation task indicating that praxis movement representations are not being ad equately transferre d across the corpus callosum in individuals with AD. A significant difference in praxis asymmetry between the two groups on the conceptual pantomime ta sk would indicate that information from action semantics representations is not being transferred across the corpus callosum in individuals with AD. Statistic al analysis of the data re vealed that there was not a significant difference in praxis asymmetry between the two groups on the conceptual

PAGE 85

73 pantomime task indicating that right hemi sphere motor areas are able to access information from action semantics representa tions. There is evidence from these analyses to suggest that information from pr axis movement representations is not being transferred across the corpus callosum in individuals with AD. Additional findings suggest that the right hemisphere is able to access information from action semantics representations in individuals with AD. Explanation. The disparity between right hand and left hand performance was used to answer this question because praxis performance with the left hand relies on interhemispheric transfer of praxis information. A loss of pyramidal neurons in the third cortical layer that project to analogous areas of the cont ralateral hemisphere via the corpus callosum results in atrophy of speci fic regions of the corpus callosum in individuals with AD. With regards to the co rpus callosum, several studies have reported a reduction in the total area of the corpus callosum while other studies have suggested reductions in specific regions of the corpus callosum (Biegon et al., 1994; Black et al., 2000; Pantel et al., 1998; Teipel et al., 1998; Teipel et al., 1999) Based on patients described by Kazui and Sawada (1993), Wats on and Heilman (1983), and Degos et al. (1987), fibers in the anterior portion of the corpus callosu m are thought to be important for interhemispheric transfer of praxis info rmation. Evidence from Weis et al. (1991) indicated a significant decreas e in volume of the anterior corpus callosum without a significant decrease in volume of the posteri or corpus callosum. Hampel et al. (1998) also noted decreased area in the most rostral (genu and anterior body) and caudal (splenium) regions of the corpus callosum without reduction of the posterior body. These

PAGE 86

74 findings suggest that degeneration of ca llosal fibers could interfere with the interhemispheric transfer of praxis in formation in individuals with AD. The results of this study i ndicated a disruption in the tr ansfer of information from praxis movement representations but not in the transfer of information from action semantics representations. It could be that the callosal disconnection that results in deficient transfer of praxis information is specific to production information such that information specific to the conceptual attributes of the movement is able to be transferred across the corpus callosum while informati on specific to the temporal and spatial specifications of the movement is not bei ng adequately transferred across the corpus callosum. Alternatively, these findings might provide evidence that praxis movement representations are localized within the left hemisphere but actions semantics representations may have a bihemispheric distri bution that allows the right hemisphere to access praxis conceptual information. So even if praxis conceptual information that is stored in the left hemisphere cannot be tran sferred across the corpus callosum, the right hemisphere may be able to access whatever act ion semantics representation is needed to complete a given task. These findings suggest th at it is likely that the conceptual apraxia in both hands of individuals with AD is re lated solely to degraded action semantics representations due to bilateral cortical atr ophy while the ideomotor apraxia in the left hand of individuals with AD can be attribut ed to a combination of degraded left hemisphere praxis movement representations and deficient interhemispheric transfer of praxis information. However, we have not addressed whether the ideomotor apraxia in individuals with AD could result from an intr ahemispheric verbal motor disconnection or whether it is the verbal input or motor re presentations that are not being adequately

PAGE 87

75 transferred across the corpus callosum in i ndividuals with AD. The final research question will address these two issues. Research Question 4 Is the disruption of praxis information transfer a result of an intrahemispheric verbal motor disconnection or an interh emispheric corpus callosum disconnection? Summary. Right hand performance of both groups on the verbal command pantomime and pantomime imitation tasks were compared to address this issue. A significant difference between right hand perf ormance of the two groups on these tasks would suggest that impaired performance of individuals with AD results from an intrahemispheric verbal motor disconnection. Th e results of this anal ysis were significant for the main effect of group but not task and there was not a significant interaction between group and task. This suggests that the impaired performance of individuals with AD on the verbal command pantomime and pa ntomime imitation tasks results from an interhemispheric disconnecti on rather than an intrahem ispheric disconnection. These results also imply that the interruption of pr axis movement representation transfer across the corpus callosum is not dependent on the transfer of verbal information. The asymmetry between right and left ha nd performance of individuals with AD on the verbal command pantomime and pantomim e imitation tasks was also compared. A significant difference between these two ta sks would provide further evidence that individuals with AD are unable to transf er information from praxis movement representations across the corpus callosum. There was a significant difference in praxis asymmetry of individuals with AD on the verbal command pantomime and pantomime imitation tasks indicating that the presen ce of verbal input cannot account for the

PAGE 88

76 deficient transfer of movement informati on across the corpus callosum in individuals with AD. Explanation. Thus far, we have examined the mechanisms of praxis information transfer using the verbal command pantomime task. The verbal command pantomime task requires processing of verbal information and an interaction between language processing centers and praxis movement repr esentations. The pantomime imitation task requires processing of visual information and in volves solely the transfer of information from praxis movement representations to motor areas for movement execution. Therefore in order to determine whether verb al input interferes with interhemispheric transfer of praxis information it was necessary to compare the performance of individuals with AD on these two tasks. If individuals with AD demonstrated impa ired performance on the verbal command pantomime task but not the pantomime imitatio n task, this would indicate that verbal input was interfering with inte rhemispheric transfer of praxis information. However, the individuals with AD in this study demonstrated impaired performance on both the verbal command pantomime and pantomime imitation tasks meaning that the information that is unable to cross the corpus callosum is motor in nature (i.e. information from praxis movement representations that contain the temporal and spatial specifications of a movement). Furthermore, if the disparity or asymmetry between ha nds is significantly different in individuals with AD, this woul d provide further evidence that information from praxis movement representations is inadequately transferred across the corpus callosum. If verbal command performan ce was more asymmetric than pantomime imitation performance (based on group mean asymmetry), this would be considered an

PAGE 89

77 intrahemispheric verbal motor disconnection but because pantomime imitation performance was more asymmetric than verb al command pantomime performance (based on group mean asymmetry), it can be concluded that there is an interhemispheric callosal motor disconnection in individuals with AD. Conclusions According to the results of this study, individuals with AD have conceptual and ideomotor apraxia in both the dominant (ri ght) and nondominant (left) hands. Based on the finding of a significant difference in ri ght hand performance on the verbal command pantomime and conceptual pantomime tasks, it can be concluded that praxis movement representations and action semantics representa tions are degraded in individuals with AD and that degraded movement representations contribute to the ideomotor apraxia while degraded semantic representations contribute to the conceptual ap raxia in individuals with AD. Previous studies have provided ev idence of neuronal loss in the areas of the brain that govern skilled m ovement systems which likely re sults in the degradation of praxis movement and conceptual represen tations. A significant difference in the asymmetry of performance of the two groups on the praxis production tasks but not the praxis conceptual task indicates that defici ent transfer of praxis information across the corpus callosum contributes to the ideomotor but not the conceptual apraxia in the left hand of individuals with AD. Other studies have suggested neuronal loss in the cortical layers that project to contralateral motor areas (i.e., corpus callosum) and this could explain the deficient transfer of praxis production information. So while the apraxia in the right hand of individuals with AD can be attributed to degraded representations, a combination of degraded movement representations and deficient interhemispheric transfer of prax is information most likely explains the

PAGE 90

78 ideomotor apraxia in the left hand. According to the results of this study, the conceptual apraxia in the left hand of individuals with AD is related to the degradation of action semantics representations but left hand perf ormance on the conceptual pantomime task was better than performance on the imitati on or command tasks because the right hemisphere can still access acti on semantics information. This may be because deficient callosal transfer of information is specific to the transfer of spatial and temporal information or because semantic information is represented in a distributed network that encompasses both hemispheres. Additionally, the study findings suggested that the disconnection in individuals with AD is an interhemispheric callosa l motor disconnection rather than an intrahemispheric verbal motor disconnection meaning that verbal information does not interfere with the interhemispheric transf er of praxis information. Deficient interhemispheric transfer is specific to the tr ansfer of information from praxis movement representations. Future studies will investigate the particular types of information from praxis movement representation that are una ble to cross the corpus callosum by using a discriminate analysis of the error data. The purpose of this study was to investig ate how praxis information processing is represented in the brain by examining the transf er of different types of praxis information across the corpus callosum. The findings of this study support the notion that praxis movement representations are localized in the left hemisphere of right handed individuals and suggest the conclusion that action seman tics representations are distributed across both hemispheres. In addition, it can be surm ised that only information from praxis movement representations is transferred acro ss the corpus callosum and that information

PAGE 91

79 from the input modality must access praxis movement representations prior to the interhemispheric transfer of praxis information. In conclusion, it is evident that the interaction between left hemisphere prax is movement representations and right hemisphere motor areas is necessary for left hand movement precision and intact praxis movement representations and action sema ntics representations are necessary for bimanual movement precision. All of these out comes provide significa nt contributions to the study of praxis mechanisms. The fact that measurements of cortical and callosal volumes were not obtained and that these measurements were not correlate d with the presence of limb apraxia is a potential weakness of this study. The absence of this data limits the ability to draw conclusions about the neuroanatomical correlat es of the praxis mechanisms described in this study. However, this does not negate th e value of the findings of the current study because it is possible that individuals with AD can still demonstrate limb apraxia in both hands without showing radiological evidence of cortical or callosal atrophy. While it is possible to measure the volume of cortical a nd callosal structures, the integrity of the pyramidal cells in the cortical layers that are responsible for transferring information within and between hemispheres is not m easurable while the individual is living (examination of senile placques and neurofibrillary tangles in the cortical layers requires post-mortem analysis). Therefore, regardle ss of the presence or absence of cortical and/or callosal atrophy in individuals with AD limb apraxia in both hands may still be present due to underlying neuranatomical pro cesses that cannot be adequately examined. Additionally, it should be noted that an attempt was made to balance the two groups for age and education level. While the experimenter succeeded in matching the

PAGE 92

80 two groups for age, there was a significant di fference in education level between the two groups. Unfortunately, low education level is one of the predictive factors for development of this type of dementia. Finding healthy elderly control subjects that matched subjects in the AD group for both ag e and education leve l was difficult. Typically, in the AD population older individual s have low levels of education while in the HC population older individuals have high levels of education. Because low education level is a predictive factor for developing AD, most individuals within the age group studied (i.e., 60-90 yrs old) who had low levels of education had developed symptoms of AD so healthy elde rly individuals that were re cruited because they matched the AD subjects for age and gender had higher levels of education. Lastly, it is important to poi nt out that general cognitive decline in the AD group is a potential alternative explanat ion for the findings of this study. It is possible that individuals with AD demonstr ated impaired performance on limb praxis tasks because their overall cognitive abilities are affected by AD and not because of degraded praxis representations or deficient in terhemispheric transfer of prax is information. In this study, the design attempted to control for the effects of general cognitive decline on limb praxis performance by enrolling individuals in the ea rlier stages of the di sease process and by excluding individuals who exhibi ted cognitive defic its that would inte rfere with their performance on the experimental tasks (like visual object agnosia and severe auditory comprehension deficits). Therefore, it is not likely that the eff ect of general cognitive decline had a significant impact on these results. Implications Rothi and Horner (1982) described two th eories of physiologic mechanisms of recovery that can be applied to rehabilitati on of individuals with neurologic disease or

PAGE 93

81 injury. Restitution of function suggests that as the lesion area heals neural pathways resume activities and the f unctions subserved by the i nvolved neural systems are restored (p. 74). Substitution of function suggests that the brain is physiologically capable of spontaneous restoration of f unction beyond the acute phase of recovery through substitution and reorganization of ne uronal structures (p. 74). Behavioral treatment approaches consistent with a restitution-of-function m odel are based on the idea that functions are lost or impaired fo llowing brain damage and that lost function must be retrained and impaired functions must be maximally stimulated in order to be maintained (p.77). Behavioral treatment a pproaches consistent with a substitution-of function model are based on the idea that the clinician treats that patient by discouraging the use of ineffective strategies while en couraging the use of new strategies not previously available to him (p. 78). Studies that have attempted to treat individuals with lim b apraxia can be subdivided into approaches that are consistent with a restitution-of -function model (Butler, 1997; Maher, Rothi, & Greenwald, 1991; Ochipa, Maher, & Rothi, 1995; Smania, Girardi, Domencali, Lora, & Aglioti, 2000; Wilson, 1988) and approaches that are consistent with a substitution-of-function model (Donkervoor t, Dekker, Stehmann-Saris, & Deelman, 2001; Goldenberg, Daumuller, & Hagmann, 2001; Goldenberg & Hagmann, 1998; van Heugten, Dekker, Dellman, van Dijk, Stehlm ann-Saris, &Kinebanian, 1998). Those studies that aimed to restore praxis func tions usually demonstrated improvement on outcome measures but the improvement was ty pically limited to gestures targeted during the treatment. Strategy training in indivi duals with apraxia was designed to teach strategies to compensate for ap raxia rather than rehabilitate the apraxia itself and these

PAGE 94

82 approaches tended to be successful in institut ing compensatory strategies that allow the patient to function more independently despite the persistence of apraxia. For individuals with AD, who have apraxia that impacts th eir ability to perform everyday activities independently, perhaps a combination of thes e two treatment methods would be useful. Additionally, because the results of this st udy have shown the individuals with AD have apraxia in both the right and left hands, assessment of apraxia should include the examination of both right hand and left ha nd performance and apraxia treatment should comprehensively be aimed at improving function in both hands. With respect to AD, limb apraxia continues to be an important area of study. This study has shown that many areas of praxis f unction are impacted by the cognitive decline that characterizes AD. Since numerous studies have shown the impact of limb apraxia on this population and the resultant burdens asso ciated with its pres ence (Foundas et al., 1995; Giaquinto et al., 1999; Saeki et al., 1995), research into the nature, assessment and treatment of this disorder in individuals with AD should continue to be vigorously pursued.

PAGE 95

83 APPENDIX A LIST OF STIMULI stimulus # Task 1VC Task 2PI Task 3CP 1 Show me how you would hold and use a paddle to play ping pong. imitate using a paddle to play ping pong pictured object: ping pong ball and table 2 Show me how you would insert a p lug into an electrical outlet. imitate inserting a plug into an electrical outlet pictured object: electrical outlet 3 Show me how you would hold and use a razor to shave your face. imitate shaving your face pictured object: unshaven face 4 Show me how you would hold and use a match to light a candle imitate lighting a candle with a match pictured object: unlit candle 5 Show me how you would hold and use a screwdriver to turn a screw into the wall. imitate using a screwdriver to turn a screw into the wall pictured object: screw sticking out of a piece of wood 6 Show me how you would throw a bowling ball. imitate throwing a bowling ball pictured object: upright bowling pins 7 Show me how you would hold and use an iron to press a shirt. imitate ironing a shirt pictured object: ironing board with a shirt on it 8 Show me how you would beat a drum with a drumstick. imitate drumming pictured object: drum set 9 Show me how you would hold and use a spatula to turn eggs in a frying p an. imitate turning eggs with a spatula pictured object: skillet with eggs in it 10 Show me how you would hold and use a knife to spread butter on bread. imitate spreading butter on bread with a knife pictured object: piece of bread with butter on it 11 Show me how you would hold and use a paint roller to paint a wall. imitate using a paint roller to paint a wall pictured object: paint roller pan

PAGE 96

84 stimulus # Task 1VC Task 2PI Task 3CP 12 Show me how you would hold and use a spoon to eat a bowl of soup. imitate using a spoon to eat soup pictured object: bowl of soup 13 Show me how you would hold and use a paintbrush to paint on an easel. imitate using a paintbrush to paint on a canvas pictured object: painter's easel and palette 14 Show me how you would hold and use a paintbrush to paint a wall in f ront of you. imitate using a paintbrush to paint a wall pictured object: open paint can 15 Show me how you would hold and use a knife to carve a turkey. imitate using a knife to carve a turkey pictured object: whole turkey 16 Show me how you would throw a dart at a dart board. imitate throwing a dart pictured object: dart board 17 Show me how you would hold and use a spoon to stir a cup of coffee. imitate stirring coffee pictured object: cup of coffee and open packet of sugar 18 Show me how you would hold and use a saw to cut wood on a sawhorse imitate sawing wood pictured object: wood on a sawhorse 19 Show me how you would hold and use a match to light a fire in a f ireplace. imitate using a match to light a fire pictured object: wood in a fireplace 20 Show me how you would hold and use lipstick to paint your lips imitate using lipstick pictured object: lips with partial lipstick 21 Show me how you would hold and use a spatula to cut and serve cake. imitate cutting and serving cake pictured object: cut bundt cake 22 Show me how you would use a jack to lift a car t hat had a flat tire. imitate using a jack to fix a flat tire pictured object: car with a flat tire 23 Show me how you would hold and use a fork to eat dinner. imitate using a fork to eat dinner pictured object: plate of food on table

PAGE 97

85 stimulus # Task 1VC Task 2PI Task 3CP 24 Show me how you would hold and use a toothbrush to brush your teeth. imitate brushing your teeth pictured object: dirty teeth 25 Show me how you would hold and use a hammer to pound a nail into the wall. imitate hammering a nail into a wall pictured object: nail sticking out of a piece of wood 26 Show me how you would hold and use a key to unlock a door. imitate unlocking a door with a key pictured object: keyhole and doorknob 27 Show me how you would hold and use a hammer to remove a bent nail f rom a piece of wood. imitate removing a bent nail from wood with a hammer pictured object: bent nail in a piece of wood 28 Show me how you would hold and use a shovel to scoop sand into a bucket. imitate scooping sand into a bucket with a shovel pictured object: sandbox with sand and pail 29 Show me how you would hold and use a comb to fix your hair. imitate combing your hair pictured object: messy hair 30 Show me how you would thread a needle. imitate threading a needle pictured object: spool of thread and a button 31 Show me how you would hold and use a turnkey to open a can of sardines. imitate using a turnkey to open a can of sardines pictured object: partially opened sardine can 32 Show me how you would hold and use a pencil sharpener to sharpen a broken pencil. imitate sharpening a pencil pictured object: broken pencil 33 Show me how you would hold and use a bottle opener to open a soda bottle. imitate opening a soda bottle with a bottle opener pictured object: soda bottle 34 Show me how you would hold and use a screwdriver to open a can of p aint. imitate using a screwdriver to open a paint can pictured object: closed paint can 35 Show me how you would turn off a dripping faucet. imitate turning off a dripping faucet pictured object: dripping faucet 36 Show me how you would throw a baseball to the catcher. imitate throwing a baseball pictured object: baseball catcher

PAGE 98

86 stimulus # Task 1VC Task 2PI Task 3CP 37 Show me how you would hold and use a hatchet to chop a log. imitate chopping a log with a hatchet pictured object: partially chopped log 38 Show me how you would hold and use tongs to serve ice imitate using tongs to serve ice pictured object: ice bucket and glass 39 Show me how you would hold and use an ice cream scoop to serve ice cream imitate using an ice cream scoop to serve ice cream pictured object: ice cream and cone 40 Show me how you would hold and use an eraser to clean a chalkboard. imitate erasing a chalkboard pictured object: scribbles on chalk board 41 Show me how you would hold and use a wrench to turn a bolt imitate using a wrench to turn a bolt pictured object: hexhead bolt 42 Show me how you would hold and use a gun to shoot at a target. imitate shooting a gun pictured object: human shaped target 43 Show me how you would roll up a car window imitate rolling up a car window pictured object: partially opened car window 44 Show me how you would hold and use clippers to trim a rose stem. imitate using clippers to trim a rose pictured object: rose and vase 45 Show me how you would hold and use scissors to cut a piece of paper imitate cutting paper with scissors pictured object: partially cut out paper doll

PAGE 99

87 APPENDIX B DESCRIPTION OF ERRORS Content P PerseverativeThe patient produces a ll or part of a previously produced pantomime. R RelatedThe pantomime is an accurately produced pantomime associated in content to the target. For example, th e participant might pantomime playing a trombone for a target of a bugle. N NonrelatedThe pantomime is an accurately produced pantomime not associated in content to the target. For example, the participant might pantomime playing a trombone for a target of shaving. H HandThe patient performs the action w ithout benefit of a real or imagined tool. For example, when asked to cut a piece of paper with scissors, they pretend to rip the paper. Another example would be turning a screw by hand rather than with an imagined screwdriver. Temporal S SequencingSome pantomimes require multiple positioning that are performed in a characteristic sequen ce. Sequencing errors involve any perturbation of this sequence includi ng addition, deletion, or transposition of movement element as long as th e overall movement structure remains recognizable. T TimingThis error reflects any altera tion from the typical timing or speed of a pantomime. May include abnorma lly increased, decreas ed, or irregular rate of production. O OccurrencePantomimes may characteris tically involve eith er single (i.e. unlocking a door with a key) or repet itive (i.e. screwing in a screw with a screwdriver) movement cycles. This error reflects any multiplication of characteristically single cycles or reduc tion of a characteristically repetitive cycle to a single event. Spatial A AmplitudeAny amplification, re duction, or irregularity of the characteristic amplitude of a target pantomime. IC Internal Configuration-This error type reflects any abnormality of the required finger/hand posture and its re lationship to the target tool. For example, when asked to pretend to br ush teeth, the participant may close the hand tightly into a fist with no space allowed for the imagined toothbrush handle. BPT Body Part as ToolThe patient uses finger, hand, or arm as the imagined tool of the pantomime. For example, when asked to pretend to smoke a cigarette, the participan t might puff on the end of an extended index finger.

PAGE 100

88 Spatial EC External configurationThis erro r type reflects any abnormality of the required finger/hand posture and its re lationship to the target object. For example, when asked to pretend to brush teeth, the participant might hold his hand next to his mouth without reflecting the distance necessary to accommodate an imagined toothbrush. M MovementWhen acting on an object with a tool, a movement characteristic of the action and n ecessary to accomplishing the goal is required. Any disturbance of the char acteristic movement of the action. For example, when asked to pantomime using a screwdriver, a participant may orient the imagined screwdriver correctly to the imagined screw but instead of stabilizing the shoulder and wrist while twisting at the elbow, the participant stabilizes the elbow and twists at the wrist or shoulder. Other C ConcretizationThe participant perfor ms a transitive pantomime not on an imagined object but instead on a real obj ect not normally used in the task. For example, when asked to pretend to saw some wood, they pantomime sawing on their leg. NR No Response UR Unrecognizable ResponseA response that is not recognizable and shares no temporal or spatial features of the target.

PAGE 101

89 LIST OF REFERENCES Afifi, A.K. & Bergman, R.A. (1998). Functional neuroanatomy: text and atlas McGraw-Hill: New York. Ball, J.A., Lantos, P.L., Jackson, M., Marsden, C.D., Scadding, J.W., & Rossor, M.N. (1993). Alien hand sign in associati on with Alzheimers histopathology. Journal of Neurology, Neurosurgery, and Psychiatry, 56, 1020-1023. Barclay, L.L., Zemcov, A., Blass, J.P., & Sans one, J. (1985). Survival in Alzheimers disease and vascular dementia. Neurology, 35 834-840. Biegon, A., Eberling, J.L., Richardson, B.C., Roos, M.S., Wong, S.T.S., Reed, B.R., & Jagust, W.J. (1994). Human corpus callosu m in aging and Alzheimers disease: a magnetic resonance imaging study. Neurobiology of Aging, 15, 393397. Black, S.E., Moffat, S.D., Yu, D.C., Parker, J., Stanchev, P., & Bronskill, M. (2000). Callosal atrophy correlates with temporal lobe volume and mental status in Alzheimers disease. The Canadian Journal of Neurological Sciences, 27, 204209. Boller, F. & Duyckaerts, C. (1997). Alzheimer di sease: clinical and anatomic aspects. In T.E. Feinberg & M.J. Farah (Eds.), Behavioral neurology and neuropsychology (pp. 521-544). New York: McGraw-Hill. Burns, A., Lewis, G., Jacoby, R., & Levy, R. (1991). Factors affecting survival in Alzheimers disease. Psychological Medicine, 21 363-370. Butler, J. (1997). Interventi on effectiveness: evidence from a case study of ideomotor and ideational apraxia. British Journal of Occupational Therapy, 60, 491-497. Canadian Study of Health and Aging Work ing Group (1994). Patterns of caring for people with dementia in Canada. Canadian Journal of Aging, 13 470-487. Cho, C., Cho, H., Cho, K., Choi, K., Oh, H., & Bae, C. (2001). Factors associated with functional dependence in Alzheimers disease. Journal of Clinical Geropsychology, 7 79-89. Cimino-Knight, A.M., Hollingsworth, A.L., Ma her, L.M., Raymer, A.M., Foundas, A.L., Heilman, K.M., & Rothi, L.J.G. (2002). Forms of recovery in ideomotor apraxia: a preliminary investig ation (abstract). Journal of the International Neuropsycholgical Society, 8, 207.

PAGE 102

90 Degos, J.D., Gray, F., Louarn, F., Ansquer, J.C., Poirier, J., & Barbizet, J. (1987). Posterior callosal infarction: cl inicopathological correlations. Brain, 110, 11551171. Della Salla, S., Lucchelli, F., & Spinnler, H. (1987). Ideomotor apraxia in patients with dementia of the Alzheimer type. Journal of Neurology, 234 91-93. DeRenzi, E., Faglioni, P., & Sorgato, P. (1982). Modality-specific and supramodal mechanisms of apraxia. Brain, 105, 301-312. DeRenzi, E. & Lucchelli, F. (1988). Ideational apraxia. Brain, 113, 1173-1188. Derouesne, C., Lagha-Pierucci, S., Thibault S., Baudouin-Madec, V., & Lacomblez, L. (2000). Apraxic disturbances in patients with mild to moderate Alzheimers disease. Neuropsychologia, 38 1760-1769. Donkervoort, M., Dekker, J., Stehmann-Saris, F.C., & Deelman, B.G. (2001). Efficacy of strategy training in left hemisphere str oke patients with apraxia: A randomized clinical trial. Neuropsychological Rehabilitation, 11, 549-566. Dumont, C., Ska, B., & Joanette, Y. (2000). Conceptual apraxia and semantic memory deficit in Alzheimers disease: two sides of the same coin? Journal of the International Neuropsyc hological Society, 6 693-703. Edwards, D.F., Deuel, R.K., Baum, C.M., & Mo rris, J.C. (1991). A quantitative analysis of apraxia in senile dementia of the Al zheimer type: stage related differences in prevalence and type. Dementia, 2 142-149. Fastenau, P.S., Denburg, N.L., & Mauer, B. A. (1998). Parallel short forms for the Boston Naming Test: psychometric propert ies and norms for older adults. Journal of Clinical and Experime ntal Neuropsychology, 20, 828-834. Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). Mini-mental State. A practical method for grading the cognitive stat e of patients for the clinician. Journal of Psychiatric Research, 12, 189-198. Foundas, A.L., Eure, K.F., & Seltzer, B. ( 1996). Conventional MRI volumetric measures of parietal and insular cortex in Alzheimers disease. Progress in Neuropsychopharmacology and Biological Psychiatry, 20, 1131-1144. Foundas, A.L., Macauley, B.L., Raymer, A.M., Maher, L.M., Rothi, L.J.G., & Heilman, K.M. (1999). Ideomotor apraxia in Alzhei mer disease and left hemispheric stroke: limb transitive and intransitive movements. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 12 161-166. Gazzaniga, M., Bogen, J., & Sperry, R. (1967). Dyspraxia following diversion of the cerebral commissures. Archives of Neurology, 16 606-612.

PAGE 103

91 Geschwind, N. (1965a). Disconnexion syndr omes in animals and man. Part I. Brain, 88 237-294. Geschwind, N. (1965b). Disconnexion syndrom es in animals and man. Part II. Brain, 88, 585-644. Geschwind, N. & Kaplan, E. (1962). A hu man cerebral disconnection syndrome. Neurology, 12 665-685. Giannakopoulos, P., Duc, M., Gold, G., Hof, P.R., Michel, J.P. & Bouras, C. (1998). Pathologic correlates of apraxi a in Alzheimer disease. Archives of Neurology, 55 689-695. Gianquinto, S., Buzzelli, S., Di Francesco, L., Lottarini, A., Montenero, P., Tonin, P., & Nolfe, G. (1999). On the prognosis of outcome after stroke. Acta Neurolgica Scandinavica, 100, 202-208. Goldenberg, G., Daumuller, M., & Hagmann, S. (2001). Assessment and therapy of complex activities of daily living in apraxia. Neuropsychological Rehabilitation, 11, 147-169. Goldenberg, G. & Hagmann, S. (1998). Therapy of activities of daily living in patients with apraxia. Neuropsychological Rehabilitation, 8, 123-141. Goldenberg, G., Wimmer, A., Holzner, F., & Wessely, P. (1985). Apraxia of the left limbs in a case of callosal disconnection: the contribution of medial frontal lobe damage. Cortex, 21, 135-148. Graff-Radford, N.R., Welsh, K., & Goders ky, J. (1987). Callosal apraxia. Neurology, 37, 100-105. Guilmette, T.J., (1997). Pocket guide to br ain injury, cognitive, and neurobehavioral rehabilitation. San Die go: Singular Publishing. Halliday, G.M., Double, K.L., Macdonald, V, & Kril, J.J. (2003). Identifying severely atrophic cortical subregions in Alzheimers disease. Neurobiology of Aging, 24, 797-806. Hampel, H., Teipel, S.J., Alexander, G.E., Ho rwitz, B., Teichberg, D., Schapiro, M.B., & Rapoport, S.I. (1998). Corpus callosum at rophy is a possible i ndicator of regionand cell typespecific neuronal degenera tion in Alzheimer Disease: a magnetic resonance imaging analysis. Archives of Neurology, 55, 193-198. Heilman, K.M. (1973). Ideationa l apraxiaa re-definition. Brain, 96 861-864. Heilman, K.M. & Rothi, L.J.G. (1993). Apra xia. In K.M. Heilman & E. Valenstein (Eds.) Clinical neuropsychology (pp.141-163). New York: Oxford University Press.

PAGE 104

92 Heilman, K.M., Rothi, L.J.G., & Valenstein, E. (1982). Two forms of ideomotor apraxia. Neurology, 32 342-346. Hof, P.R. & Morrison, J.H. (1999). The cellu lar basis of cortical disconnection in Alzheimer disease and related dementing conditions. In R.D. Terry, R. Katzman, K.L. Bick, & S.S. Sisodia (Eds.), Alzheimers disease (2nd ed) (pp. 207-232). Philadelphia: Lippincott, Williams, & Wilkins. Jacobs, D.H., Adair, J.C., Williamson, D.J. G., Na, D.L., Gold, M., Foundas, A.L., Shuren, J.E., Cibula, J.E., & Heilman, K.M. (1999). Apraxia and motor-skill acquisition in Alzheimers disease are dissociable. Neuropsychologia, 37 8975880. Janowsky, J.S., Kaye, J.A., & Carper, R.A. (1996). Atrophy of the corpus callosum in Alzheimers disease versus healthy aging. Journal of the American Geriatrics Society, 44, 798-803. Kaplan, E.F., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test. Philadelphia: Lea & Febiger. Kato, M., Meguro, K., Sato, M., Shimada, Y ., Yamazaki, H., Saito, H., Yamaguchi, S., & Yamadori, A. (2001). Ideomotor apraxia in patients with Alzheimer disease: why do they use their body parts as objects? Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 14 45-52. Kazui, S. & Sawada, T. (1993). Ca llosal apraxia without agraphia. Annals of Neurology, 33, 401-403. Kertesz, A. (1980). Western Aphasia Battery London, Ontario: University of Western Ontario Press. Kramer, M.S. & Feinstein, A.R. (1981). Clin ical biostatistics: The biostatistics of concordance. Clinical pharmacology and therapeutics, 29, 111-117. LeClerc, C.M. & Wells, D.L. (1998). Use of a content methodology process to enhance feeding abilities threatened by ideational apraxia in people with Alzheimers-type dementia. Geriatric Nursing, 19 261-268. Lezak, M.D. (1995). Neuropsychological assessment (3rd ed.). New York: Oxford University Press. Liepman, H. (1980). The left hemisphe re and action. (A translation from Munchener Medizinische Wochenschrift 1905, 48-49). Translations from Liepmanns essays on apraxia. In Research Bulletin #506 Department of Psychology, The University of Western Ontario.

PAGE 105

93 Lyoo, K., Satlin, A., Lee, C.K., & Renshaw, P.F. (1997). Regional atrophy of the corpus callosum in subjects with Alzheimers disease and multi-infarct dementia. Psychiatry Research: Neuroimaging, 74, 63-72. Maher, L.M. & Ochipa, C. (1997). Management and treatment of limb apraxia. In L.J.G. Rothi & K.M. Heilman (Eds.). Apraxia: the neuropsychology of action (pp. 75-92). East Sussex, UK: Psychology Press. Maher, L.M., Rothi, L.J.G., & Greenwal d, M.L. (1991). Treatment of gesture impairement: a single case. ASHA, 33, 195. Marks, R.G. (1999). Designing a research project: the art of doing science. Department of Statistics in the College of Medicine (Biostatistics), University of Florida, Gainesville, FL. Unpublished. McKhann, G., Drachman, D., Folstein, M., Ka tzman, R., Price, D., & Stadlan, E.M. (1984). Clinical diagnosis of Alzheimers disease: report of the NINCDS-ADRDA work group under the auspices of the Depa rtment of Health and Human Services task force on Alzheimers disease. Neurology, 34 939-944. Morris, J.C. (1999). Clinical presentation a nd course of Alzheimer disease. In R.D. Terry, R. Katzman, K.L. Bick, & S.S. Sisodia (Eds.), Alzheimer disease (2nd ed) (pp. 11-24). Philadelphia: Lippincott, Williams, & Wilkins. Ochipa, C., Maher, L.M., & Rothi, L.J.G. (1995) Treatment of ideomotor apraxia. (abstract) Journal of the Inte rnational Neuropsyc hological Society, 2, 149. Ochipa, C., Rothi, L.J.G, & Heilman, K.M. ( 1989). Ideational apraxia: a deficit in tool selection and use. Annals of Neurology, 25, 190-193. Ochipa, C., Rothi, L.J.G, & Heilman, K.M. (1992). Conceptual apraxia in Alzheimers disease. Brain, 115 1061-1071. Pantel, J. Schonknecht, P., Essig, M., Schrode r, J. (2004). Distribution of cerebral atrophy assessed by magnetic resona nce imaging reflects patterns of neuropsychological deficits in Alzheimers dementia. Neuroscience Letters, 361, 17-20. Pantel, J., Schroder, J., Essig, M., Minakaran, R., Schad, L.R., Friedlinger, M., Jauss, M., & Knopp, M.V. (1998). Corpus callosum in Alzheimers disease and vascular dementiaa quantitative ma gnetic resonance study. Journal of Neural Transmission, 54, 129-136. Pantel, J., Schroder, J., Jauss, M., Essig, M., Minikaran, R., Schonknecht, P., Schneider, G., Schad, L.R., & Knopp, M.V.(1999). Topography of callosal atrophy reflects distribution of regional ce rebral volume reduction in Alzheimers disease. Psychiatry Research: Neuroimaging, 90, 181-192.

PAGE 106

94 Poizner, H., Mack, L., Verfaellie, M., Roth i, L.J.G., & Heilman, K.M. (1990). Three dimensional computer graphic analysis of apraxia. Brain, 113 85-101. Poizner, H., Soechting, J.F., Bracewell, M., Rothi, L.J.G., & Heilman, K.M. (1989). Disruption of hand and joint kinematics in limb apraxia. Society of Neuroscience Abstracts, 15 196.2. Rapcsak, S.Z., Croswell, S.C., & Rubens, A.B. (1989). Apraxia in Alzheimers disease. Neurology, 39, 664-668. Raymer, A.M. & Ochipa, C. (1997). Concep tual praxis. In L.J.G. Rothi & K.M. Heilman (Eds.), Apraxia: the neuropsychology of action (pp. 51-60). East Sussex, UK: Psychology Press. Riddoch, M.J. & Humphreys, G.W. (1993). Birmingham Object Recognition Battery (BORB). New York: Psychology Press. Rocca, W.A., Bonaiuto, S., Lippi, A., Luciani, P., Turtu, F., Cavarzeran, F., & Amaducci, L. (1990). Prevalence of clinically di agnosed Alzheimers disease and other dementing disorders: a door-to-door su rvey in Appignano, Macerata Province, Italy. Neurology, 40 626-631. Rothi, L.J.G. & Heilman, K.M., & (1997). Intr oduction to limb apraxi a. In L.J.G. Rothi & K.M. Heilman (Eds.). Apraxia: the neuropsychology of action (pp. 1-6). East Sussex, UK: Psychology Press. Rothi, L.J.G., Heilman, K.M., & Watson, R. T. (1985). Pantomime comprehension and ideomotor apraxia. Journal of Neurology, Neur osurgery, and Psychiatry, 48 207210. Rothi, L.J.G. & Horner, J. (1982). Restituti on and substitution: Two theories of recovery with application to neurobehavioral treatment. Journal of Clinical Neuropychology, 5, 73-81. Rothi, L.J.G., Mack, L., Verfaellie, M., Brown, P. & Heilman, K.M. (1988). Ideomotor apraxia: error pattern analysis. Aphasiology, 2 381-387. Rothi, L.J.G., Ochipa, C., & Heilman, K. M. (1991). A cognitive neuropsychological model of limb praxis. Cognitive Neuropsychology, 8 443-458. Rothi, L.J.G., Ochipa, C., & Heilman, K.M. (1997a). A cognitive neuropsychological model of limb praxis and ap raxia. In L.J.G. Rothi & K.M. Heilman (Eds.), Apraxia: the neuropsychology of action (pp. 29-49). East Sussex, UK: Psychology Press. Rothi, L.J.G., Raymer, A.M., & Heilman, K.M. (1997b). Limb praxis assessment. In L.J.G. Rothi & K.M. Heilman (Eds.), Apraxia: the neuropsychology of action (pp. 61-73). East Sussex, UK: Psychology Press.

PAGE 107

95 Roy, E.A. & Square, P.A. (1985). Common c onsiderations in the study of limb, verbal, and oral apraxia. In E.A. Roy (Ed.), Neuropsychological studies of apraxia and related disorders (pp. 111-161). North Holland: Elsevier Science Publishers. Saeki, S., Ogata, H., Okubo, T., Takahashi, K ., & Hoshuyama, T. (1995). Return to work after stroke: a follow-up study. Stroke, 26, 399-401. Schwartz, R.L., Adair, J.C., Raymer, A. M., Williamson, D.J.G., Crosson, B., Rothi, L.J.G., Nadeau, S.E., & Heilman, K.M. ( 2000). Conceptual apraxia in probable Alzheimers disease as demonstrated by the Florida Action Recall Test. Journal of the International Neuropsychological Society, 6 265-270. Smania, N., Girardi, F., Domenicali, C., Lora E., & Aglioti, S. (2000). The rehabilitation of limb apraxia: a study in le ft brain damaged patients. Archives of Physical Medicine and Rehabilitation. 81, 379-388. Tanaka, Y., Iwasa, H., & Obayashi, T. (1990) Right hand agraphia and left hand apraxia following callosal damage in a right-hander. Cortex, 26, 665-671. Teipel, S.J., Bayer, W., Alexander, G.E., Ze buhr, Y., Teichberg, D., Kulic, L., Schapiro, M.B., Moller, H.J., Rapoport, S.I., & Hamp el, H. (2002). Progression of corpus callosum atrophy in Alzheimer Disease. Archives of Neurology, 59, 243-248. Teipel, S.J., Hampel, H., Alexander, G.E., Schapiro, M.B., Horwitz, B., Teichberg, D., Daley, E., Hippius, H., Moller, H.J., & Rapoport, S.I. (1998). Dissociation between corpus callosum atrophy and wh ite matter pathology in Alzheimers disease. Neurology, 51, 13811385. Teipel, S.J., Hampel, H., Pictrini, P., Alexa nder, G.E., Horwitz, B., Daley, E., Moller, H.J., Schapiro, M.B., & Rapoport, S.I. ( 1999). Region-specific corpus callosum atrophy correlates with the regional pattern of corti cal glucose metabolism in Alzheimer Disease. Archives of Neurology, 56, 467-473. Tekin, S., Fairbanks, L.A., OConnor, S., Rosenberg, S., & Cumm ings, J.L. (2001). Activities of daily living in Alzheimers disease: neuropsychiat ric, cognitive, and medical illness influences. American Journal of Geriatric Psychiatry, 9 81-86. Thompson, P.M., Hayashi, K.M., Zubicaray, G. Janke, A.L., Rose, S.E., Semple, J., Herman, D., Hong, M.S., Dittmer, S.S., D oddrell, D.M., & Toga, A.W. (2003). Dynamics of gray matter loss in Alzheimers disease. The Journal of Neuroscience, 23, 994-1005. Thompson, P.M., Mega, M.S., Woods, R.P., Z oumalan, C.I., Lindshield, C.J., Blanton, R.E., Moussai, J., Holmes, C.J., Cummings J.L., & Toga, A.W. (2001). Cortical change in Alzheimers disease detected with a disease-sp ecific population-based brain atlas. Cerebral Cortex, 11 1-16.

PAGE 108

96 Tombaugh, T.N., & McIntyre, N.J. (1992). The Mini-Mental State Examination: A comprehensive review. Journal of the American Geriatric Society, 40 922-935. Travniczek-Marterer, A., Danielczyk, W., Sima nyi, M., & Fischer, P. (1993). Ideomotor apraxia in Alzheimers disease. Acta Neurologica Scandinavica, 88 1-4. van Heugten, C.M., Dekker, J., Deelman, B.G ., van Dijk, A.J., Stehmann-Saris, J.C., & Kinebanian, A. (1998). Outcome of stra tegy training in stroke patients with apraxia: a phase II study. Clinical Rehabilitation, 12, 294-303. Vermersch, P., Roche, J., Hamon, M., Daem s-Monpeurt, C., Pruvo, J.P., Dewailly, P., & Petit, H. (1996). White matter magnetic resonance imaging hyperintensity in Alzheimers disease: correlations with corpus callosum atrophy. Journal of Neurology, 243, 231-234. Vermersch, P., Scheltens, P., Barkhof, F., Stei nling, M., & Leys, D. (1993). Evidence for atrophy of the corpus callosum in Alzheimers disease. European Neurology, 34, 83-86. Watson, R.T. & Heilman, K.M. (1983). Callosal apraxia. Brain, 106, 391-403. Weis, S., Jellinger, K., & Wenger, E.(1991). Morphometry of the corpus callosum in normal aging and Alzheimers disease. Journal of Neural Transmission,33, 35-38. Willis, L., Behrens, M., Mack, W., & Chui, H. (1998). Ideomotor apraxia in early Alzheimers disease: time and accuracy measures. Brain and Cognition, 38 220233. Wilson, B.S. (1988). Remediation of apraxia fo llowing an anaesthetic accident. In West, J. & Spinks, P. (Eds.) Case studies in Clinical Psychlogy. Bristol: John Wright. Yesavage, J.A., Brooks, J.O, Taylor, J., & Tinkelberg, J. (1993). Development of aphasia, apraxia, and agnosia and decline in Alzheimers disease. American Journal of Psychiatry, 150 742-747).

PAGE 109

97 BIOGRAPHICAL SKETCH Ann Marie Knight received her doctoral degree in speech-language pathology from the University of Florida, Department of Communication Sciences and Disorders. She received a Bachelor of Arts degree from the Un iversity of South Florida and a Master of Arts degree from the University of Florida. Dr. Knight ha s published articles in several peer reviewed publications and has presented re search findings at scientific meetings. Dr. Knight is currently participating in da ta collection and study design at the VA RR&D Brain Rehabilitation Research Center in Gaines ville, Florida, and is an instructor of neuroanatomy and adult language and motor di sorders in the University of Florida, Department of Communication Sciences and Disorders. Her primary research and clinical interests include investigation of the nature and rehabilitation of acquired neurologic deficits such as spasmodic dysphonia, limb apraxia, agrammatism, alexia, and phonological processing disorders. Additionally, Dr. Knight maintains an active national certification and state license in speech pathology as well as several state and national professional affiliations, and has held leadership positions in various graduate level university organizations.