University of Southern California Dissertations and Theses

Behavioral, muscular and dynamical changes in low force dexterous manipulation during development and aging

(USC Thesis Other)

Behavioral, muscular and dynamical changes in low force dexterous manipulation during development and aging

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content BEHAVIORAL, MUSCULAR AND DYNAMICAL CHANGES IN LOW FORCE
DEXTEROUS MANIPULATION DURING DEVELOPMENT AND AGING
by
Sudarshan Dayanidhi
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulllment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOKINESIOLOGY)
August 2012
Copyright 2012 Sudarshan Dayanidhi
Epigraph
\The sheer versatility of the hand, its essential beauty, its contribution to everything we
do and the devastation that results from loss of its control all justify the time, energy,
resources and imagination that are needed to conclude a successful outcome to this quest."
Roger N. Lemon (1999)
ii
Dedication
To my parents and my brother for their love and support
iii
Acknowledgements
A doctoral research presents one with many opportunities and diculties, most of which
are achieved and overcome not by one on their own but by help from many people. While
this dissertation is my doctoral dissertation there are a large number of people whose
presence in these PhD years have helped me achieve my personal goals. Additionally,
while this dissertation presents successful outcomes to the research objectives the real
learning has happened from the dead ends, the self doubts and the unsaid aspects of this
process.
First and foremost I am very grateful to Professor Valero-Cuevas, who while recog-
nizing that I was a nontraditional student never let me think that it would in anyway
prevent me from entering a new domain of learning. As an adviser Francisco provided me
with every resource I could ask for, always promoted my personal learning while man-
aging to keep me grounded in reality and on track to nish up my PhD. In the course
of the past 5 years I have had the opportunity to present our work in many conferences,
attend meetings, meet numerous researchers, participate in various collaborative projects
nationally as well as internationally, all of which have been an exceptional learning expe-
rience. Finally, I would like to thank Francisco, Erika, Marco and Eva for their wonderful
iv
hospitality during the many \annual" get togethers at their place as well as for the lab
ski trip.
All of my guidance and dissertation committee members were a tremendous support
and kept on challenging my thinking. Professor Winstein was a wealth of information and
always managed to nd time to meet with me as needed, in spite of having a very busy
schedule. In addition, the opportunity to be a part of the Rehabilitation Engineering
Research Center (RERC) at USC was a fabulous experience and for that I am very
thankful. Professor Sanger brought in his encyclopedic resource about developmental
aspects in children and in particular, I am very thankful to him for introducing me to
the fabulous work done by John Martin's group. In addition, I am thankful to him for
giving me the opportunity to be part of the Taskforce on Childhood Movement Disorders
in 2008 and 2009. I would also like to thank Professor Schweighofer for taking the time
to participate in my guidance committee and be a part of my qualifying exams.
Professor Hans Forssberg, although not ocially in my dissertation committee was
very instrumental in the design and development of the study in children. He was most
gracious and kind with his time and advice during my many visits to Karolinska Institutet.
Also I am thankful for his kind hospitality when I was in Stockholm. In addition, despite
his busy schedule he made time to participate in my qualifying exams and his input was
invaluable.
Professor Jason Kutch was very supportive and willing to share his AMPL facilities
and time for me to do one of my dissertation studies. It turned out to be a very interesting
study and I am thankful that I was able to do it. In addition, I would also like to thank
him for all his advice for my postdoctoral work. Jason was also a friend, who provided
v
a lot of good musical entertainment in many of the \annual" get togethers, in particular
his rendition of the music of Nick Drake.
Dr.

Asa Hedberg is an equal contributing author on the rst study and I am very
thankful to her for all her time, hospitality and kindness shown during my many visits
to Karolinska Institutet. In addition I would like to thank Dr. Elena Pavlova, Linda
Junker, Isak H agg and Novalie Lilja for their assistance. Many individuals from Rancho
Los Amigos Rehabilitation Hospital were very helpful with recruitment and assistance
in the RERC project. In particular I would like to thank Phil Requejo, Juan Garibay,
Allison Chu, Kathleen Shaneld for all their help.
Professor Jim Gordon always had something nice and supportive to say to me when-
ever I have presented in his lab meetings or at the division seminar. The Division of
Biokinesiology and Physical Therapy were very supportive with their generous support
as were the discussions with my fellow BKN and DPT students.
Kornelius R acz and Manish Kurse, were my comrades in arms throughout the PhD
process. Manish has been a good friend, a true motivator, an inspiration for camaraderie
and organizer extraordinaire. He has been there for these past 5 years through the ups and
downs of graduate education. Kornelius, likewise has been a good friend, always willing
to discuss my data and suggest ideas for improvement. His minimalist (i.e primarily
visual) approach to presentations has helped me a lot in my own presentations. Finally,
the classes I took with both Manish and Kornelius without the pressure of trying to nish
up my research were probably the most fun aspect of graduate school. I can honestly
say that it would have been very dicult to complete my PhD without their friendship,
discussions and support.
vi
There are many other friends in the lab who contributed in ways which will not be
acknowledged in research papers. I have to thank Brendan Holt for being a fellow public
transport user and for introducing me to \A Song of Ice and Fire", which helped maintain
my sanity during a tough last year (\Valar Morghulis"), Alex Reyes for being a fellow
microbrew acionado and chess wizard, Josh Inouye for his amazing acronyms (\Grazie
Mille!"), Na-Hyeon Ko for being a fellow pediatric therapist and social butter
y (\Gam-
bae!"), Nora Nelson for her multi-lingual humor and discussions about the Karolinska
project, Alison Hu for her help with the RERC project, discussions about my data and
above all for asking me some very tough questions, Jon Weisz for his coding skills and
rare kaiten sushi nights and Evangelos Theodorou for some very interesting philosophi-
cal, supportive discussions and for introducing me to some good music. Finally, all the
other unnamed people in the lab were a great support and provided an excellent work
environment.
vii
Table of Contents
Epigraph ii
Dedication iii
Acknowledgements iv
List of Tables xi
List of Figures xii
Abstract xiv
Chapter 1: Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Statement of the Problem . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Phylogenetic Considerations for Dexterous Manipulation . . . . . . 3
1.1.3 Ontogenetic Considerations for Dexterous Manipulation . . . . . . 5
1.1.4 Development of Control . . . . . . . . . . . . . . . . . . . . . . . . 8
1.1.5 Assessment of Fine Motor Abilities & Manipulation . . . . . . . . 10
1.1.5.1 International Classication of Function, Disability and
Health (ICF) . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.1.5.2 Clinical Assessment Tools . . . . . . . . . . . . . . . . . . 12
1.1.5.3 Quality Analysis of Clinical Measurements . . . . . . . . 13
1.1.5.4 ICF-High Quality . . . . . . . . . . . . . . . . . . . . . . 13
1.2 Previous Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.3 Signicance of Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.4 Dissertation outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4.1 Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4.2 Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4.3 Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.4.4 Chapter 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.4.5 Chapter 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
viii
Chapter 2: Development of Dexterous Manipulation in Childhood 24
2.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.3.1 Instrumentation for Dexterity Instrument . . . . . . . . . . . . . . 27
2.3.2 Experimental Procedure . . . . . . . . . . . . . . . . . . . . . . . . 28
2.3.3 Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3.4 Statistical Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.4.1 Improvements in Individual Spring Forces . . . . . . . . . . . . . . 32
2.4.2 Improvements in Dexterity on Dexterity Score . . . . . . . . . . . 33
2.4.3 Relationship between Strength & Dexterity . . . . . . . . . . . . . 33
2.4.4 Change in Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.4.5 Relationship between hand anthropometrics and dexterity . . . . . 37
2.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.6 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Chapter 3: Characteristics of Muscle Twitch of the First Dorsal Interosseous
(FDI) in Children and Adults during Submaximal Contractions 45
3.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.3.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.3.2 Experimental Protocol . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.3.3 Data Reduction and Analysis . . . . . . . . . . . . . . . . . . . . . 51
3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.4.1 EWA in adults and children . . . . . . . . . . . . . . . . . . . . . . 52
3.4.2 Changes in dexterous manipulation abilities . . . . . . . . . . . . . 53
3.4.3 Relationship between time-to-peak and dexterous manipulation . . 55
3.4.4 Relationship between pinch strength and time-to-peak . . . . . . . 55
3.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Chapter 4: In
uence of Surface and Visual Conditions on Dexterous Ma-
nipulation 62
4.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.3.1 Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.3.2 Experimental Procedure . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3.3 Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.4.1 Relationship between nger forces and surface and vision conditions 66
4.5 Discussion and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.6 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
ix
Chapter 5: Change in Dexterous Manipulation with Aging 76
5.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.3.1 Instrumentation for Dexterity Instrument . . . . . . . . . . . . . . 79
5.3.2 Experimental Procedure . . . . . . . . . . . . . . . . . . . . . . . . 79
5.3.3 Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.4.1 Changes in dexterous manipulation abilities with aging . . . . . . 81
5.4.2 Relationship between dexterous manipulation and pinch strength . 82
5.4.3 Change in dynamics of control with aging . . . . . . . . . . . . . . 85
5.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Chapter 6: Dynamical Analysis to Quantify Changes in Low Force Dex-
terous Manipulation across the Lifespan 90
6.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
6.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.3.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.3.2 Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.4.1 Lifespan changes in the Dexterity Score . . . . . . . . . . . . . . . 95
6.4.2 Lifespan changes in dynamics of control . . . . . . . . . . . . . . . 96
6.4.3 Detrended Fluctuation Analysis . . . . . . . . . . . . . . . . . . . . 100
6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.6 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Chapter 7: Conclusions and Future Work 104
Bibliography 106
x
List of Tables
1.1 Clinical instruments of ne motor ability in adults and children . . . . . . 14
1.2 Assessment of quality of clinical instruments for ne motor ability . . . . 15
1.3 Clinical tests for ne motor ability in the context of the ICF . . . . . . . . 17
2.1 Spring specications of the experimental setup . . . . . . . . . . . . . . . 29
4.1 Maximal hold Thumb forces under dierent conditions . . . . . . . . . . . 68
4.2 Maximal hold Index forces under dierent conditions . . . . . . . . . . . . 68
4.3 Range for Thumb forces under dierent conditions . . . . . . . . . . . . . 68
4.4 Maximal hold Index forces under dierent conditions . . . . . . . . . . . . 69
6.1 Enrollment Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
xi
List of Figures
1.1 The International Classication of Function, Disability and Health (ICF) 11
2.1 The Strength-Dexterity plane for manual skills . . . . . . . . . . . . . . . 28
2.2 The Strength-Dexterity setup demonstrating the four springs and the hard-
ware for the force data capture . . . . . . . . . . . . . . . . . . . . . . . . 29
2.3 Improvement in individual spring forces through adolescence. . . . . . . . 32
2.4 Dexterity Score throughout childhood and adolescence . . . . . . . . . . . 34
2.5 Maximum pinch strength cannot explain performance on the dexterity device 35
2.6 Changing dynamics of control . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.7 Relationship between nger lengths and dexterous manipulation abilities
for spring 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.8 Relationship between nger lengths and dexterous manipulation abilities
for spring 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.9 Relationship between hand size and nger length for spring 1 . . . . . . . 40
2.10 Relationship between hand size and nger length for spring 2 . . . . . . . 41
3.1 Experimental setup for EMG weighted average showing the surface EMG
on the First Dorsal Interosseous, the 6-axis load cell with a tube attachment
and visual feedback around 2N of force . . . . . . . . . . . . . . . . . . . . 50
3.2 The Time to Peak from the EWA waveform for all 36 subjects for all trials 53
3.3 Change in dexterous manipulation and Time-to-Peak between young adults
and children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.4 Representative examples of the EWA waveform, with identied time-to-
peak, in young adults and children . . . . . . . . . . . . . . . . . . . . . . 55
xii
3.5 Relationship between Time-to-Peak and dexterous manipulation capabili-
ties in young adults and children . . . . . . . . . . . . . . . . . . . . . . . 56
3.6 Relationship between time-to-peak and pinch strength (Tip-to-Tip and
Key Pinch) in adults and children . . . . . . . . . . . . . . . . . . . . . . 57
4.1 Dynamic precision grip setup . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.2 Sample time series from experiment . . . . . . . . . . . . . . . . . . . . . 67
4.3 Variance in maximal nger forces across dierent conditions . . . . . . . . 69
4.4 Finger forces under low friction condition . . . . . . . . . . . . . . . . . . 70
4.5 Finger forces under high friction condition . . . . . . . . . . . . . . . . . . 71
4.6 Coecient of variation of nger forces under all conditions . . . . . . . . . 72
5.1 The Strength-Dexterity setup demonstrating the spring and the hardware
for the force data capture . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2 Change in dexterous manipulation capabilities with aging. . . . . . . . . . 82
5.3 Boxplots showing the signicant change in dexterous manipulation capa-
bilities with aging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.4 Relationship between pinch strength and dexterous manipulation . . . . . 84
5.5 Dynamics of control in adults . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.1 Schematic of improvement of stability . . . . . . . . . . . . . . . . . . . . 91
6.2 The Strength-Dexterity setup demonstrating the four springs and the hard-
ware for the force data capture . . . . . . . . . . . . . . . . . . . . . . . . 93
6.3 Lifespan change in low force dexterous manipulation capabilities . . . . . 96
6.4 Symbolic regression using Eureqa . . . . . . . . . . . . . . . . . . . . . . . 97
6.5 Lifespan change in dynamics of control . . . . . . . . . . . . . . . . . . . . 99
6.6 Stability in Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.7 Stability in Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
xiii
Abstract
This dissertation focuses on the change in low force (< 3N) dexterous manipulation capa-
bilities across the lifespan. A simple device based on the mechanical properties of springs
allowed us to systematically test the control and the change of dexterous manipulation
skills across the lifespan over 240 participants from 4-89 years of age. Dexterous manipu-
lation capabilities improve dramatically during early childhood and adolescence, followed
by gradual declines from the middle age. Here we show that the timelines of development
of dexterity are much longer than previously thought and continue well into late adoles-
cence, matching known changes in neural development. In addition, these improvements
appear to be poorly predicted by changes in strength and hand anthropometrics.
Muscle twitch properties of the hand muscles, specically the time-to-peak of the
rst dorsal interosseous are shown to be slower in early adolescent children and periph-
eral changes could also be contributing to changes in dexterous manipulation. This was
discovered by applying a previously developed noninvasive method, the EMG weighted
average (EWA) used for the rst time used in children. The parameters used by sensory
modalities for dexterous manipulation appear to be dierent from those used for static
tasks. Starting in the middle ages (45-65 years) there is a decline in both the control of
ngertip force direction as well as the ability to overcome and postpone instabilities. By
xiv
observing the dynamics of the whole time series we are able to show that elderly people
and young children share certain aspects of control of the dynamics of manipulation.
xv
Chapter 1
Introduction
1.1 Background
1.1.1 Statement of the Problem
Manual dexterity can be dened as the control of ngertip force direction (Valero-Cuevas,
Smaby, Venkadesan, Peterson & Wright 2003). The thumb and index ngers are the most
independently controlled (Schieber & Santello 2004) and consequently, precision grip is
considered a very dexterous task with specic direct neural control (Muir & Lemon 1983).
The hand and the neural control associated with it has evolved over millions of years and
in primates skill acquisition of dexterity is a feature of development (Kuypers 1962) and
is a particularly protracted one (Armand 1982, Armand, Olivier, Edgley & Lemon 1997).
Typically developing children grasp objects with whole handed movements around 2-
3 months of age and show improvements in grasping to use the thumb and the index
ngers (precision grasp) by around 10 months of age (Forssberg, Eliasson, Kinoshita,
Johansson & Westling 1991). Children continue to show improvements in their precision
grasp with adult-like behaviors starting around 6-8 years with renements continuing
1
into the second decade of life (Forssberg et al. 1991) These improvements are due to
both neurological maturation as well as learning (Lawrence & Kuypers 1968a, Elias-
son, Gordon & Forssberg 1991). Neuroimaging studies indicate the growth of the cor-
ticospinal tract and diuse grasping network occurs well into early adulthood (Armand
et al. 1997, Fietzek, Heinen, Berweck, Maute, Hufschmidt, Schulte-Monting, Lucking &
Korinthenberg 2000, Lebel, Walker, Leemans, Phillips & Beaulieu 2008, Muller, Homberg
& Lenard 1991, Paus, Zijdenbos, Worsley, Collins, Blumenthal, Giedd, Rapoport &
Evans 1999) with decrements from the middle ages (Pieperho, H omke, Schneider, Habel,
Shah, Zilles & Amunts 2008). Consequently appropriate timelines matching structure and
function for manual dexterity need to be elucidated in detail.
Perinatal cerebral palsy aects around 11,000 newborn children every year (Hirtz,
Thurman, Gwinn-Hardy, Mohamed, Chaudhuri & Zalutsky 2007), with preterm infants
having a 30-fold risk of developing cerebral palsy (Pharoah, Platt & Cooke 1996). Chil-
dren with cerebral palsy have signicant motor delays and lifelong diculties in mobility
and manipulation depending on the extent of the injury (Eliasson et al. 1991). Develop-
ment of precision grip is severely impaired in these children, with 6-8 year old children
demonstrating performance comparable to that achieved by age 1 in typically developing
children (Eliasson et al. 1991). Hand rehabilitation in these children is very challenging
and requires prolonged periods to show improvements. Importantly, cerebral palsy results
in lifelong impairments that lead to poor quality of life and restricted participation in the
society.
2
1.1.2 Phylogenetic Considerations for Dexterous Manipulation
Evolutionarily some motor systems are highly conserved across species; the Corticospinal
Tract (CST) is present in all mammalian species (Nudo & Masterton 1990). Brie
y
the CST, a wide diuse connection between multiple cortical areas and subcortical areas
(Dum & Strick 1991), terminates onto the intermediate zone (dorsolateral & ventrolateral)
and onto the alpha motor neurons of the spinal cord (Lemon 2008). Majority of the CST
crosses over to form the lateral CST while a small percentage (10-15%) continues as
the ventral CST. The CST works in parallel with brainstem pathways mediated via the
corticobulbar tract (Lawrence & Kuypers 1968b).
Part of the CST has direct (monosynaptic) corticomotoneuronal connections onto
the alpha motor neuron of the spinal cord (Bernhard & Bohm 1954). This system has
been attributed to being responsible for fractionation movements in the hand (Lawrence
& Kuypers 1968b) and has a specic role in ne control via precision grip (Muir &
Lemon 1983). It has been shown that many of the monosynaptic connections project
onto the alpha motor neurons related to the intrinsic muscles of the hand (Rathelot &
Strick 2006). While the CST is not the only motor system responsible for dexterity,
evolutionarily there appear to be two primary changes, namely: a) Increase in the size
of the neocortex and the CST (Hener & Masterton 1983) b) Increase in the projections
of the monosynaptic Corticomotor (CM) system with it being very well developed in the
primates and most so in the humans (Lemon & Griths 2005). While it is dicult to nd
causality in these changes they are believed to at least partly provide the neural substrate
for dexterity in humans (Lemon 1999, Lemon 1993). Additionally there has also been an
3
increase in the size of the lateral cerebellum (Matano & Hirasaki 1997), which has a role
in development of neural representations (internal models) for skilled activities such as
dexterity (Wolpert, Miall & Kawato 1998).
These studies on elucidating the role of the dierent pathways were performed in non-
human primates (Lawrence & Kuypers 1968a, Lawrence & Kuypers 1968b, Nakajima,
Maier, Kirkwood & Lemon 2000, Dum & Strick 2002) and dierences exist between
human and other species (Courtine, Bunge, Fawcett, Grossman, Kaas, Lemon, Maier,
Martin, Nudo, Ramon-Cueto, Rouiller, Schnell, Wannier, Schwab & Edgerton 2007).
The exact function of the phylogenetically older pathways in functional movements in
humans remains unclear. It has been suggested that the C3-C4 propriospinal (PN) system
mediated through the rubrospinal and reticulospinal tracts play a main role in control of
the forelimbs in cats, but during evolution their in
uence over limb control was weakened
and taken over by the CST, particularly the CM system (Nakajima et al. 2000, Lemon
& Griths 2005). There have been a series of studies that have shown that the PN
system might in fact be present in humans, but might be under inhibitory control by the
cortex (Pierrot-Deseilligny 2002). However in people with stroke or spinal injuries this
inhibition might be aected and could be of signicant importance in neurorehabilitation
(Burke 2001). In addition, while the rubrospinal tract can in
uence the alpha motor
neuron it probably cannot take over the function of the CST after lesions (Nathan &
Smith 1982). While there is an increased importance of the CM system in the control of
the human hand, all the descending systems interact with each other in a concerted way
to control movement (Kuypers 1964).
4
1.1.3 Ontogenetic Considerations for Dexterous Manipulation
In primates including humans the protracted period of development of hand control
includes increased monosynaptic projections for the development of the CM system
(Armand, Edgley, Lemon & Olivier 1994), maturation of the CST with increased myeli-
nation, conduction velocity and increase in axon diameter (Muller et al. 1991, Paus
et al. 1999, Fietzek et al. 2000). While there are corticospinal projections at birth (Eyre,
Taylor, Villagra, Smith & Miller 2001) during development there are a larger number
of terminations in the spinal cord than that seen later in development and maturity
(Martin 2005). There is a change in a cat model of the ipsilateral and contralateral ter-
minations between an immature kitten and older mature cat with gradual elimination
and increased contralateral specicity. In fact, one could argue that part of development
is to eliminate transient terminations and increase specicity of actions by making it
predominantly under the control of one hemisphere.
A number of studies have proposed the idea that the competition between hemispheres
for control of the corticospinal projection during development is important for achiev-
ing this predominantly unilateral control of the limb (Eyre 2007, Friel & Martin 2007).
Specically, this competition is used to prune the exuberant ipsilateral projections and
to increase the density of axon termination onto the spinal cord. Martin and colleagues
(Martin, Friel, Salimi & Chakrabarty 2007) propose that this competition is activity-
based and works along with the activity- independent processes during specic critical
5
periods. It is important to note that many of the studies in development of the corti-
cospinal (CS) connections as well as on plastic changes in the motor system, specically
in the spinal cord are performed in rats and cats.
This inter-hemisphere competition hypothesis was initially tested in kittens during
postnatal CS renement period, i.e. 3-7 weeks, when the terminations on the spinal cord
are rened with elimination of the transient connections (Martin & Lee 1999). During
this period unilateral inactivation of the sensorimotor cortex (with muscimol) demon-
strated that the CS terminations from the silenced side were very limited (mainly to
the contralateral side) while the active cortex maintained immature bilateral projections.
Compared to controls the spinal gray area showed a substantial decrease in terminations
from the silenced side along with an increase from the active side. These results suggest
there is denitely a role of the activity of the neural cortex during this period on rene-
ment of the transient terminations and more importantly the silenced cortex changes the
organization of the terminations from both cortices.
In a follow-up study the role of bilateral versus unilateral inactivation was studied
to
esh out the role competition between cortices & activation plays on the organiza-
tion of the CS terminations (Martin & Lee 1999). Most interestingly they reported that
bilateral inactivation showed that the transient terminations onto the spinal cord gray
matter persisted and importantly were extensive than that seen with unilateral inacti-
vation. These results provide some evidence to the hypothesis that activity-dependent
competition between hemispheres during early development shapes the topography of the
CS terminations on to the spinal gray matter. Impairments were seen in behaviors in the
cats on the contralateral side to the inactivation (in the unilateral inactivation case) and
6
to a less extent on both sides following this period for months in spite of training (Martin,
Donarummo & Hacking 2000). More recently (Friel & Martin 2007) followed this idea
with a cross-over design where the unilateral inactive M1 became the active cortex in a
later post-natal period (week 7-11); the previously active cortex was inactivated. Their
results showed a complete recovery of previously impaired motor movements re
ecting
an importance of bilateral interactions of the CS system for achieving a balance between
the contralateral and ipsilateral CST connections.
In human studies (Eyre 2007, Eyre et al. 2001) TMS has played a role in elaborating
on activity driven plasticity of the CS projections. (Eyre et al. 2001) tested a small (n=9)
longitudinal sample along with a large (n=85) cross-sectional sample to characterize the
development of the ipsilateral and contralateral projections of the CS connections on to
the spinal cord. They show the presence of ipsilateral projections in neonates, which
are withdrawn with competitive maturity. The withdrawal of the ipsilateral projections
occurred between 3-18 months when the threshold for ipsilateral responses increased.
This is believed to be due to activity-dependent corticospinal axonal withdrawal during
development. While TMS provides us with indirect estimates of CS development, it
cannot give us precise information as can be obtained from the animal models.
In primates including humans the protracted period of development of hand con-
trol includes increased monosynaptic projections for the development of the CM system
(Armand et al. 1994), maturation of the CST with increased myelination, conduction
velocity and increase in axon diameter (Fietzek et al. 2000, Muller et al. 1991, Paus
et al. 1999). In a study exploring the eects of activity on development of the CST
(and other white matter tracts) in pianists and non-pianists (Bengtsson, Nagy, Skare,
7
Forsman, Forssberg & Ullen 2005), it was seen that there were increased white matter
cytoarchitecture particularly in the posterior part of the internal capsule with both age
of start of training and with number of hours practiced in childhood.
In children after a stroke at birth there is a signicantly increased withdrawal of
corticospinal axons from the infracted hemisphere although there might have been con-
nections present at the time of the stroke (Eyre 2007). There appears to be at least an
18-month period over which this withdrawal occurs. In children with hemiplegia there is
reorganization of the central motor pathways and a number of children have persistent
ipsilateral CS projections (Carr, Harrison, Evans & Stephens 1993). Hand impairments
seem to be worse in those children who only have ipsilateral projections than in those who
have mixed control or only contralateral (lesioned) cortical control (Holmstr om, Vollmer,
Tedro, Islam, Persson, Kits, Forssberg & Eliasson 2010, Vandermeeren, Sebire, Grandin,
Thonnard, Schlogel & De Volder 2003). Interestingly there are no monosynaptic (CM)
connections in the ipsilateral CST (Muir & Lemon 1983), development of which might
be essential for dexterity.
1.1.4 Development of Control
The role of development of anticipatory control is believed to be very important for
dexterous manipulation. During grasping it has been demonstrated in adults there is
feedforward control of force based on various parameters of the object (Johansson &
Flanagan 2009, Johansson & Westling 1988). In children during development, it appears
that after 2-3 years of age there is a change from feedback control to feedforward control
(Forssberg et al. 1991). At younger ages children are unable to preprogram the forces
8
required to pick up objects; they take longer duration in the phase before the object is
lifted, in addition to having a poor scaling between load and grip forces. This is also
seen for the use of specic parameters such as weight (Forssberg, Kinoshita, Eliasson,
Johansson, Westling & Gordon 1992), tactile conditions (Forssberg, Eliasson, Kinoshita,
Westling & Johansson 1995) and visual cues (Gordon, Forssberg, Johansson, Eliasson
& Westling 1992) which is thought be related to development of internal models for
dexterity.
Children with cerebral palsy have poor anticipatory control (Eliasson, Gordon &
Forssberg 1992) and do not seem to be able to develop appropriate cues to information
about weight but can adapt to some extent to tactile information (Eliasson, Gordon
& Forssberg 1995). However with practice in a blocked fashion with longer number
of trials they can adapt to both weight and tactile information (Eliasson, Gordon &
Forssberg 1995) indicating that they do have the ability for sensorimotor transformations
with anticipatory control. While children with typical development control ngertip forces
similar to that seen in adults by the age of eight, children with cerebral palsy require
higher amounts of practice in a predictable fashion, perhaps for use of error in learning
internal models for eventual development of anticipatory control.
In summary, it can be concluded that the development of dexterity is based on a
complex interaction of activity-independent factors and activity-dependent factors during
childhood aecting neuroanatomical structures as well as neural control.
9
1.1.5 Assessment of Fine Motor Abilities & Manipulation
1.1.5.1 International Classication of Function, Disability and Health (ICF)
There are a large number of clinical metrics for evaluation of ne motor function and
manipulation abilities as a means of planning and evaluating ecacy of intervention
(Eliasson, Forssberg, Hung & Gordon 2006, Sakzewski, Boyd & Ziviani 2007, Gilmore,
Sakzewski & Boyd 2009, van de Ven-Stevens, Munneke, Terwee, Spauwen & van der
Linde 2009, Greaves, Imms, Dodd & Krumlinde-Sundholm 2010). However evaluat-
ing them using the International Classication of Function, Disability and Health (ICF)
framework (Fig. 1.1) can provide a comprehensive and clear assessment of the utility and
applicability in daily life of these tools as well as provide a means to understand impact of
disablement (Nagi 1964), plan treatment, facilitate communication across disciplines and
recognize their value for evidence based practice (Stucki, Ewert & Cieza 2002, Jette 2006).
Historically there have been a few models of disablement focused on medical, social
and biopsychosocial aspects of disability (Jette 2006). The medical model focused on
dening the individual based on their disability for professional intervention (correction or
compensation) for the problem, while the social model aimed at dening the social factors
causing the disability as means of political and environmental change. However, neither
of these models worked for dening the impact of chronic conditions on the functional
aspects of life of the people with disability. The biopsychosocial model integrates the two
models and the Nagi (Nagi 1964) & ICF (WHO, 2001) are two commonly used frameworks
in the eld of rehabilitation.
10
Body Function (BF)
&
Body Structure (BS)
Participation (P) Activity (A)
Environmental
Factors
Personal
Factors
Contextual Factors
Health Condition
(disorder or disease)
Figure 1.1: The International Classication of Function, Disability and Health (ICF)
[modied from (Jette 2006) ]
11
The ICF framework attempts to provide a view of the impact of the health condition
on the person from a biological, personal and social perspective. The main relevant
domains in this framework are:
1. Body Functions & Structure - Body Functions (BF) are physiological functions of body
systems while Body Structure (BS) are anatomical parts of the body such as organs, limbs.
2. Activity & Participation- Activity (A) is execution of a task/action by an individual
while Participation (P) is involvement in daily life.
In addition A &P includes two important qualiers: Capacity and Performance. Capacity
is dened as the ability to perform an action or task (i.e. can do without the use of any
assistance), while performance describes what an individual does (does do) in a real world
environment (with the use of whatever assistive devices the individual uses daily, if any).
So in essence the dimensions under which clinical tools should be evaluated are BF, BS,
A&P capacity and A&P participation. Each of them provide relevant information and a
true understanding of a persons disability should integrate these domains.
1.1.5.2 Clinical Assessment Tools
The commonly used clinical tools for assessment of hand function will be evaluated for
both quality and for the dimension in the ICF that they test (Table 1.1). These do not
include tests which are based on questionnaires of participation such as the Childrens
Assessment of Participation and Enjoyment (CAPE), School Function Assessment (SFA)
etc. because these do not test and measure participation but are scored based on general
observation and information from children and caregivers [for a comprehensive review see
(Sakzewski et al. 2007)].
12
1.1.5.3 Quality Analysis of Clinical Measurements
These clinical assessment scales were evaluated for quality (1-6) on the basis of studies
showing validity (construct, criterion and content) and reliability (interrater, intrarater
and test-retest), in addition to showing specicity and applicability in children and adults
(if the test is used with both). Minimal quality was given if the studies were published in
a peer- reviewed journal and were indexed on Medline. This is summarized in Table 1.2.
This classication of quality of clinical measures is subjective and not a meta-analysis
considering levels of evidence. The scales with score 4 and higher were considered to have
high quality and are compared in the ICF framework. Of the fteen clinical measures
considered ve met this score of high quality, of which four are used in children and one
in adults.
1.1.5.4 ICF-High Quality
The ve tests were reviewed against which dimension in the ICF framework they test.
This is summarized in Table 1.3.
Overall the highest quality tests measure performance and participation in daily life
based on observation of a number of functional tasks which correlate well with the ICF
framework. However these studies do not measure specic force or dexterity decits,
13
Clinical Tools of Fine Motor Ability Target Popula-
tion
1.Range of Motion (Gajdosik & Bohannon 1987) Adults/Children
2.Pinch Strength (Mathiowetz, Weber, Volland &
Kashman 1984, Mathiowetz, Volland, Kashman &
Weber 1985, Mathiowetz, Wiemer & Federman 1986)
Adults/Children
3.Muscle Tone-Modied Ashworth Scale (Bohannon &
Smith 1987)
Adults/Children
4.Purdue Pegboard(Mathiowetz et al. 1986) Adults/Children
5.Nine Hole Test (Kellor, Frost, Silberberg, Iversen &
Cummings 1971)
Adults/Children
6.Box and Blocks (Mathiowetz et al. 1985) Adults/Children
7.(Pediatric) Jebsen-Taylor (Jebsen, Taylor, Tri-
eschmann, Trotter & Howard 1969, Taylor, Sand &
Jebsen 1973)
Adults/Children
8.Assisting Hand Assessment (AHA) (Krumlinde-
Sundholm, Holmefur, Kottorp & Eliasson 2007)
Children
9.House classication (House, Gwathmey & Fidler
1981)
Children
10.Shriners Hospital Upper Extremity Evaluation
(SHUEE) (Davids, Peace, Wagner, Gidewall, Black-
hurst & Roberson 2006)
Children
11.Manual Ability Classication (MACS) (Eliasson
et al. 2006)
Children
12.Melbourne Assessment (Johnson, Randall, Reddi-
hough, Oke, Byrt & Bach 1994)
Children
13.Quality of Upper Extremity Skills Test (QUEST)
(DeMatteo, Law, Russell, Pollock, Rosenbaum &
Walter 1993)
Adults/Children
14.Fugl Meyer (Fugl-Meyer, Jaasko, Leyman, Olsson
& Steglind 1975)
Adults
15.Strength-Dexterity (Valero-Cuevas et al. 2003,
Vollmer, Holmstr om, Forsman, Krumlinde-Sundholm,
Valero-Cuevas, Forssberg & Ull en 2010)
Adults/Children
Table 1.1: Clinical instruments of ne motor ability in adults and children
14
Clinical
Tools of
Fine Motor
Ability
Quality Validity Reliability Specicity/
Norms
Comments
1.Range of
Motion
2 - (Gajdosik
&
Bohannon
1987)
- Some reliabil-
ity
2.Pinch
Strength
2 - (Mathiowetz
et al. 1984)
(Mathiowetz
et al. 1985,
Mathiowetz
et al. 1986)
Norms in
children and
adults but not
reliability
3.Muscle
Tone (Mod-
ied Ash-
worth)
1 Poor
(Fleuren,
Voerman,
Erren-
Wolters,
Snoek,
Rietman,
Hermens &
Nene 2010)
(Fleuren
et al. 2010)
- No validity, re-
liability in chil-
dren
4.Purdue
Pegboard
2 (Tin &
Asher 1948)
(Tin &
Asher 1948,
Gallus &
Mathiowetz
2003)
(Mathiowetz
et al. 1986)
No validity
in children,
norms in14-19
year olds
5.Nine Hole
Test
3 (Smith,
Hong &
Presson
2000)
(Smith
et al. 2000)
(Oxford Grice,
Vogel, Le,
Mitchell,
Muniz &
Vollmer
2003)
norms in 5-10
year olds
6.Box and
Blocks
3 (Platz,
Pinkowski,
van Wi-
jck, Kim,
di Bella &
Johnson
2005)
(Platz et al.
2005)
(Mathiowetz
et al. 1985,
Platz
et al. 2005)
No validity
studies in
children
7.Jebsen-
Taylor
2 (Davis Sears
& Chung
2010)
- - No reported
validity, relia-
bility studies
in children
Table 1.2: Assessment of quality of clinical instruments for ne motor ability
15
8.Assisting
Hand As-
sessment
(AHA)
6 (Krumlinde-
Sundholm
et al. 2007)
(Krumlinde-
Sundholm
et al. 2007,
Holmefur,
Krumlinde-
Sundholm
& Eliasson
2007)
(Krumlinde-
Sundholm
et al. 2007)
Well tested for
bimanual con-
trol in children
with hemiple-
gia
9.House
Classication
1 - - (House
et al. 1981)
Not much
information on
reliability and
validity
10.Shriners
Hospital
Upper Ex-
tremity
Evaluation
(SHUEE)
3 (Davids
et al. 2006)
(Davids
et al. 2006)
(Davids
et al. 2006)
Only 11 chil-
dren with
hemiplegia
were tested
11.Manual
Ability Clas-
sication
(MACS)
5 (Eliasson
et al. 2006)
(Eliasson
et al. 2006)
(Eliasson
et al. 2006)
Good validity,
reliability for
a classication
system.
12.Melbourne
Assessment
5 (Johnson
et al. 1994)
(Randall,
Carlin,
Chondros &
Reddihough
2001)
- Valid for use
in children
with cerebral
palsy from
5-15 years of
age
13.Quality
of Upper
Extremity
Skills Test
(QUEST)
4 (DeMatteo
et al. 1993)
(DeMatteo
et al. 1993)
- Valid for use
in children
with cerebral
palsy from
1.5-8 years of
age
14.Fugl-
Meyer
4 (Platz et al.
2005)
(Platz et al.
2005)
- Valid for use in
adults
15.Strength-
Dexterity
Test
3 (Vollmer
et al. 2010)
(Valero-
Cuevas
et al. 2003,
Vollmer
et al. 2010)
- Valid for use
in typically de-
veloping chil-
dren 4-17 years
of age
Table 1.2, Continued
16
Test ICF BF
BS
ICF A& P
Capacity
ICF A&P
Perfor-
mance
Functions Mea-
sured
1.Fugl-Meyer * * - Quality of move-
ment
2.QUEST * * - Quality of move-
ment
3.Melbourne - * - Speed of move-
ment. Functional
tasks like drawing
4.MACS - - * Participation in
daily life. If assis-
tance is required.
Does not directly
measure capacity.
5. AHA - - * Bimanual abil-
ities in daily
life. Daily life
tasks like grasp &
release, coordina-
tion.
Table 1.3: Clinical tests for ne motor ability in the context of the ICF
which in combination with one of these tests could provide more information for rehabil-
itation planning, intervention as well as follow-up. More studies are required evaluating
the validity, reliability and specicity of measures of ICF BF BS and their correlations
with the measures of ICF A&P.
1.2 Previous Work
In a series of innovative studies the development of precision grip in children with typical
development (Eliasson, Forssberg, Ikuta, Apel, Westling & Johansson 1995, Forssberg
et al. 1991, Forssberg et al. 1995, Forssberg et al. 1992, Gordon et al. 1992) and in
children with cerebral palsy (Eliasson et al. 1991, Eliasson et al. 1992, Eliasson, Forssberg,
17
Ikuta, Apel, Westling & Johansson 1995, Forssberg, Eliasson, Redon-Zouitenn, Mercuri
& Dubowitz 1999) was elaborated using a stable grasp paradigm. Traditional methods
of evaluating dexterity have been used to look at composite aspects of motor control
and dexterity typically using tasks which use the whole upper extremity or with timed
tasks (Gilmore et al. 2009, Johnson et al. 1994, Mathiowetz et al. 1986, Poole, Burtner,
Torres, McMullen, Markham, Marcum, Anderson & Qualls 2005, Smith et al. 2000, Taylor
et al. 1973). With all these methods the common underlying factor is that they have used
stable objects while not specically looking at the control of ngertip force direction. In
the real world there is a signicant cost associated with failure (like breaking objects)
and dexterity depends on precision of task performance.
The manipulation ability of the hand is dened by the mechanical eect each of the
ngertips can produced on the object being held (Valero-Cuevas et al. 2003). For eective
manipulation the magnitude and direction of the ngertip forces are dynamically regu-
lated. Based on this rigorous denition of the building blocks of dexterous manipulation
the Strength-Dexterity paradigm was designed and developed to quantify the dynamic
interaction between ngertip forces (strength) and directional accuracy (dexterity). The
strength-dexterity paradigm is an innovative method of looking at hand function based on
the principle of buckling of compression springs. Under this paradigm strength require-
ment is dened as the force necessary to compress the spring; the dexterity requirement is
the ability to regulate the ngertip movement and consequently the direction of ngertip
force vectors to compress the slender springs without buckling.
The mechanical independence of the strength and dexterity requirements in a spring
was used to create a set of springs with dierent strength and dexterity requirement, where
18
each spring has a specic pair of requirements. In other words each spring represents a
point on the strength-dexterity plane. This plane was approximated by a kit with 87
springs with dierent combinations of stiness and slenderness. The original kit has 8
dexterity levels (A-H) with the dexterity index values ranging from 0.28 (the spring will
never buckle even with both sides free to rotate and shift) to 2.33 (the spring can buckle
even when both ends are held parallel to each other) and 14 strength levels (from 1 to 92
N). The dexterity index is dened by the mean diameter of the spring, spring constants
depending on material properties, springs free length, and the maximum distance the
spring can be compressed before reaching solid length (Valero-Cuevas et al. 2003). This
version of the SD paradigm has been found to be repeatable and sensitive in the elderly
(Valero-Cuevas et al. 2003).
The SD paradigm was the basis of a subsequent study evaluating sensorimotor capabil-
ities for dynamic manipulation in adults using a single nger (Venkadesan, Guckenheimer
& Valero-Cuevas 2007). Bifurcation theory provides methods to analyze the behavior of
very complex nonlinear dynamical systems (Venkadesan et al. 2007), such as dynamic
ngertip forces. By using bifurcation theory they were able to quantify how the limits
of dynamic ngertips forces are reached. Analyzing the nonlinear dynamical behavior of
how participants delay or prevent spring buckling provides vital information about how
neuro-muscular-skeletal interactions produce dynamic manipulation. A clear transition
to instability (bifurcation) is observed in 3-D position data past a critical compression
load. In addition, the projection of the position data close to the instability onto a hor-
izontal plane lies along a straight line (R
2
0.8), which is indicative of reduction of
dimensionality characteristic of bifurcated systems. Importantly, this work also showed
19
that compressing such springs utilizes, and is informative of, a sensorimotor integration
process but not a passive peripheral strategy. The maximal compressive force was similar
across healthy people regardless of their hand strength, and was sensitive to the integrity
of sensory signals.
Subsequent work further conrmed that the SD paradigm can be used as a means to
quantify cerebral sensorimotor processes that produce dexterous manipulation. Given the
complex interaction between the central nervous system and manipulation, a critical ques-
tion is whether the musculoskeletal properties of the periphery dominate this behavior,
or if there is signicant cortical involvement associated with controlling dexterous manip-
ulation in the SD paradigm. Three published studies (Mosier, Lau, Wang, Venkadesan
& Valero-Cuevas 2011, Talati, Valero-Cuevas & Hirsch 2005, Holmstr om, Lennartsson,
Eliasson, Flodmark, Clark, Tedro, Forssberg & Vollmer 2011) show signicant corti-
cal involvement when using the SD paradigm. Specically there are dierent networks
associated with Strength/Stable tasks and with Dexterity tasks. Importantly, this cor-
tical involvement is very sensitive to the sensory conditions and small changes in the
mechanical requirements and stability characteristics of the SD spring used. Therefore,
the SD paradigm is an eective means to bilaterally engage a wide variety of sensorimotor
networks in the brain, and is therefore, uniquely suited to assess the integrity of brain
function specic to dexterous manipulation.
In children with typical development the SD paradigm has demonstrated internal
scale validity as a means to measure improvement of ngertip dexterity with age (Vollmer
et al. 2010). In addition the correlation of the SD paradigm with gross manual dexter-
ity (Box and Blocks test) and pinch strength was examined. The SD paradigm has a
20
signicant non-strength, non-box and blocks variance, which re
ects sensorimotor inte-
gration for dexterity. In essence the SD paradigm is indicative of a latent sensorimotor
performance trait that improves with age and is dierent from, and complementary to,
strength and gross hand-arm function. Importantly this test demonstrated some evidence
for sensitivity to hand impairment in children with cerebral palsy and with renement
can be made into a clinically useful metric of ngertip force coordination.
This landmark study demonstrates the use of an unstable grasp paradigm in children
with typical development. However by using the rst version of the SD paradigm it eval-
uates a combined performance of strength and dexterity and not all of the tested springs
demonstrate instability. Secondarily while informative of development of hand function
in children, it is not necessarily informative specically of development of dexterity. Ad-
ditionally it requires children to be tested on a large number of items; upwards of at least
60 that have varying strength and dexterity requirements, for which a binary score is
computed (1 or 0 i.e. successful in compressing the spring or not). The composite score
in and of itself might or might not be informative of control of ngertip force direction.
Finally the use of that setup does not allow an understanding of the dynamical behavior
at the limits of performance, which might yield insights into directional control.
1.3 Signicance of Research
These innovative results indicate that the strength-dexterity paradigm is an ideal sys-
tem for challenging the developing neuromuscular system to be able to quantify the
development of dexterous manipulation. In particular, there is a need for appropriate
21
measurement of dexterity during development and in children with cerebral palsy. Not
only can such a system help measure impairments in children with disabilities but also
can help plan treatments specically focused on challenging the system for skill acquisi-
tion. A primary goal of this dissertation is to elaborate on the specic timelines and test
the processes involved in the improvements of dexterous manipulation during the course
of development and in aging. It is anticipated that the results of this work will signi-
cantly advance the current state of knowledge regarding acquisition of manual dexterity
in children, elderly people as well as promote methods for rehabilitation in children with
cerebral palsy and in people aging with and into a disability.
1.4 Dissertation outline
1.4.1 Chapter 2
This chapter is about the development of dexterous manipulation abilities in children
(4-16 years of age). This was done in collaboration with Karolinska Institutet, Sweden.
Professors Hans Forssberg and Valero-Cuevas have guided the project with Dr.

Asa
Hedberg and I being equal contributors to this study. Part of this work has been presented
at the 35th annual meeting of the American Society of Biomechanics in 2011 and at the
22nd Annual Meeting of the Society for Neural Control of Movement in 2012.
1.4.2 Chapter 3
This chapter elaborates on the study looking at muscle twitch properties with the use
of the EMG weighted average (EWA) in early adolescent girls. This work was done in
22
collaboration with the Applied Mathematical Physiology Lab (AMPL) at USC under the
guidance of Professors Jason Kutch and Valero-Cuevas.
1.4.3 Chapter 4
This chapter looks at the in
uence of dierent surface and vision conditions on dexterous
manipulation abilities. This pilot work was done under the guidance of Professor Valero-
Cuevas and was presented at the 19th Annual Meeting of the Society for Neural Control
of Movement in 2009.
1.4.4 Chapter 5
This chapter looks at the change in low force dexterous manipulation in adults (18-89
years of age). This was done as part of the Rehabilitation Engineering Research Center
(RERC) on Technologies for Successful Aging with Disability at USC and Rancho Los
Amigos National Rehabilitation Center under the guidance of Professor Valero-Cuevas.
Part of this work and some earlier forms have been presented at the Annual Conference
of the Rehabilitation Engineering and Assistive Technology Society of North America in
2009 and at the 35th annual meeting of the American Society of Biomechanics in 2011.
1.4.5 Chapter 6
This chapter looks at the change in dexterous manipulation capabilities across the lifespan
by combining data from Chapters 2, 3 and 5. This was done under the guidance of
Professor Valero-Cuevas and part of the work has been presented at the 35th annual
meeting of the American Society of Biomechanics in 2011.
23
Chapter 2
Development of Dexterous Manipulation in Childhood
2.1 Abstract
Neural control of dexterous manipulation is attributed to specic neuroanatomical struc-
tures whose connectivity and function are known have a prolonged period of development
into late adolescence. In contrast, functional improvements in dexterous manipulationas
measureable by current developmental and clinical milestonesshow few changes past the
age of eight because most measures of hand function saturate. We now show that an
extension of our prior work bridges this apparent discrepancy and establishes a novel
and clear link between known neuroanatomical development and dexterous manipulation
well in to late adolescence. Importantly, musculoskeletal growth and strength are poorly
correlated with these functional improvements in dexterity. These results begin to clarify
the behavioral benets of such neural maturation, enable the systematic study of specic
neuroanatomical structures, their connectivity, and plasticity. For example, neuroimag-
ing studies to disambiguate the dierential roles and contributions of maturation of the
corticospinal tract vs. the emergence of fronto-parietal and cortico-striatal-cerebellar
24
networks. Clinically, this extends the ages for which therapeutic interventions can be
considered fruitful, and provides a clinically-practical means to chart functional devel-
opment of dexterous manipulation in typically developing children, and children with
neurological conditions.
2.2 Introduction
Dynamic control of ngertip force magnitude and direction is the paramount require-
ment for manipulation of small, deformable and fragile object. Neural control of dex-
terous manipulation is attributed to a distributed network of control for small force,
precise dexterous manipulation; specically the primary sensory motor cortex, the dorsal
premotor area (PMd), the ventral premotor area (PMv) and the supplementary motor
area (SMA) (Bernhard & Bohm 1954, Ehrsson, Fagergren & Forssberg 2001, Muir &
Lemon 1983, Porter 1985, Rathelot & Strick 2006, Bonnard, Gall ea, De Graaf & Pailhous
2007, Davare, Andres, Cosnard, Thonnard & Olivier 2006, Ehrsson, Fagergren, Jonsson,
Westling, Johansson & Forssberg 2000, Gall ea, de Graaf, Bonnard & Pailhous 2005, Holm-
str om et al. 2011, Kuhtz-Buschbeck, Ehrsson & Forssberg 2001, Mosier et al. 2011), all of
which form a part of the CST. Their connectivity and function are known have a prolonged
period of development into late adolescence (Armand et al. 1997, Fietzek et al. 2000, Lebel
et al. 2008, Muller et al. 1991, Paus et al. 1999). In contrast, the functional improve-
ments in dexterous manipulation | as measured by current developmental and clinical
milestones | show few changes past the age of 8 (Deutsch & Newell 2001, Deutsch &
25
Newell 2002, Eliasson, Forssberg, Ikuta, Apel, Westling & Johansson 1995, Forssberg
et al. 1991, Forssberg et al. 1995, Forssberg et al. 1992, Gordon et al. 1992).
The Strength-Dexterity Test (SD) provides an innovative way to dynamically test
the control of dexterous manipulation (Mosier et al. 2011, Valero-Cuevas et al. 2003,
Venkadesan et al. 2007), dened as the ability to control ngertip force magnitudes and
directions (Valero-Cuevas et al. 2003). Recently, this paradigm was shown to capture
the development of a latent behavioral trait of dynamic ngertip force coordination in
typically developing children through adolescence (Vollmer et al. 2010) over a range of
voluntary force magnitudes. That study was the rst to demonstrate behavioral correlates
in dexterous manipulation to known neural maturation throughout adolescence but, by
testing over the range of voluntary force magnitudes, the latent variable we detected is
a compound of both sensorimotor processing and nger strength. Most importantly, the
change in the ability to control small forces that require a high amount of precision was
shown to be the area in manual skills that displays large switch during development (Fig.
2.1).
This study focuses on clearly establishing the relationship between the known time-
lines of neuroanatomical development and this novel means to grade behavioral improve-
ments in dexterous manipulation. We extend the previous work by (i) specically focusing
on a subset of the Strength-Dexterity Plane informative of dexterous manipulation ca-
pabilities requiring low strength; (ii) removing the enslaving eects between the index
and middle nger by only testing index-thumb precision pinch; (iii) using a brief (<15
min), clinically-practical version of the Strength-Dexterity Test; (iv) uniformly sampling
26
performance in 130 children spanning 4-16 yrs years of age; and (v) demonstrating an em-
pirical metric of the development of dexterous manipulation that has sensitivity through
adolescence, and which has also been shown to be associated with specic fronto-parietal
and cortico-striatal-cerebellar networks.
2.3 Methods
130 children (4-16 yrs, 76F/54M) participated in this study. Ethical approval was ob-
tained from an institutional review board and all subjects and parents consented to
participate in this study.
2.3.1 Instrumentation for Dexterity Instrument
Four custom-made springs (Century Springs Corp., Los Angeles, CA) that require low
force for complete compression and have the same stiness (k=0.49 lb/inch) were used
(Table 2.1). These four springs were chosen to provide higher resolution in dexterity while
maintaining a low strength requirement identied from (Vollmer et al. 2010) (Fig. 2.1).
In addition the following criteria were used: less 4 N of force (Ehrsson et al. 2001) and
length between 2-4 cms. Pilot studies conrmed that the properties of the custom-made
springs would possibly cover the dexterity range for the ages 4 to 16 years.
Two compression load cells (ELB4-10, Measurement Specialties, Hampton, VA) were
mounted at the spring endcaps (Fig. 2.2). The load cells were connected to a signal-
conditioning box, an USB-DAQ (Measurement Computing, Norton, MA) sampled the
data at 400 Hz using a custom written MATLAB program (Natick, MA) and a deadweight
calibration procedure was used for conversion from voltage to force.
27
Dexterity Requirements
Strength Requirements
Low
High
High
(1-2 N) (77-80 N)
Figure 2.1: The Strength-Dexterity plane (Valero-Cuevas et al. 2003) created to have
increasing strength requirements along the x-axis and increasing dexterity requirements
along the y-axis. The results from (Vollmer et al. 2010), color-coded for diculty, demon-
strate the large switch in diculty in performance in children when they go from item
G2 to H2, implying that a higher resolution in this transition could provide a better
gradation of changes in dexterity. The strength requirements in this column 2 is only
between 2.2 -2.7 N.
2.3.2 Experimental Procedure
The aim of the task was to keep a sustained compression, with the index nger and
the thumb of their dominant hand, for at least three seconds at the highest individual
level of control of force magnitude and force directions. After a brief familiarization with
28
Spr 1 Spr 2 Spr 3 Spr 4
Free Length (cm) 3.96 3.6 3.24 2.90
Solid Length(cm) 0.69 0.69 0.69 0.69
Force Range(N) 0-2.84 0-2.5 0-2.19 0-1.89
Table 2.1: Spring specications of the experimental setup. The four springs used had the
same stiness (k= .8581 N/cm) but dierent length and low forces were required for full
compression of the springs.
1 2 3 4
Figure 2.2: Strength-Dexterity Setup demonstrating the four springs and the hardware
for force data capture. The springs were custom made such that the spring stiness
was maintained the same (k=0.86 N/cm) across all the four springs. The lengths of the
springs varied from 2.90- 3.96 cm, while the maximum force required for compression of
the springs ranged from 2-3 N of force. Compression load cells were mounted on custom
ABS plastic endcaps with double-sided tape. The springs were presented in a sequential
order from the shortest, i.e. spring 4 and the test spring was chosen such that it was the
rst spring the subject was not able to compress fully, i.e. the minimally-impossible-to-
compress spring. The gure on the right shows an example compression and hold.
all springs and the task, the springs were presented in order, starting with the shortest
(Spring 4) to identify a spring the subject could not compress fully, i.e., Spring 3, 2 or 1.
This spring was identied as the test-spring. After test-spring identication, the subjects
were asked to rstly compress their test-spring to the point at which the device would
slip out of their hand, in order to identify their level of control. They were then asked to
compress the spring as close as possible to that point and maintain that compression at
a steady level for at least three seconds (Fig. 2.2). At least three successful compression
holds were collected per subject.
29
Pinch strength with the index nger and the thumb of the dominant hand using a
tip-to-tip pinch was measured using pinch gauge (B&L Engineering, Tustin, CA, USA).
The subjects were asked to compress the pinch meter with maximum force for a couple
of seconds and the maximal value of two attempts was used. Fine motor precision was
tested using subtest one for Fine Manual Control of the Bruininks-Oseretsky Test of
Motor Prociency, Second Edition (BOT-2). The summed raw scores of the subtest were
used to conrm typical ne motor development in the population.
For anthropometrics measures, the dominant hand was photographed in three po-
sitions; dorsal view, palmar view and radial view. This data were obtained only in
94 subjects. Anthropometric data were extracted regarding thumb length, index nger
length, hand size as well as 22 other measures.
2.3.3 Data Reduction
A custom written MATLAB
c
(Mathworks, Natick, MA) program was used to visually
identify (ginput) the sustained compression phases based on the force and force rate. To
facilitate this in the presence of high frequency dynamic changes in the force, we used
a loess smoother with a span of 10% before the force rate was computed. We dened a
sustained compression phase when the rate was bounded within 1 standard deviation of
the mean force rate. The start was identied when the rate was close to zero and the end
when the rate went out of bounds and the force dropped towards baseline. The sustained
compression phase data were downsampled to 100 Hz, low pass ltered at 25 Hz while
maintaining phase (Butterworth, ltlt). The average of the two nger force time series
were computed to create a representative force.
30
The springs were weighted such that for each subject a Dexterity score was computed
by summing the maximum force of each spring they could compress fully, plus the max-
imal force they could keep in their test-spring, normalized to that maximally possible
over the four springs. In addition dead band of force between the springs were removed.
So if a child were able to compress spring 1 completely they would get a score of 100%.
For most children in our study Spring 1 or 2 were used as the test-spring (three children
of the age of four used Spring 3). In order to characterize the dynamics of how children
control the dexterity device during the sustained compression we examined the data in
state space. We calculated the normalized sum of Euclidean distances of the trajectories
through the time series of the hold phase.
2.3.4 Statistical Analyses
The independent variables of interest were age and gender while the dependent variables
were a) the Mean of the three maximal holds, b) Maximal hold, c) Dexterity score during
maximal compression and d) Dynamics during maximal hold compression. The dexterity
score was computed as a weighted average based on the mean hold force on the impossible-
to-compress spring and normalized to the maximal possible value. In addition dead band
in forces in the high and low ranges were removed for the springs. Regression analyses were
performed to look at shared variances between pinch strength and maximum performance
in dexterity as well as between hand size and maximal performance in dexterity.
31
2.4 Results
2.4.1 Improvements in Individual Spring Forces
There is a consistent increase in the ability to overcome instability as seen with the
increased force in both spring 1 & spring 2. Statistical dierences (p<0.05) are seen in
the average of the best 3 hold phases across the ages as seen in the boxplots (Fig. 2.3).
0
100
200
300
7−9 10−11 12−13 14−16
Subject Ages
Force(grams)
Mean Force
7
20 22 19
*
Spring 1
*
0
100
200
300
4−5 6−7 8−9 10−11
Force(grams)
7 17 18
14
*
*
*
Spring 2
Figure 2.3: The median hold force children are able to generate for spring 1(Top) or
spring 2 (Bottom). The children (n=126) used either spring 1 or spring 2 as their minimal
impossible-to-compress spring. The dierence in length between the two springs is 0.36
cm. The x-axis is childrens ages while the y-axis is the force in gram force (0-300 gmf).
Signicant dierences indicated with an asterisk, are based on a 95% condence interval
around the median values for the age groups and show changes throughout childhood and
adolescence.
32
2.4.2 Improvements in Dexterity on Dexterity Score
The change in combined dexterity score is best represented by a Fourier t with a plateau
phase achieved in late adolescence (Fig. 2.4, top). Statistically signicant dierences
(p<0.05) are seen throughout childhood and adolescence (Fig. 2.4, bottom).
2.4.3 Relationship between Strength & Dexterity
The maximum strength a child can produce does not appear to be predictive of their
performance during dexterous manipulation, with low r
2
values for both spring 1 (r
2
=
0.18, Fig. 2.5, Red) & spring 2 (r
2
= 0.185, Fig. 2.5, Blue).
2.4.4 Change in Dynamics
The dynamics of control during the hold phase improves with age as seen by the reduction
in the normalized Euclidean distance in the state space (F -
_
F -

F ). Fig. 2.6 (top) shows
representative children for each of the age groups (4-6, 7-9, 10-12, 13-16). The greatest
reduction is seen in early childhood between 4-6 and 7-9, however there are also signicant
changes (p<0.05) in late adolescence (Fig. 2.6, bottom).
33
4 6 8 10 12 14 16
40
60
80
100
Dexterity Score
Change in Dexterity During Development

SSE=2926
RMSE=4.819
73.89 −10.66*cos(x*.2251)−6.029*sin(x*.2251)
Spring 1
Spring 2
Spring 3
Fourier Robust Fit
95% prediction bounds
40
60
80
100
4−5 6−7 8−9 10−11 12−13 14−16
Dexterity Score
*
*
*
Age (Years)
Figure 2.4: Dexterity Score throughout childhood and adolescence. The dexterity score
was computed as a weighted average based on the mean hold force on the impossible-to-
compress spring and normalized to the maximal possible value. In addition dead band in
forces in the high and low ranges were removed for the springs. A regression line ( 95%
prediction bounds) based on a Fourier robust t showed the best t for the Dexterity
Score with respect to age (Top). The test spring is colored for spring 1 (red), spring 2
(blue) and spring 3 (magenta). Signicant dierences indicated by asterisks, based on
95 % condence levels around the median are seen in adjacent age bins across childhood
and adolescence.
34
100 200 300
0
5
10
15
Sustained Compression Force
Maximum Pinch Strength
Hold Force Vs. Maximum Pinch Strength

Spring 1
r
2
= 0.18
Spring 2
r
2
= 0.185
Figure 2.5: Regression between maximum pinch strength and maximal hold phase for
spring 1 (red circles) and spring 2 (blue squares). Pinch strength does not appear to
be predictive of ability in dexterous manipulation with only 18% of the variance being
explained by strength.
35
0
50
100
140
4−6 7−9 10−12 13−16
Ages
Euclidean length in state space
per second
Dynamics across Ages (n=130)
28
68
25
22
[17] [36] [47] [30]
Age 16
F
.
.
F
.
.
F
.
.
Age 11
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
Age 4
˙
F
F
.
.
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
Age 9
˙
F
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
*
*
*
*
Figure 2.6: Changing dynamics of control quantied by Euclidean length in the state space. Representative children for each
age range show improvements across childhood and adolescence (Top). Median values are signicantly dierent, indicated by the
asterisk in the earlier age range (4-6 years) as well as the end age range (13-16 years) (Bottom). The highest rate of change is in
the earliest years, i.e. 4-6 years.
36
2.4.5 Relationship between hand anthropometrics and dexterity
The length of the hand, length of the thumb and the index nger in itself does not
appear to be entirely predictive of ability to perform dexterous manipulation. Pearsons
correlation of coecient between dexterity and thumb length, index nger length and
hand length for spring 1 was 0.24, 0.26 and 0.34 respectively, while for spring 2 this
was 0.48, 0.55 and 0.62 respectively. A multiple regression with interaction eect was
performed to evaluate the predictive relationship of the hand anthropometric variables on
dexterity. The adjusted coecient of determination (R
2
) for this multi-linear regression
model was 0.19 (spring 1) and 0.32 (spring 2). Fig. 2.7 and 2.8 show the regression of
the nger lengths with respect to the ability to perform dexterous manipulation. The
variance in the hand length could be explained by the other anthropometric variables
(R
2
=0.74, R
2
=0.88 ), but not entirely (Fig. 2.9, Fig. 2.10 ).
2.5 Discussion
Development of dexterous manipulation abilities shows improvements well into late ado-
lescence as seen by higher maximal hold phases, as well as normalized Euclidean dis-
tances being achieved by the children in the 14-16 year olds. We are able to provide a
link between known timelines of neuroanatomical changes and behavioral improvements
in dexterous manipulation well past the age of eight, independent of changes in strength.
These developmental changes parallel known exponential increases in the structure of the
corticospinal tract where changes are observed till the early 20s (Lebel et al. 2008, Paus
et al. 1999). Region specic changes are seen in the CST with extensive practice of
37
5
8
11
5
8
11
100
200
300
R
2
=0.19
Index Finger Length(cm)
Relationship between Finger Lengths and Dexterous Manipulation (Spr 1)
Thumb Length (cm)
Maximal Mean Hold Force(gramforce)
Figure 2.7: Multiple regression between nger lengths and dexterous manipulation for
spring 1. Individual lengths do not appear to be predictive of dexterous manipulation
ability with combined variance in nger lengths only accounting for 19% for spring 1.
38
R
2
=0.32
Figure 2.8: Multiple regression between nger lengths and dexterous manipulation for
spring 2. Individual lengths do not appear to be predictive of dexterous manipulation
ability with combined variance in nger lengths only accounting for 32% for spring 2.
39
5
8
11
5
8
11
10
15
20
R
2
=0.74
Index Finger Length(cm)
Relationship between Finger Lengths and Hand Length (Spr 1)
Thumb Length (cm)
Hand Length (cm)
Figure 2.9: Multiple regression between nger lengths and hand lengths for spring 1.
Combined variance in nger lengths account for 74% variance in hand length.
40
Figure 2.10: Multiple regression between nger lengths and hand lengths for spring 2.
Combined variance in nger lengths account for 88% variance in hand length.
41
manual skills (Bengtsson et al. 2005) and impaired ability to coordinate ngertip forces
are seen in children with congenital hemiplegia (Forssberg et al. 1999) , which is highly
correlated with CST dysgenesis (Duque, Thonnard, Vandermeeren, Sebire, Cosnard &
Olivier 2003).
Here we are able to extend the previous work (Vollmer et al. 2010) to show changes
in dexterous manipulation abilities with a high resolution for small forces that span the
age range from 4-16. In addition by using only the thumb and index nger; the two most
individuated ngers in the hand (H ager-Ross & Schieber 2000) that can be controlled
independently during manipulation (Edin, Westling & Johansson 1992) we are eliminating
any in
uence of enslavement eects over the middle nger on the task. The change in the
normalized Euclidean distance provides an innovative way to quantify the control during
dexterous manipulation. The dynamics of control seen in our study show the greatest
change in the initial years, 4-6 years similar to the changes seen in white matter density
in the internal capsule (Paus et al. 1999). Interestingly, while improvements in tactile
spatial resolution and in dexterity do not appear to be associated, maximum change in
sensory acuity is seen in the 4-6 year range (Bleyenheuft, Wilmotte & Thonnard 2010).
It is important to note that the musculoskeletal changes in hand size as well as strength
increases are not predictive of dexterous manipulation capabilities with low values for
coecient of determination. This reinforces previous work (Venkadesan et al. 2007) that
the use of such a paradigm allows one to measure changes in neuromotor performance and
processing. Additionally while it is known that the increase in hand strength is highly
correlated with changes in anthropometrics such as hand length in children from 4-16
42
years of age (H ager-Ross & R osblad 2002) it appears that the rate of development of
dexterity is relatively independent from that of strength or hand anthropometrics.
The method presented in our current work provides a simple, brief (<15 min) and
yet clinically-practical version of the Strength-Dexterity Test (Valero-Cuevas et al. 2003).
Other clinical metrics for dexterity (Mathiowetz et al. 1986, Poole et al. 2005) are timed
tasks that require sequencing, motor planning, utilizing the whole arm and do not nec-
essarily provide specic information about coordination of ngertip forces while they
might have shown changes in children past the age of 8-10. Recently it has been argued
(Steenbergen & Utley 2005) that this specic information could be relevant and infor-
mative for planning of rehabilitation in individuals with neurological injuries. Here we
propose that the prolonged period of development of dexterity in typically developing
children- longer than previously known- provides a larger window for rehabilitation in
children with neurological injuries. In children with hemiplegia there is an improvement
in hand function even in adolescent years following specic rehabilitation (Bonnier, Elias-
son & Krumlinde-Sundholm 2006, Eliasson et al. 2006, Gordon, Schneider, Chinnan &
Charles 2007, Kuhtz-Buschbeck, Sundholm, Eliasson & Forssberg 2000).
Finally while the CST development has been attributed to provide the neural sub-
strate for dexterity, recent evidence has shown a diuse cortical network for dexter-
ous manipulation (Bonnard et al. 2007, Davare et al. 2006, Ehrsson et al. 2000, Gall ea
et al. 2005, Holmstr om et al. 2011, Kuhtz-Buschbeck et al. 2001, Mosier et al. 2011).
Ehrsson et al. have demonstrated that there is a greater activity in the fronto-parietal
sensorimotor areas during the control of smaller forces than larger forces, implying that
during dynamic manipulation, there is a high modulation of control of objects requiring
43
low strength that is dierent from the control required for increases in force magnitude.
Force magnitude and direction appear to be controlled by dierent aspects of the grasping
network (Holmstr om et al. 2011) with poor modulation of the control of force direction
by the M1. The cortical representation for the control of force vectors diers from that
of grip strength and require modulation by cortical-striatal-cerebellar networks (Mosier
et al. 2011). Importantly, Mosier et al. have shown that dierent neural circuitries are
associated with grasping stable and unstable objects with a signicant involvement of the
basal ganglia modulation of the premotor and motor areas in the presence of instabilities.
All of these studies have tested adults where the sensorimotor system has matured and
in order to get a deeper understanding of neural control of dexterity, the developmental
aspects of control of unstable objects needs further exploration.
2.6 Acknowledgements
We acknowledge the assistance of Elena Pavlova with data collection, Jonathan Weisz,
Nora Nelson and Linda Junker with technical assistance during the initial stages. Veron-
ica Santos and Ruben Wong, from Arizona State University extracted the hand anthro-
pometric information from the photographs.

Asa Hedberg and Sudarshan Dayanidhi
contributed equally to this study. NSF EFRI, RERC Grant 84-133E2008-8 to FVC.
44
Chapter 3
Characteristics of Muscle Twitch of the First Dorsal
Interosseous (FDI) in Children and Adults during
Submaximal Contractions
3.1 Abstract
During childhood and adolescence improvements in dexterous manipulation abilities have
been attributed to myelination and maturation of the corticospinal pathways. It is not
known if there are any peripheral changes in the muscle structural properties and if
these could contribute to improvement in manipulation. In order to answer this we use a
previously developed non-invasive technique, the EMG weighted average (EWA) to study
the muscle twitch properties of the First Dorsal Interosseous (FDI) in early adolescent
girls and in young adults. Children appear to have a slower muscle twitch as seen by a
slower time-to-peak compared to young adults. In addition we were able to show again
that there are dierences in the manipulation abilities between children and adults. It
is not clear if the slower muscle twitch properties are related to dexterous manipulation
abilities and longer cross-sectional studies are required. However we clearly show that
45
there could be a peripheral component of the developing system which could also play a
role in improvement of low force dexterous manipulation skills.
3.2 Introduction
During the course of development the improvements seen in dexterous manipulation abil-
ities in childhood and adolescence are primarily attributed to changes in the central
nervous system; specically the corticospinal tract (CST) (Armand et al. 1997, Fiet-
zek et al. 2000, Lebel et al. 2008, Muller et al. 1991, Paus et al. 1999). The high de-
gree of modulation required for manipulation tasks can be achieved by changing the
corticospinal excitability to either the rst dorsal interosseous or the opponens pollicis
(Bonnard et al. 2007). Finer adjustments of low force dexterous manipulation also require
appropriate ring of motor units with more than 50 % of the motor unit pool of the rst
dorsal interosseous being potentially activated for just 2N force output (Milner-Brown,
Stein & Yemm 1973c, Fuglevand 2011). Recent work has shown that there are behavioral
improvements in dexterity well into late adolescent years (Dayanidhi, Hedberg, H agg,
Lilja, Forssberg & Valero-Cuevas 2011, Vollmer et al. 2010), contrary to the idea that
there are few signicant changes past the age of ten (Deutsch & Newell 2001, Deutsch
& Newell 2002, Eliasson, Forssberg, Ikuta, Apel, Westling & Johansson 1995, Forssberg
et al. 1991, Forssberg et al. 1995, Forssberg et al. 1992, Gordon et al. 1992). While
strength gains are a dominant feature of muscular changes in the pre-pubescent and
pubescent years currently it is unclear if there are any peripheral contributing factors for
the observed improvements in dexterous manipulation.
46
Muscular development is a signicant aspect of change in pre-pubescent and pubescent
children with a dramatic, three-fold increase in hand strength after the age of 10 (H ager-
Ross & R osblad 2002). The increases in muscle strength during adolescent years have
been attributed primarily to increases in physiological cross-sectional area (PCSA), mo-
ment arm length and activation level but not to changes in specic tension of the muscles
(O'Brien, Reeves, Baltzopoulos, Jones & Maganaris 2010). Direct measurements of tem-
poral properties of motor units in children have been few and far due to the invasive nature
of such studies and have focused on dierent aspects such as the ring rates of motor units
and recruitment over a wide age range in people with developmental disabilities (Rose &
McGill 2005) or have used maximal level activation (Belanger & McComas 1989) which
might not be informative of submaximal twitch characteristics. Importantly all of the
studies have focused on lower extremity muscles. The developmental aspects of muscle
twitch properties in hand muscles and their role, if any in improvements in dexterous
manipulation are not known.
Spike-Triggered Average (STA) is a commonly used invasive technique to study prop-
erties of muscles (Milner-Brown, Stein & Yemm 1973a, Thomas, Ross & Stein 1986).
First developed by (Stein, French, Mannard & Yemm 1972), while invasive it has become
the de rigeur for evaluating single motor unit properties and performed over a number
of motor units to create an average STA. Electromyography Weighted Average (EWA)
is a recent noninvasive technique developed by (Kutch, Kuo & Rymer 2010) to extract
the spatiotemporal characteristics of average motor unit ring, including the time-to-peak
(similar to contraction time). Conceptually the information about the average motor unit
characteristics obtained by this method is similar to that from the averaged STA (Kutch
47
et al. 2010). Importantly the EWA might be a noninvasive way to extract information
about muscle twitch properties such as contraction time and be used in children with
relative ease. The goal of this study was to utilize the EWA to evaluate if there are any
dierences in muscle twitch characteristics, specically the time-to-peak between children
and adults as well as to observe if there is an association between the time-to-peak and
dexterous manipulation capabilities.
3.3 Methods
Thirty six subjects; twenty three adults (28.3 2.7, 13 F, 1 Left handed ) and thirteen
children (11.78 0.44, 13 F, 2 Left handed) participated in this study. The protocol
was approved by an Institutional Review Board and informed consents and assents were
obtained from the children/their parents and from the adult subjects.
3.3.1 Experimental Setup
A 6-axis load cell (20E12, 100N, JR3,Inc., Woodland, CA) was mounted to an adjustable
base. A 10 cm, custom hollow tube was created such that the base of it could be securely
attached to the load cell, while the other end provided space to insert a ngertip (Fig.
3.1). An adjustable height platform was attached to four magnetic bases, such that it
could support the forearm of the subjects. A dowel was attached to one end of the
platform, to allow one to wrap their hand around it. A 16-bit data acquisition device
(USB-6251, National Instruments, Austin, TX) was utilized to collect data from the
load-cell and from a EMG system (Bagnoli-16, Delsys
R
, Boston, MA) (Fig. 3.1) using
48
a custom written Matlab
R
program (MathWorks, Natick, MA). The data were acquired
at 4000 Hz with a bandwidth of 20-2000Hz for the EMG signals.
Two single-axis miniature load cells ( ELB4-10, Measurement Specialties, Hampton,
VA) were mounted at the endcaps of a slender linear spring(Fig. 2.2). The load cells
were connected to a signal-conditioning box, an USB-DAQ (USB-1408FS, Measurement
Computing, Norton, MA) sampled the data at 400 Hz using a custom written Matlab
R
program (MathWorks, Natick, MA) and a 3 point deadweight calibration procedure was
used for conversion from voltage to force.
3.3.2 Experimental Protocol
The subjects were seated with their dominant arm supported by the platform in around
20

shoulder abduction, 90

elbow
exion, wrist in mid prone, index nger in a few degrees
of
exion and 0

of metacarophalengeal abduction, the remaining ngers were wrapped
around a padded dowel (Fig 3.1). Dominance was determined by using a combination
of the Edinburgh Handedness test, verbal questions of hand use for writing and daily
activities to both the child subjects and parents. The single dierential EMG electrode
was placed on the rst dorsal interosseous (FDI) diagonally to run along the length of
the muscle bers, ensuring that the skin was appropriately cleaned and did not have
any lotions. The EMG signals were tested to ensure minimal baseline noise as well as
appropriate gain to avoid clipping. The subjects were provided with a visual feedback of
the vertical force and asked to apply forces vertically with their index nger such that
they were in the target zone of 2N 0.5 and maintain it as steady as possible for 20-25
49
6-Axis Load Cell
Visual Feedback
Surface EMG (FDI)
Figure 3.1: Experimental setup for EMG weighted average showing the surface EMG
on the First Dorsal Interosseous, the 6-axis load cell with a tube attachment and visual
feedback around 2N of force.The subjects were asked to wrap their ngers around the
padded dowel, apply an isometric force of around 2N upwards against the tube with their
index nger; around the distal interphalangeal joint and maintain it as steady as possible
for around 20 seconds.
50
seconds. The target used was a low sub maximal force, less than 5% of the maximum
isometric voluntary contraction for all subjects.
To test the dexterous manipulation abilities, dened as the ability to control nger
tip force direction the subjects were asked to compress and hold the instrumented spring
device. This device was designed such that the force required for full compression was
only 2-3 N but as the spring was compressed the instability increased requiring a higher
control of nger tip force direction with higher compression, consequently making it
impossible-to-compress fully. The dominant forearms were supported in mid prone, the
subjects were asked to pick up and compress this device with their index and thumb with
no help from the other ngers as much as possible so as to maintain the compression
for at least 3 seconds. In addition the maximal tip-to-tip as well as key pinch force was
measured on the dominant hand using a pinch meter( B and L Engineering, Santa Ana,
CA). One of the children only participated in the EWA experiment and did not complete
the dexterous manipulation and strength testing components of the protocol. Her data
was included in the analysis of the EWA component of the experiment.
3.3.3 Data Reduction and Analysis
Custom written programs in MATLAB were used for reducing the EWA and dexterous
manipulation data. The steady state periods from the unltered force data were graph-
ically selected using the function ginput and the force and full wave rectied EMG data
were cross correlated. The steady state periods were at least 20-25 seconds in duration
for both the adults and children and the maximum lag used for cross correlation was
100ms. To facilitate the identication of the time-to-peak of the EWA waveform the
51
ndpeaks function was used on data smoothed with a moving window of 10 ms. All the
data were visually inspected to ensure that no erroneous identication of time-to-peak
were performed. For the dexterous manipulation the mean of the force level at which
the ngers could compress and hold the spring device was computed for each of the hold
attempts. A representative time series was created as an average of the force from the
index and thumb load cells and the mean of the maximal three force in the hold phases
were computed for each of the subjects.
The independent variable was age, while the dependent variables were time-to-peak,
mean force for dexterous manipulation and maximal compression force for tip-to-tip and
key pinch. One-way ANOVAs were performed to compare dierences between the age
groups in both mean time-to-peak and mean force of dexterous manipulation. In addition
linear regression between EWA time-to-peak and dexterous manipulation abilities were
conducted, as well as between pinch strengths and EWA time-to-peak.
3.4 Results
3.4.1 EWA in adults and children
Histograms of all the trials in all the subjects show a mean time-to-peak across all trials
for adults is 51ms while that for children is 74ms (Fig 3.2). A one-way ANOVA shows
highly signicant dierences (p<0.0001, F=19.19) between adults and children in time-
to-peak (Fig 3.3, Right) with children having a median time of 74ms and adults of 49ms;
the mean values were respectively 66.54 (12.4) and 50.37 (6.47) ignoring the outliers
52
(n=3). Representative EWA waveforms are shown in (Fig 3.4) with dierent time-to-
peaks shown across the two age groups.
20 40 60 80 100 120 140
20
40
60
Time to Peak(ms)
Trial Count
All EWA Trials for All 36 (23+13) Subjects
Adults −51ms
Children −74 ms
Figure 3.2: The Time to Peak from the EWA waveform for all 36 subjects for all trials.
All the individual trials for all adults and children are illustrated along with the mean
for both groups, with children having a longer time to peak.
3.4.2 Changes in dexterous manipulation abilities
Highly signicant dierences were observed on a one-way ANOVA between adults and
children for small force dexterous manipulation abilities (p<0.0001, F=79.71) with adults
having a mean value of 212.35 (30.75) while the children have a mean value of 125.05
(19.25)(Fig 3.3, Left).
53
100
200
300
Mean Performance (gramforce)
Dexterous Manipulation
CHILDREN ADULTS
122 gmf
200 gmf
40
80
120
Time−to−Peak (ms)
Mean Time to Peak
CHILDREN ADULTS
49 ms
74ms
*
*
Figure 3.3: Change in dexterous manipulation (Left) and Time-to-Peak (Right) between
young adults and children. Signicant dierences indicated with an asterisk, are based
on a 95 % condence interval around the median values for the age groups and show
signicant changes for both time-to-peak as well as dexterous manipulation abilities.
54
0 20 40 60 80 100
Time(milliseconds)
74.25ms
Arbitrary Units
Representative EWA Waveforms
49.5ms
Adult
Child
Figure 3.4: Representative examples of the EWA waveform, with identied time-to-peak,
in young adults (blue) and children (red).
3.4.3 Relationship between time-to-peak and dexterous manipulation
Individual linear regressions for both ages show a non-signicant association(p=0.18,
adults, p=0.1, children) between the time-to-peak and the mean hold force in dexterous
manipulation (Fig 3.5).
3.4.4 Relationship between pinch strength and time-to-peak
The time-to-peak does not appear to be predictive of pinch strengths; the linear regression
for time-to-peak and pinch strengths show a non-signicant regression line for both adults
and children for key as well as tip-to-tip pinch(Fig 3.6).
55
20 40 60 80
0
100
200
300
Time to Peak(ms)
Mean Performance (gramforce)
Relationship between Dexterous Manipulation Ability and Time to Peak

R
2
=0.09
NS (p=0.18)
R
2
=0.3
NS(p=0.1)
Linear
Children
Adults
Figure 3.5: Relationship between Time-to-Peak and dexterous manipulation capabilities
in young adults (Top) and children (Bottom). The regression lines for each of the age
groups were non-signicantly dierent from zero, indicating no association between these
two variables within each age groups.
56
0 20 40 60 80 100
5
10
15
Relationship between Time to Peak and Pinch Strength
Time to Peak (ms)
Pinch Strength (kg)
R
2
=0.02
NS (p=0.58)
R
2
=0.08
NS (p=0.21)

Key Pinch
Tip Pinch
Linear
20 40 60 80 100
0
5
10
Children
Time to Peak (ms)
Pinch Strength (kg)
R
2
=0.01
NS (p=0.82)
R
2
=0.09
NS (p=0.42)

Key Pinch
Tip Pinch
Linear
Adults
Figure 3.6: Relationship between time-to-peak and pinch strength (Tip-to-Tip and Key
Pinch) in adults(Top) and children(Bottom). The regression lines for both the key pinch
and tip-to-tip pinch are non-signicantly dierent from zero, which is indicative of no
association between maximal strength and mean time-to-peak.
57
3.5 Discussion
Children show signicant increase in their time-to-peak, which for the rst time demon-
strates that there are peripheral aspects of small force muscle twitch in hand muscles
which are dierent from adults. This in part could explain the observed improvements
in dexterous manipulation capabilities observed past the age of 10. The values we ob-
served in our adults (50.37 6.47ms) is similar to what has been reported before with
both the EWA (51.511.5ms) (Kutch et al. 2010) and the STA (5512ms) (Milner-
Brown, Stein & Yemm 1973b), but importantly included a large number of young adults
(n=22). The relationship between EWA time-to-peak and muscle strength appears to not
be associated with each other. In addition, we observe dierences in low force dexterous
manipulation capabilities between children, who are older than 10 and adults, similar to
what we have reported before (Dayanidhi et al. 2011). While the time-to-peak decreases
with development and increases are seen in the mean hold force, i.e. a measure of dexter-
ous manipulation the discontinuity in our data prevents us from performing a regression
analysis across age.
The time-to-peak has been observed to the same in the Triceps Surae in children, 11-
14 years of age and adults in both sexes (Davies, White & Young 1983), while contrary
to our observation suggests that there do not appear to be a gender based dierence in
the physiological properties of the muscle twitch. The contraction time of the dorsi
exors
has been shown to have signicant increases in value in late adolescence, but this again
is not seen for the plantar
exors (Belanger & McComas 1989). Twitch contraction times
in the extensor hallucis brevis has been reported to be similar in children from a young
58
age, although there is a consistent increase in muscle twitch tension until adolescent years
(McComas, Sica & Petito 1973). However, in other mammals in developmental studies it
has been reported that there is a decrease in the time to peak from birth till adolescent
years and then similar till adulthood (Close 1964). All of the human studies appear to
be done only in lower extremity muscles and given the adult like walking pattern by the
age of 5 (Sutherland 1997), it is unclear whether one would expect these results in case of
the hand muscles which have a higher number of monosynpatic corticospinal connections
(Rathelot & Strick 2006) as well as a prolonged period of improvement (Armand et al.
1997).
In children under low force(< 5 % MVC) visual feedback conditions no dierences
have been seen in isometric force output across ages, with improvements seen through-
out childhood with sensorimotor organization of force at higher force levels (Deutsch &
Newell 2001). Increased time-to-peak leading to dierences in twitch properties have been
attributed to improved Ca
++
uptake by the sarcotubules (Brody 1976) in adulthood. The
dierence in time-to-peak could potentially also be explained on the basis of changes in
the structure of muscle bers and a shift of percentage of slow and fast-twitch muscle
bers changing during adolescent years. In infants the time-to-peak in the soles muscle
based on stimulation shows an increase from 75 ms to 110 ms at around a year (Elder &
Kakulas 1993), when they reach adult like values. Concomitantly at this time there is a
change in percentage of type I muscle bers, which also reach adult like percentages, i.e.
79 % around the age of 18 months. However in the same study the authors also reported
on data obtained by autopsy on 20 individuals from newborn to 28 years. The percentage
of type I bers in the vastus laterals and gastrocnemius was signicantly higher in the
59
children between 3-16 years of age, however the sample sizes were small and they did
not report on contraction times of those muscles. It is speculated that there is a period
of change of ber type distribution seen not just in infancy but also in the adolescent
years. However it does not appear that there are dierences in ber type distribution in
childhood compared to adults, at least for the vastus lateralis (Bell, Macdougall, Billeter
& Howald 1980, Glenmark, Hedberg, Kaijser & Jansson 1994). Currently the research
on changes in ber type proportions during later development, if any are very sparse, un-
clear and inconclusive (Kraemer, Fry, Frykman, Conroy & Homan 1989, Martin, Dore,
Twisk, Van praagh, Hautier & Bedu 2004).
It does not appear that there is any clear relationship between contraction times and
ber type. Long contraction times have been observed in human muscles with large pro-
portion of bers rich in mitochondria; Soleus around 74 ms while Tibialis Anterior was
around 58 ms (Buchthal & Schmalbruch 1970) as well as in cat muscles (Burke & Tsairis
1974). However classication of motor units based on the basis of the physiological prop-
erties have not been successful with poor correlations seen in the rst dorsal interosseous
between contraction time and force (Milner-Brown et al. 1973a, Thomas, Bigland-Richie
& Johansson 1991). Interestingly the nding by (Stephens & Usherwood 1977) have
shown moderate correlations between contraction times and twitch tensions in the rst
dorsal interosseous. The relationship between contraction time/time-to-peak, twitch ten-
sion, and ber types is not very clear given the many dierent techniques, muscles and
contraction levels used.
60
3.6 Conclusions
Here we present for the rst time that there are changes seen in the muscle twitch prop-
erties in submaximal contractions in hand muscles in children. These changes may help
explain a part of the reason for improvements in dexterous manipulation capabilities in
children during adolescent years. In addition here we show an application of the per-
viously developed noninvasive EWA method as a simple way to look at developmental
changes in muscle twitch properties.
61
Chapter 4
In
uence of Surface and Visual Conditions on Dexterous
Manipulation
4.1 Abstract
Precision grip provides insights into corticospinal motor neuronal connections. Prior hu-
man studies of precision grip have focused primarily on isometric tasks using two ngers,
which while useful in understanding basic two-nger force interactions, are limited for un-
derstanding manipulation. In our previous work we have demonstrated the signicance
of dynamically testing a system to the edge of instability to gain meaningful information
about sensorimotor integration in the hand. In addition, we know from human and animal
studies that precision grip takes years to develop. In this paper we extend our previous
work on the dynamical control of one digit to the measurement of dynamic precision grip
of instrumented unstable hand-held objects. We mounted compression load cells on the
end-caps of a slender spring, which required 3-4 N of force for complete compression and
was unstable under compression. The methodology was tested using three trials each for
two conditions of the surface of the load cell: low friction (Te
on) and high friction (ne
62
sand paper, 120 grits) and two conditions of vision: vision allowed and vision occluded.
The results suggest there is an increase in complexity in the dynamics of the precision grip
with a decrease in friction, which may re
ect higher sensorimotor integration demands
and error correction for the low friction condition. In addition the parameters for low
force dexterous manipulation might be dierent from what has been seen in static grasp
studies. As suggested by a prior study of single digit manipulation, lower grip forces in
response to lower friction for these unstable objects appears contrary to the concept of
increased grip forces for higher safety margins for solid objects. Further studies are war-
ranted to conclusively establish these dierences between stable and unstable precision
grip.
4.2 Introduction
Precision grip studies provide an insight into dexterous manipulation capabilities specif-
ically using monosynaptic connections of the corticospinal tract (Muir & Lemon 1983,
Kuypers 1960, Kuypers, Fleming & Farinholt 1960). In human studies with grip and lift
studies it has been shown that people generate higher grip forces for the same weight in
case of slippery/smoother surfaces(Johansson & Westling 1984, Edin et al. 1992, Forss-
berg et al. 1995, Westling & Johansson 1984). For example, if you were to lift an object
which had condensation on it you would apply higher grip forces on it as a way to prevent
it from slipping out of your ngers. This increase (10-40%) above the minimal required
force to lift the object has been considered a safety margin to prevent slips or allows one
to successfully do the task by activating the Fast Adapting Type-I (FA-1) receptors in
63
the ngers. However this strategy of increased grip force will not work in case of dealing
with a fragile or deformable object or worse a slippery fragile object.
The Strength-Dexterity paradigm, based on the propensity of slender springs to buckle
under compression allows one to test the dynamic ability to regulate ngertip force di-
rection and magnitude (Valero-Cuevas et al. 2003, Venkadesan et al. 2007). Our goal was
to dierentially test the in
uence of dierent sensory modalities of surface condition and
vision on dexterous manipulation abilities. By utilizing the paradigm of manipulating at
the edge of instability (Venkadesan et al. 2007) and confounding the sensorimotor system
with changing sensory conditions we are testing the relative importance of the parameters
required for dexterous manipulation.
4.3 Methods
Seven healthy young subjects (4 M/3 F, mean age 26 years) participated in this pilot
study. Ethical approval was obtained from an institutional review board and all subjects
consented to participate in this study.
4.3.1 Instrumentation
Two 6-axis load cells (ATI Instrumentation, Apex, NC) and a slender compression spring
that required 4-5 N for full compression were used. The spring was chosen such that
it could not be compressed fully but required low force for full compression. As shown
by (Venkadesan et al. 2007) with compression the instability of the system increased
requiring a greater amount of control of ngertip force direction in order to succeed at
partially compressing and holding the spring. The load cells were mounted on to the
64
end caps of the spring with double sided tape, the weight was around 20 grams and the
length of the device was around 2.7 inches (6.75 cms). A custom written MATLAB
c
(Mathworks, Natick, MA) program was used to acquire data at 400 Hz.
4.3.2 Experimental Procedure
The subjects were seated with their dominant hand forearm resting on the table. A 2x2
factorial design [Friction(High/Low); Vision(Allowed/Occluded)] was used, with block
randomized trials collected for each subject. Sandpaper of 120 grits was used for the
High Friction condition while Te
on was used for the Low Friction condition. For the
Vision condition the subjects were either able to see their hands or wore glasses masked
with black tape to proven them from seeing anything. They were asked to pick up the
spring device only with their Thumb-Index ngers and compress it as much as they could
while ensuring that the device would not slip or buckle and to hold it while applying
that force for at least 5 seconds (Fig. 4.1). The experimenter encouraged the subjects to
compress as much as they could on the impossible-to-compress spring to try to get them
to compress and hold close to the edge of instability. Three trials in 4 conditions [High
Friction/Low Friction; Vision/Vision Occluded] were collected for each subject.
4.3.3 Data Reduction
The mean for each of the hold forces for each condition and for each nger were computed
using a custom MATLAB program. The trials with the maximal hold force for each of the
conditions were chosen to be representative for each subject. Two-way ANOVAs(Vision
65
vs. Surface conditions) was used to compare the dierences in grip force for the Thumb
and Index ngers.
6- Axis Load Cells
Surface
Figure 4.1: The experimental setup showing the two 6-axis load cells being grasped using
a precision grip. The surface of both the load cells were either a rough surface or a smooth
surface.
4.4 Results
4.4.1 Relationship between nger forces and surface and vision conditions
The mean forces of the thumb were 3.97(0.47) N for low friction and 4.15 (1.05)
N for high friction, in the vision condition and were 4.11(0.78) N for low friction and
3.88 (0.82) N for high friction. The index nger forces were 3.95(0.48) N and 4.03
(0.9) N for the low and high friction respectively with vision. With vision occluded
the forces were 4.09( 0.8) N and 3.78 (0.9) N for the low and high friction.None of
the dierences were detected as signicant (Table 4.1, 4.2). The range for all the surface
66
1000 2000 3000
0
1
2
3
4
5
Data Points (100Hz)
Force (N)
Finger Forces With A Dynamic Precision Grip

Thumb
Index
Figure 4.2: Time series of normal force from the thumb (blue) and index (red) nger load
cells. The subjects were asked to compress the spring device maximally while ensuring it
will not buckle or slip and maintain for at least 5 seconds.
67
and vision conditions for both Thumb and Index ngers are shown in Table 4.3,4.4. The
variance seen was highest under the high friction and vision condition while it was the
lowest for low friction and vision (Fig. 4.3).
Thumb Force High Friction Low Friction
Vision (mean sd) 4.16 1.05 3.97 0.47
No Vision (mean sd) 3.88 0.80 4.11 0.78
Table 4.1: Maximal hold forces for the Thumb under the high and low friction and vision
and no vision conditions.
Index Force High Friction Low Friction
Vision (mean sd) 4.03 0.93 3.95 0.48
No Vision (mean sd) 3.78 0.82 4.11 0.80
Table 4.2: Maximal hold forces for the Index nger under the high and low friction and
vision and no vision conditions.
Thumb Range (N) High Friction Low Friction
Vision 3.28 1.45
No Vision 2 2.14
Table 4.3: Range for maximal hold forces for the Thumb nger under the high and low
friction and vision and no vision conditions.
4.5 Discussion and Conclusions
During low force dexterous manipulation under dierent conditions of vision and surfaces
no signicant dierences were observed in between any of the conditions. The results
68
Index Range High Friction Low Friction
Vision 2.85 1.53
No Vision 2.08 2.22
Table 4.4: Range for maximal hold forces for the Index nger under the high and low
friction and vision and no vision conditions.
High
Vision
High
No Vision
0.4
0.8
1.2
Variance
Variance for all Conditions

Thumb
Index
Low
Vision
Low
No Vision
Figure 4.3: Variance across subjects in thumb (blue) and index (red) forces for low and
high friction surface condition for all subjects under vision and no vision condition. Note
the large change in variance from the High friction- Vision condition to the Low friction-
Vision condition.
69
Low Friction
2
4
6 Thumb Forces
Force (N)
Vision Vision
Occluded
Index Forces
2
4
6
Vision Vision
Occluded
Figure 4.4: Thumb and Index nger forces for low friction surface condition for all subjects
under vision and no vision condition. Note that one of the subjects was identied as an
outlier in the vision condition for both the thumb and index ngers.
70
2
4
6
Vision
Vision
Occluded
Index Forces
2
4
6
Thumb Forces
Force (N)
Vision
Vision
Occluded
High Friction
Figure 4.5: Thumb and Index nger forces for high friction surface condition for all
subjects under vision and no vision condition.
71
2
4
6
8
Hi
Visn
Hi
No Visn
Lo
Visn
Lo
No Visn
Index
Coefficient of Variation
Coefficient of variation for all Conditions
2
4
6
8 Thumb
Hi
Visn
Hi
No Visn
Lo
Visn
Lo
No Visn
Figure 4.6: The coecient of variation (Coev) for the thumb and index nger across low
and high friction surface condition and under vision and no vision condition.
72
for the vision were similar to what has been reported before in dynamic manipulation
(Venkadesan et al. 2007) that the role of vision becomes important only in the absence
of tactile sensibility. The lack of dierences in case of handling objects with low friction
versus those with high friction are in contrast to what has been reported previously with
higher grip forces seen in case of slippery surfaces (Johansson & Westling 1984, Edin
et al. 1992, Forssberg et al. 1995, Westling & Johansson 1984). While the role of safety
margin in anticipatory control with static objects is well known (Cole, Rotella & Harper
1999, Johansson & Flanagan 2009) it is not clear if the same parameters are used for
dexterous manipulation. One factor which is very importantly dierent from the static
grip and lift tasks was the weight of the object, the weight used in many of those studies
is 100 grams or greater (Johansson & Flanagan 2009) while our device was less than 20
grams in weight. This suggests that for low force dexterous manipulation weight might not
be an important parameter for internal models (Wolpert, Diedrichsen & Flanagan 2011)
of manipulation given that many of the objects being manipulated, such as keys weigh
less than 20 grams and are denitely lighter than 100 grams.
The lowest variance is seen across subjects for the Low Friction-Vision condition (Fig.
4.3) and highest for High Friction- Vision condition. The maximal hold forces for the
thumb ranged from 3.28 to 1.45 N (Table 4.3) while that for the index was 2.85 to 1.53
N (Table 4.4). This can be seen in the whiskers of box plots for Low Friction (Fig. 4.4)
versus that for High Friction (Fig. 4.5). These results could be interpreted to mean that
when there is a high amount of safety, i.e. there is no requirement for the system to
perform optimally. On the other hand under conditions of Low Friction-Vision one of
73
the sensory modalities is challenged leading to a convergence on an optimal performance
across subjects.
The most interesting result we were able to observe was that the metric of performance
in our experiment was not the mean grip force a person could generate while holding
the spring maximally compressed (i.e. compressed partially but to the best of their
abilities) rather was the Coecient of Variation (Coev). The Coev for the most dicult
of the conditions, Low Friction-No Vision was the highest implying that this was the
most challenging of the 4 conditions for the subjects to perform. While this was not
signicant (p=0.10) it is still an interesting nding that illustrates that this seemingly
simple task can be challenging to the sensorimotor system. In our healthy young adults
they were perhaps able to modify their motor unit recruitment, re
ected by their decrease
in steadiness (Jesunathadas, Marmon, Gibb & Enoka 2010) under the most testing of the
conditions which shows some adaptability as is expected. However this might not hold
true in clinical populations and this can be modied to create a challenging task to test
for dierential contribution of sensory modalities in dierent clinical conditions.
This study showed some promising results, which while not signicant reveal two
important things; the most challenging of the condition, i.e. low friction and no vision
changes the task requirement and solely challenging the tactile system seemed to narrow
the task variance. Our pilot project requires to be expanded to see if there are signicant
dierences under the dierent friction and vision conditions and particularly if Coecient
of Variation would be a good metric to evaluate the change. The importance of this work
would be to see if the parameters for static manipulation tasks such as surface conditions,
weight remain the same in case of low force dexterous manipulation.
74
4.6 Acknowledgements
We are thankful to Kornelius R acz for his technical assistance with data acquisition.
75
Chapter 5
Change in Dexterous Manipulation with Aging
5.1 Abstract
The control of ngertip force direction appears to change with aging and is presumed
to cause impairments in manipulation abilities. By using a task which requires one to
control the ngertip force vector near normal in order to succeed in the task and by
testing people from the young through the old age (range 18-89 years) we were able to
systematically study the eect of aging on it. We extend our previous work to show a
simple, instrumented hand held device that requires low force but a high control of nger
tip force direction can be used without any ceiling eects to eectively show changes across
the whole of adulthood. Importantly the decline in the ability to overcome instabilities
with improved control of force direction appears to decline starting in the middle ages
and is dissociated from the changes in strength.
76
5.2 Introduction
Dramatic decline in strength is seen in elderly people (Lexell, Henriksson-Lars en, Winblad
& Sj ostr om 1983) and while balance and falls are a major health concern in older age
(Close, Ellis, Hooper, Glucksman, Jackson & Swift 1999) appropriate daily interactions
require ne motor abilities. With aging a decline in dexterity has been demonstrated on
pick and place tasks, as well as static tasks (Cole et al. 1999, Desrosiers, Hbert, Bravo
& Rochette 1999, Parikh & Cole 2012, Ranganathan, Siemionow, Sahgal & Yue 2001)
with a multitude of changes after the age of 65 (Carmeli, Patish & Coleman 2003).
Importantly, older people appear to have an incorrect directional bias and when asked to
generate normal forces are unable to control the direction of the ngertip force vector to
be perfectly normal to the surface during isometric tasks (Cole 2006). Consequently this
could lead to inappropriate external moment generation (Parikh & Cole 2012) leading to
functional diculties handling small objects. A perfect experiment to test if indeed there
is a directional bias in ngertip force direction with aging requires that the subjects be
constrained to have to produce a perfectly normal force direction in order to succeed at
the task. The Strength-Dexterity paradigm, based on the propensity of slender springs
to buckle under compression denes dexterity as the dynamic ability to regulate ngertip
force direction (Valero-Cuevas et al. 2003) and provides a clear scientic framework in
which to study if aging causes a directional bias in ngertip force.
During the transition from early adulthood to middle age there appears to be changes
in precision of ngertip force (Cole et al. 1999, Lindberg, Ody, Feydy & Maier 2009)
77
especially at low forces but not in maximal force generation (Lindberg et al. 2009). How-
ever this decrement has not been observed consistently (Lowe 2001) and currently it is
not clear if there is a change in ngertip force direction during a dynamic task and if
the directional bias observed in the elderly has an earlier onset. Given that with aging
there is a loss of alpha motor neurons, reinnervation leading to an increased number of
muscle bers per motor unit (Lexell et al. 1983, Brown, Strong & Snow 1988) and that
there is a high percentage of intrinsic hand muscle motor units recruited for low forces
(Milner-Brown et al. 1973b, Fuglevand 2011), leading to loss of renement of recruitment
we hypothesize that past the age of 65 adults will be worse in their low force dexterous
manipulation abilities, specically their ability to control their ngertip force direction,
but these would start from the middle age and these will be independent of changes in
strength. The goal of this study was to evaluate how the ability of individuals to con-
trol their ngertip force direction during low force manipulation changes with age and to
explore its association with strength.
5.3 Methods
32 young adults (18-34 years) and 66 adults (45-89 years) participated in this study.
They were divided into three age distributions with around 33 subjects in each group;
young, 18-34 (28.38 3.69), middle aged, 45-65 (57.06 6.58) and older, 66-89 (75.6
7.15). Ethical approval was obtained from an institutional review board and all subjects
consented to participate in this study.
78
5.3.1 Instrumentation for Dexterity Instrument
A linear spring with stiness, k=0.49 lb/inch that requires low force for complete com-
pression were used (Table 2.1)(# 4268, Century Springs Corp., Los Angeles, CA) . This
was the longest spring, part of the spring setup used in children in a complementary
study. The spring was chosen to provide higher resolution in dexterity while maintaining
a low strength requirement (Fig. 5.1). In addition the following criteria were used: less
4 N of force (Ehrsson et al. 2001) and length between 2-4 cms.
Two compression load cells (ELB4-10, Measurement Specialties, Hampton, VA) were
mounted at the spring endcaps (Fig. 2.2). The load cells were connected to a signal-
conditioning box, an USB-DAQ (Measurement Computing, Norton, MA) sampled the
data at 400 Hz using a custom written MATLAB program (Natick, MA) and a deadweight
calibration procedure was used for conversion from voltage to force.
5.3.2 Experimental Procedure
The aim of the task was to maintain a sustained compression, with the index nger and
the thumb of their dominant hand, for at least three seconds at the highest individual
level of control of force magnitude and force directions. After a brief familiarization
with the spring and the task the subjects were asked to compress the spring as close as
possible to the point beyond which the device will slip from their hand and maintain
that compression at a steady level for at least three seconds (Fig. 5.1). The subjects
were encouraged to try to push their limits to ensure they were being tested to their true
abilities. At least three successful compression holds were collected per subject.
79
1
Figure 5.1: Strength-Dexterity setup demonstrating the longest spring and the hardware
for force data capture. The spring had a stiness, k=0.86 N/cm, the length of the spring
was 3.96 cm, while the maximum force required for compression of the spring was around
3 N of force. Compression load cells were mounted on custom ABS plastic endcaps with
double-sided tape. The spring was chosen such that while the force required to compress
it was minimal, with higher compression the instabilities increased and the subject was
not able to compress it fully, i.e. it was a impossible-to-compress spring. The gure of
the right shows an example compression and hold.
Pinch strength with the index nger and the thumb of the dominant hand using a
tip-to-tip pinch was measured using pinch gauge (B&L Engineering, Tustin, CA, USA).
The subjects were asked to compress the pinch meter with maximum force for a couple
of seconds and the maximal value of two attempts was used.
5.3.3 Data Reduction
A custom written MATLAB
c
(Mathworks, Natick, MA) program was used to visually
identify (ginput) the sustained compression phases based on the force and force rate. To
facilitate this in the presence of high frequency dynamic changes in the force, we used
a loess smoother with a span of 10% before the force rate was computed. We dened a
sustained compression phase when the rate was bounded within 1 standard deviation of
80
the mean force rate. The start was identied when the rate was close to zero and the end
when the rate went out of bounds and the force dropped towards baseline. The sustained
compression phase data were downsampled to 100 Hz, low pass ltered at 25 Hz while
maintaining phase (Butterworth, 2
nd
order, ltlt). A time series was computed as the
average of the two nger force time series to create a representative force. A mean of the
the highest three hold phases was computed for each subject.
In order to characterize the dynamics of how children control the dexterity device
during the sustained compression we examined the data in the state space i.e. (

F -
_
F -F ).
We calculated the sum of Euclidean distances (norm) of the trajectories through the time
series of the maximal hold phase per subject normalized to the time held.
The independent variables were age distributions (18-34, 45-65, 66-89), while the
dependent variables were mean hold phase force, maximum pinch strength and state
space Euclidean norms/second.
5.4 Results
5.4.1 Changes in dexterous manipulation abilities with aging
Signicant dierences (p<0.0001, F=17.99) were observed on a one-way ANOVA across
the three age groups (18-34, 45-65, 66-89)(Fig. 5.2, Fig. 5.3). Schee's test for multi-
ple comparisons was used post-hoc to test for pair wise signicance and demonstrated
signicant dierences between all three groups. There was a reduction in the mean hold
phase force across the three age groups [mean(SEM)]; 213.79( 6.8), 185.73(6.6), and
81
156.15 6.8, for the 18-34, 45-65 and 66-89 respectively. Prior to statistical analysis the
groups were tested for normality using the Kolmogorov-Smirnov test.
10 30 50 70 90
100
200
300
Change in Adults low force dexterous manipulation abilities(n=98)

Age (Years)
Mean Force (gmf)
Young Adults
Middle Age
Older Adults
Figure 5.2: The mean hold force for the 98 subjects from 18-89 years is shown here.There
is a decline in low force dexterous manipulation capabilities between young adults (ma-
genta/left) and older adults (red/right).
5.4.2 Relationship between dexterous manipulation and pinch strength
There does not appear to be any association between maximal pinch strength and dex-
terous manipulation capabilities (Fig. 5.4). The regression line between maximal hold
force and maximal tip-to-tip pinch strength is not signicantly dierent from the null
hypothesis that the slope of the line is equal to zero (p=0.07, R
2
=0.05). In other words,
pinch strength is not predictive of low force dexterous manipulation capabilities. Two
82
100
200
300
18−34 45−65 66−88
Age Bins (Years)
205
Mean Force (gmf)
Change in low force dexterous manipulation abilities with aging
188
147
*
*
Figure 5.3: There is a signicant decline in low force dexterous manipulation capabilities
between young adults (left) and older adults (right), between middle age group (middle)
and older adults (right), and also between young and middle age adults. Signicant
dierences indicated with an asterisk, are based on a one-way ANOVA with post hoc
multiple comparisons.
83
subjects were detected to be outliers in their strength (Fig. 5.4, inset) and were not
included for the analysis. Including them while making the regression line signicantly
dierent from a zero slope, still maintains the main result that only a small percentage
of variance in dexterous manipulation could be explained by changes in strength.
0.1 0.2 0.3
0
5
10
15
Max Hold Force (Kg)
Maximum Tip−to−Tip Pinch Strength (kg)
Relationship between dexterous manipulation and pinch strength
R
2
= 0.05
NS (p=0.07)
5
10
Tip−to−Tip Pinch Strength
Figure 5.4: Regression between maximum tip-to-tip pinch strength and maximal hold
phase. Pinch strength does not appear to be predictive of ability in dexterous manipu-
lation with only 5% of the variance being explained by strength. Note that two subjects
were detected to be outliers for strength (gray), the boxplot in the inset shows the data
for the tip-to-tip pinch strength and the two outliers.
84
5.4.3 Change in dynamics of control with aging
There are signicant dierences in the dynamics of control, measured by the Euclidean
distance/second seen on a one-way ANOVA (p<0.001, F=7.99). Post-hoc multiple com-
parison test (Schee's) revealed the dierences were only present between the younger (18-
34, 23.933.5) and middle (45-65, 34.883.4) age group as well as between the younger
(18-34, 23.933.5) and the older (66-89, 40.463.5) age group (Fig. 5.5) but not between
the middle and older age groups. Prior to statistical analysis the groups were tested for
normality using the Kolmogorov-Smirnov test and the outliers as seen in Fig. 5.5 were
removed.
0
50
100
140
Ages
Euclidean length in state space
per second
Dynamics across Adults (n=98)
[32] [34] [32]
0.5
.
.
0.5
Age 59
*
*
18−34 45−65 66−88
33 31
21
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
F
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
F
Age 84
.
.
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
F
Age 29
.
.
Figure 5.5: Changing dynamics of control quantied by Euclidean length in the state
space. Representative adults for each age range show some changes with aging (Top).
Mean values are signicantly dierent, indicated by the asterisks, in the young adults
(18-34 years) compared to the older adults in both groups (45-65 & 66-89) (Bottom).
85
5.5 Discussion
With aging there is a consistent decline in the ability to control nger tip force direction
during low force dexterous manipulation, with higher decline seen after the age of 65. In
addition, there are some early changes in the ability to stabilize an unstable object with
dierences seen even in the middle aged group, specically how they are able to maintain
stability in the state space. Finally we were able to show that the dexterous manipulation
capabilities were not related to pinch strength and suggesting while loss in strength is a
dramatic change in old age, change in ne motor abilities are not explained by the loss
of strength. Importantly, even for such low forces we are able to show signicant changes
across adulthood without saturation eects.
Our results support the landmark study by (Cole 2006) which showed that there is a
directional bias in older people in addition to supporting the early changes in precision
from middle age (Lindberg et al. 2009). The two limitations of the (Cole 2006) study
were only inclusion of people past the age of 75 and importantly not requiring a task
constraint that the force vector direction be constrained to be normal to the surface
against which force is being applied. In our setup by design the subjects were required to
maintain relatively normal ngertip forces in order to succeed at the task. The metric of
performance, the compression force re
ects the ability to dynamically control the ngertip
force direction with higher forces requiring higher amounts of control of ngertip force
direction. Additionally here we include people from 45-89 and show that this ability to
dynamically regulate ngertip force direction declines signicantly much earlier than 75
years of age.
86
There are a number of changes in the muscular system which can account for the
observed changes. There is minimal change in proportion of type 1 ber % seen beyond
the age of 60, but strength loss is more related to marked reduction in number of bers
and reduction in ber size of type 2 bers with an almost 25 % reduction in relative area
occupied by type 2 bers (Lexell et al. 1983, Lexell, Taylor & Sj ostr om 1988, Lexell &
Downham 1992). Tip-to-tip pinch strength has been observed to be relatively stable till
around 60 years of age, after which there is a gradual decline (Mathiowetz et al. 1985),
this again reinforces the observation that the early change in dexterous manipulation
capabilities observed in our study are probably not related to strength. Our results of
poor predictive abilities of strength observed across the adult ages of 18-89 has also been
observed in studies in development in children (Dayanidhi et al. 2011) as well as in adults
using one nger manipulation (Venkadesan et al. 2007).
Even low force activation of around 2 N, such as that used for manipulation can recruit
around 50% of the motor units of the rst dorsal interosseous (Milner-Brown et al. 1973b),
which has been suggested to be an added level of control (Fuglevand 2011). The muscle
twitch time to peak using spike-triggered average force of the rst dorsal interosseous
(FDI) in the elderly people has been reported to be similar to that observed in young
adults (Galganski, Fuglevand & Enoka 1993). With aging there is a loss of alpha motor
neurons and consequent reinnervation leading to a increased number of muscle bers
per motor unit (Lexell et al. 1983, Brown et al. 1988). Steadiness in maintaining force
even at low values has been observed to be worse in older adults (Enoka, Christou,
Hunter, Kornatz, Semmler, Taylor & Tracy 2003, Galganski et al. 1993, Jesunathadas
et al. 2010) and partly contributes to changes in functional pick and place tests such
87
as purdue pegboard test (Marmon, Pascoe, Schwartz & Enoka 2011, Marmon, Gould &
Enoka 2011). The reduction in steadiness could potentially be a reason for decline in the
mean hold force seen in the older people.
From early adulthood there is some degenerative changes in the brain volume, mainly
in cortical areas of the sensorimotor system (Pieperho et al. 2008), and it appears in
particular to have an impact on the diuse cortical circuits which have been shown to be
associated with precision grasp behavior (Bonnard et al. 2007, Ehrsson et al. 2001, Holm-
str om et al. 2011, Mosier et al. 2011). This change at a cortical level might explain some of
the changes seen in our study in middle aged adults, in particular the ability to maintain
the dexterous manipulation device stably in the state space might re
ect this decrement
in control of the precision dynamics. In addition slowing of central and peripheral (motor
and sensory) conduction velocities have been observed in people older than 60 years of
age (Dorfman & Bosley 1979), but the signicance of this in the context of dexterous
manipulation is not known. Finally while tactile impairments are seen with aging just
degradation of tactile sensibility does not appear to be able to explain dierences in ma-
nipulation abilities (Cole, Rotella & Harper 1998) but obviously tactile sensibility and
vision do play a role (Johansson & Flanagan 2009) with vision compensating to some
extent for lack of tactile sensibility (Venkadesan et al. 2007). With aging in the elderly
there appears to be a complex interaction between cortical, spinal and motor unit changes
as well as peripheral changes in tactile sensibility and muscle strength all of which have
an impact on their ability to control their ngertip force direction and perform dexterous
manipulation tasks.
88
5.6 Conclusion
Control of nger tip force direction show deterioration with age, even at low forces (< 3
N). Here we present a proof of principle in ninety eight healthy adults from 18-89 years of
age and demonstrate the capacity of a simple device to capture the changing capabilities
in low force dexterous manipulation with aging. Importantly, we are able to show there
are signicant changes in this capabilities from the middle age onwards. In addition,
we are able to show that the pinch strength is not predictive of dexterous manipulation
abilities.
Acknowledgments
We acknowledge the assistance of Kathleen Shaneld, Allison Chu, Juan Garibay, Wenhsin
Hu, Emily Lawrence, Na-hyeon Ko, Isak H agg, Novalie Lilja and Phil Requejo with data
collection and recruitment, Nora Nelson and Jon Weisz with technical assistance. NSF
EFRI 0836042, NIDRR RERC 84-133E2008-8 to FVC.
89
Chapter 6
Dynamical Analysis to Quantify Changes in Low Force
Dexterous Manipulation across the Lifespan
6.1 Abstract
In order to understand the changes in low force dexterous manipulation capabilities across
the lifespan we combine the data from 3 of the previous studies to look at the changes in
240 individuals (142 children, 32 young adults, 66 older adults) ranging from 4-89 years of
age. There are improvements in dexterity seen throughout the childhood and adolescent
years and appears to peak only in early adulthood. There are changes in the ability to
dynamically regulate and maintain the ngertip forces during dexterous manipulation,
most of which are seen in early childhood but there are some late adolescent changes
as well. During adulthood there is a gradual decline in both the dynamic abilities to
overcome and postpone instabilities and in the stability with which this is achieved.
The use of simple dynamical state space plots and nonlinear analysis help reveal a rich
dynamical source of change in dexterous manipulation across the lifespan.
90
6.2 Introduction
Conceptually the analysis of dynamic behaviors can yield a wealth of information on the
development of control. (Newell, Liu & Mayer-Kress 2001) among others have proposed
how dynamical systems analysis can provide a framework to explore the stability of a
developing system as well as the role of bifurcations in changing dynamics over large
timescales can help explain dierent patterns of skill acquisition. A conceptual schematic
of relationship between age and stability is shown in Fig. 6.1, where over the course of
years the stability of particular behaviors is reinforced and is more robust to perturba-
tions.
Young Child
STABILITY
Adult
Perturbation
Perturbation
Perturbation
Control Control Control
Figure 6.1: Conceptual schematic of development of stability through changing attractor
landscape. The perturbation is trying to change the current state of the system while
the control is trying to overcome the perturbation and maintain the current state of the
system. The spring device in our case is considered to be a perturbation challenging the
control while the children/subjects are trying to stably maintain a hold force. Stable
behaviors are more robust to perturbations.
In this study the spring is considered to be a perturbation challenging the control while
the children are trying to stably maintain a hold force. Since the use precision grip and
91
dexterity is a behavior that evolves through a complex interaction of neuromaturation as
well experience dependent processes, it is expected to re
ect stability with improvements
in dexterous manipulation. In particular, it is important to also consider the need for
challenging the developing system in children with cerebral palsy eectively so as to
promote the appropriate development of functionally meaningful stable behaviors. The
real problem is to identify both the appropriate challenge level for skill acquisition (Plautz,
Milliken & Nudo 2000) as well the specic behavior to train in rehabilitation.
The dynamics of the processes involved in overcoming instabilities can reveal the
underlying change in stability. Anomalous diusion laws relate mean square displacement
and time intervals,
< x
2
>/ t
2H
This can be expressed as,
< (x
t
x
t
0
)
2
>/ (tt
0
)
2H
,
where H is the Hurst Exponent, 0 H 1. A value of 0.5 would be a random walk,
which suggests the time series has no autocorrelation or memory processes in it, while a
value between 0-0.5 is negatively correlated (i.e. diusion is suppressed, anti-persistence),
whereas a value between 0.5-1 suggests positive correlation (i.e. diusion is enhanced,
persistence) (Kantz & Schreiber 2004). Diusion analyses have been used in biomechanics
to look at postural adjustments during quiet standing such as the random walk analysis
(Collins & De Luca 1993, Collins & De Luca 1994, Collins & De Luca 1995). Rather than
the use of the simple Hurst exponent it has been shown that the use of the Detrended
Fluctuation Analysis (DFA) to calculate it is superior to the computation as described
92
above (Kantz & Schreiber 2004, Peng, Havlin, Stanley & Goldberger 1995), especially
in the presence of nonstationarities and biological noise. The goal of this study was to
summarize the changes in low force dexterous manipulation capabilities during develop-
ment through adolescence as well as aging in the middle and later years. In particular
we wanted to characterize the changes seen in the state space of the task and utilize
nonlinear methods of diusion analysis to understand the changes.
6.3 Methods
240 subjects (130 children, 4-16 yrs of age, 32 young adults, 18-34 years of age, 66 older
adults, 45-89 years of age) participated in this study (see Chapters 2, 3 and 5). They
represented the entire lifespan from 4-89 years of age (Table 6.1).
1 2 3 4
Figure 6.2: Strength-Dexterity Setup demonstrating the four springs and the hardware
for force data capture. The springs were custom made such that the spring stiness
was maintained the same (k=0.86 N/cm) across all the four springs. The lengths of the
springs varied from 2.90- 3.96 cm, while the maximum force required for compression of
the springs ranged from 2-3 N of force. Compression load cells were mounted on custom
ABS plastic endcaps with double-sided tape. The springs were presented in a sequential
order from the shortest, i.e. spring 4 and the test spring was chosen such that it was the
rst spring the subject was not able to compress fully, i.e. the minimally-impossible-to-
compress spring. The gure on the right shows an example compression and hold.
93
Age by decade (years) Number of Subjects
0-10 66
11-20 78
21-30 22
31-40 10
41-50 5
51-60 16
61-70 24
71-80 10
81-90 11
Table 6.1: Table showing the subjects enrolled in all the studies by decades.
6.3.1 Experimental Setup
Brie
y, low force dexterous manipulation capabilities were tested by using a slender
spring(s) with two compression load cells attached to the end caps (Fig 6.2). The sub-
jects were tested on their minimally impossible-to-compress spring, i.e. when the springs
are presented in the order of lengths the rst spring which the subjects are unable to
compress fully. All the subjects were able to compress spring 4, the shortest spring and
except for 3 children all the child subject used spring 2 or 1 as their test spring. All adult
subjects used spring 1 as their test spring. The subjects were asked to compress the test
spring as much as they could and maintain it for at least 3 seconds without letting the
spring buckle or slip. For details of the experiment see Chapter 2, 3, 5.
6.3.2 Data Reduction
The maximal hold phase for each of the subjects was converted to the dexterity score as
described in Chapter 2. Eureqa (Schmidt & Lipson 2009) was used to nd the optimal
form of the regression function. Additionally the Euclidean length in the state space
94
(

F -
_
F -F ) was calculated and normalized to time, similar to described in Chapter 2 and
Chapter 5. Detrended
uctuation analysis (DFA) was chosen as the nonlinear analysis to
evaluate any correlative memory and diusion processes in the 100-1000ms time window
in the maximal hold forces.
6.4 Results and Discussion
6.4.1 Lifespan changes in the Dexterity Score
Dramatic improvements in low force dexterous manipulation capabilities are seen in chil-
dren throughout the childhood and adolescent years, appearing to improve till early adult-
hood (Fig. 6.3). This appears to match the changes in the corticospinal tract (CST) in
the developmental years changing well into early adulthood (Lebel et al. 2008, Lebel &
Beaulieu 2011, Paus et al. 1999). On the other side of the peak there is a gradual de-
cline from the middle ages and presumably this is accentuated with any other illnesses
or confounds in the later years.
The whole change across the lifespan appears to be best approximated by an exponen-
tial form with increase till early adulthood and gradual consequent decline (Fig. 6.3) with
the RMSE and R
2
4.63 and 0.73 respectively. The exponential t was cross -validated
using Eureqa and seems to match up well. This form of the equation was based only on
part of the data which was used as a training dataset and was validated on the remaining
points (Fig. 6.4, top). In addition this form of equation was chosen based on a Pareto
front of error and complexity of expressions (Fig. 6.4, bottom) ensuring low error and
low complexity such that the sparse data set was not over tted. This unique method for
95
nding the form of equation for biological data appears to be superior to the traditional
approaches, such as polynomial tting (Kurse, Lipson & Valero-Cuevas 2012, Schmidt &
Lipson 2009).
10 30 50 70 90
40
50
60
70
80
90
100
Lifespan change in low force dexterous manipulation abilities (n=240)
Age (Years)
Dexterity Score (%)

Children
Young Adults
Older Adults
Exponential
Fourier
Eureqa Crossvalidated
Eureqa Exp
RMSE=4.63
R
2
=0.73
Fourier/Exp
RMSE=4.68
R
2
=0.76
Figure 6.3: Lifespan change in low force dexterous manipulation capabilities from 4-89
years of age. This includes 240 subjects;142 children (magenta), 32 young adults (blue)
and 66 older adults (red). The exponential regression appears to represent the data well.
This was cross validated with the use of symbolic regression (Eureqa), which used part
of the data for training and validated it based on the remaining points.
6.4.2 Lifespan changes in dynamics of control
Children in the years before 6 appear to have greater problems with maintaining their
ngertip force direction and magnitude stably and occupy a large area in the state space
(Fig. 6.5). With development this improves and children have improved stability in
96
Eureqa computation of regression line based on training and validation set as well
as on a Pareto front of error and complexity of the expressions
Figure 6.4: Regression using training (cyan) and validation (blue) data sets (Top) and
computation of a Pareto front of error and complexity of expressions (Bottom). The
solution which had a combination of low error and low complexity and did not over t
the data was chosen as the solution.
97
overcoming the gentle perturbations of the spring device with the adolescents and the
young adults having similar Euclidean lengths per second. Both middle aged adults and
older adults appear to have worse stability than the younger adults and adolescents in
overcoming the instabilities (Fig. 6.5).
This lifespan change in dynamics of control with improvements in the ability to over-
come instabilities and perturbations and then a gradual decline appears to match our
conceptualized schematic of the changing attractor landscape during the lifespan. The
control against perturbations and instabilities appears worse in the early childhood years
and begins to deteriorate starting in the middle ages. The important point here is also
that in the late adolescent years and early adulthood the forces which the people are
higher leading to higher amounts of perturbations they are able to overcome while main-
taining their ngertip forces stably.
98
Age 16
F
.
.
F
.
.
F
.
.
Age 11
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
Age 4
˙
F
F
.
.
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
Age 9
˙
F
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
.
.
Age 59
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
F
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
F
Age 84
.
.
−0.4
0
0.5
−0.9
0
1
−0.4
0
0.5
F
˙
F
F
Age 29
.
.
[32] [34] [32]
*
*
23−34 45−65 66−88
33 31
21
0
50
100
140
4−6 7−9 10−12 13−16
Ages
Euclidean length in state space
per second
28
68
25 22
[17] [36] [47] [30]
*
*
*
*
Dynamics across Lifespan (4-89 yrs)
Figure 6.5: Changing dynamics of control across the lifespan quantied by Euclidean length in the state space. Representative
subjects for each age range show changes throughout the lifespan (Top). Median values are signicantly dierent, indicated by
the asterisks in both developmental years and with aging (Bottom).
99
6.4.3 Detrended Fluctuation Analysis
The Detrended Fluctuation Analysis (DFA) results in children primarily indicate that
the youngest and oldest children are signicantly dierent from each other (Fig. 6.6,
F=8.26, p<0.005). All the children younger than 6 had Hurst exponent values greater
than 0.5, indicating they had a positive correlation processes or demonstrated persistence.
The young adults had values for their Hurst exponent which are great than 0.5, which
were signicantly dierent from both the middle aged adults and older adults (Fig. 6.7 ,
F=7.53, p<.001).
These results can be understood only in the context of the associated changes in the
lifespan, in particular with the improvement in the dexterity score during development
and the decline in aging. During development the high Hurst exponent values and in
particular the fact that all the children below the age of 6 had high values indicate they
their diusion was enhanced , i.e. the system was persistent. This is possibly due to delays
in the sensorimotor pathways in this age (Fietzek et al. 2000). In the case of the young
adults who performed the best among all in the lifespan, the high Hurst exponent values
probably indicate the lack of constant feedback control, with the controller stepping in
when needed. In the case of the older people who were performing poorly the higher
Hurst exponent values probably indicate that the strategy they use depends on negative
feedback control, i.e. they are constantly correcting presumably to be able to succeed at
the task.
100
0
0.5
1
4−6 13−16
Ages
Hurst Exponent
Changing Stability between Youngest and Oldest Children
Persistent
Anti−Persistent
*
Figure 6.6: Dierences in Hurst Exponent values between youngest (4-6 years of age, left)
and oldest (13-16 years of age, right) children. Values higher than 0.5 indicate positive
correlations are induced i.e. diusion is enhanced while values less than 0.5 indicates
diusion is suppressed. Hurst exponent values of 0.5 indicate Brownian motion or a
random walk, where all time steps are uncorrelated. Signicance values are based on a
one-way ANOVA (p<0.005).
101
0
0.5
1
18−34 45−65 66−89
Ages
Hurst Exponent
Diffusion Analysis for Adults
Persistent Anti−Persistent
*
*
Figure 6.7: Dierences in Hurst Exponent values between young adults (18-34 years of
age, left), middle aged (45-65 years of age, middle) and older adults (66-89 years of age,
right). Values higher than 0.5 indicate positive correlations are induced i.e. diusion
is enhanced while values less than 0.5 indicates diusion is suppressed. Hurst exponent
values of 0.5 indicate Brownian motion or a random walk, where all time steps are un-
correlated. Signicance values are based on a one-way ANOVA (p<0.005).
102
6.5 Conclusions
For low force dexterous manipulation large improvements and gradual declines are seen
across the lifespan in the ability to overcome and postpone instabilities as well as in the
control of ngertip force direction.
6.6 Acknowledgments
We thank Kornelius R acz for his input and assistance with the time series analysis meth-
ods focusing on the DFA.
103
Chapter 7
Conclusions and Future Work
In spite of the low force (< 3N) required for the task developed for the study we are able
to show substantial changes in dexterous manipulation capabilities in healthy humans
across development and in aging. Importantly, we are able to show there is development
of dexterity well into the adolescent years, much longer than previously understood. This
appears to be relatively independent of strength and hand anthropometrics. Children
in early adolescent years appear to have slower hand muscle twitch time-to-peak than
young adults, demonstrated for the rst time with noninvasive method of the EMG
weighted average (EWA). In the case of the adults, the surprising nding was that adults
seem to start to decline in their capabilities even at this low force from the middle age.
Additionally we use some simple linear and nonlinear methods to show the change in the
ability to control dexterous manipulation.
Our current work provides the foundation across the lifespan which needs to be elab-
orated and built upon. Much work is needed in the clinical domain, in particular in
children with cerebral palsy in order to understand and promote the development of
dexterity. The rehabilitation window, so to speak, appears to be wider than previously
104
thought in children and both the timelines as well as the appropriate challenge levels for
this needs to be elucidated. The hope for the future is that this work leads to innovative
and novel methods of appropriate training at specic challenge levels for improvements
in dexterity in clinical populations as well as in healthy aging.
105
Bibliography
Armand, J. (1982), The Origin, Course and Terminations of Corticospinal Fibers in
Various Mammals, Vol. 57, Elsevier, pp. 329{360.
Armand, J., Edgley, S. A., Lemon, R. N. & Olivier, E. (1994), `Protracted postnatal
development of corticospinal projections from the primary motor cortex to hand
motoneurones in the macaque monkey', Exp Brain Res 101(1), 178{82.
Armand, J., Olivier, E., Edgley, S. A. & Lemon, R. N. (1997), `Postnatal development
of corticospinal projections from motor cortex to the cervical enlargement in the
macaque monkey', J. Neurosci. 17(1), 251{266.
Belanger, A. & McComas, A. (1989), `Contractile properties of human skeletal muscle
in childhood and adolescence', European Journal of Applied Physiology and Occupa-
tional Physiology 58(6), 563{567.
Bell, R. D., Macdougall, J. D., Billeter, R. & Howald, H. (1980), `Muscle-ber types
and morphometric analysis of skeletal-muscle in 6-year-old children', Med Sci Sports
Exerc 12(1), 28{31.
Bengtsson, S. L., Nagy, Z., Skare, S., Forsman, L., Forssberg, H. & Ullen, F. (2005), `Ex-
tensive piano practicing has regionally specic eects on white matter development',
Nat Neurosci 8(9), 1148{1150.
Bernhard, C. & Bohm, M. (1954), `Cortical representation and functional signicance of
the corticomotoneuronal system', Archives of Neurology and Psychiatry .
Bleyenheuft, Y., Wilmotte, P. & Thonnard, J.-L. (2010), `Relationship between tactile
spatial resolution and digital dexterity during childhood', Somatosensory & Motor
Research 27(1), 9{14.
Bohannon, R. W. & Smith, M. B. (1987), `Interrater reliability of a modied ashworth
scale of muscle spasticity', Phys Ther 67(2), 206{207.
Bonnard, M., Gall ea, C., De Graaf, J. B. & Pailhous, J. (2007), `Corticospinal control of
the thumb-index grip depends on precision of force control: a transcranial magnetic
stimulation and functional magnetic resonance imagery study in humans', European
Journal of Neuroscience 25(3), 872{880.
Bonnier, B., Eliasson, A. C. & Krumlinde-Sundholm, L. (2006), `Eects of constraint-
induced movement therapy in adolescents with hemiplegic cerebral palsy: a day
camp model', Scand J Occup Ther 13(1), 13{22.
106
Brody, I. A. (1976), `Regulation of isometric contraction in skeletal muscle', Experimental
Neurology 50(3), 673{683.
Brown, W. F., Strong, M. J. & Snow, R. (1988), `Methods for estimating numbers of
motor units in biceps-brachialis muscles and losses of motor units with aging', Muscle
Nerve 11(5), 423{32.
Buchthal, F. & Schmalbruch, H. (1970), `Contraction times and bre types in intact
human muscle', Acta Physiologica Scandinavica 79(4), 435{452.
Burke, D. (2001), `Clinical relevance of the putative c-3{4 propriospinal system in hu-
mans', Muscle & Nerve 24(11), 1437{1439.
Burke, R. E. & Tsairis, P. (1974), `The correlation of physiological properties with his-
tochemical characteristics in single muscle units', Annals of the New York Academy
of Sciences 228(1), 145{158.
Carmeli, E., Patish, H. & Coleman, R. (2003), `The aging hand', J Gerontol A Biol Sci
Med Sci 58(2), M146{152.
Carr, L. J., Harrison, L. M., Evans, A. L. & Stephens, J. A. (1993), `Patterns of central
motor reorganization in hemiplegic cerebral palsy', Brain 116(5), 1223{1247.
Close, J., Ellis, M., Hooper, R., Glucksman, E., Jackson, S. & Swift, C. (1999), `Preven-
tion of falls in the elderly trial (profet): a randomised controlled trial', The Lancet
353(9147), 93{97.
Close, R. (1964), `Dynamic properties of fast and slow skeletal muscles of the rat during
development', The Journal of Physiology 173(1), 74{95.
Cole, K. J. (2006), `Age-related directional bias of ngertip force', Exp Brain Res
175(2), 285{91.
Cole, K. J., Rotella, D. L. & Harper, J. G. (1998), `Tactile impairments cannot explain the
eect of age on a grasp and lift task', Experimental Brain Research 121(3), 263{269.
Cole, K. J., Rotella, D. L. & Harper, J. G. (1999), `Mechanisms for age-related changes
of ngertip forces during precision gripping and lifting in adults', The Journal of
Neuroscience 19(8), 3238{3247.
Collins, J. J. & De Luca, C. J. (1993), `Open-loop and closed-loop control of posture: a
random-walk analysis of center-of-pressure trajectories', Exp Brain Res 95(2), 308{
18.
Collins, J. J. & De Luca, C. J. (1994), `Random walking during quiet standing', Phys
Rev Lett 73(5), 764{767.
Collins, J. J. & De Luca, C. J. (1995), `Upright, correlated random walks: A statistical-
biomechanics approach to the human postural control system', Chaos 5(1), 57{63.
107
Courtine, G., Bunge, M. B., Fawcett, J. W., Grossman, R. G., Kaas, J. H., Lemon,
R., Maier, I., Martin, J., Nudo, R. J., Ramon-Cueto, A., Rouiller, E. M., Schnell,
L., Wannier, T., Schwab, M. E. & Edgerton, V. R. (2007), `Can experiments in
nonhuman primates expedite the translation of treatments for spinal cord injury in
humans?', Nat Med 13(5), 561{566.
Davare, M., Andres, M., Cosnard, G., Thonnard, J. L. & Olivier, E. (2006), `Dissociating
the role of ventral and dorsal premotor cortex in precision grasping', J Neurosci
26(8), 2260{8.
Davids, J. R., Peace, L. C., Wagner, L. V., Gidewall, M. A., Blackhurst, D. W. &
Roberson, W. M. (2006), `Validation of the shriners hospital for children upper
extremity evaluation (shuee) for children with hemiplegic cerebral palsy', J Bone
Joint Surg Am 88(2), 326{333.
Davies, C., White, M. & Young, K. (1983), `Muscle function in children', European Jour-
nal of Applied Physiology and Occupational Physiology 52(1), 111{114.
Davis Sears, E. & Chung, K. C. (2010), `Validity and responsiveness of the jebsen-taylor
hand function test', The Journal of Hand Surgery 35(1), 30{37.
Dayanidhi, S., Hedberg,

A., H agg, I., Lilja, N., Forssberg, H. & Valero-Cuevas, F. (2011),
Dynamic control of ngertip forces: Development in childhood and decline with ag-
ing, in `Proceedings of the 35th Annual Meeting of the American Society of Biome-
chanics'.
DeMatteo, C., Law, M., Russell, D., Pollock, N., Rosenbaum, P. & Walter, S. (1993),
`The reliability and validity of the quality of upper extremity skills test', Physical &
Occupational Therapy in Pediatrics 13(2), 1{18.
Desrosiers, J., Hbert, R., Bravo, G. & Rochette, A. (1999), `Age-related changes in
upper extremity performance of elderly people: a longitudinal study', Experimental
Gerontology 34(3), 393{405.
Deutsch, K. M. & Newell, K. M. (2001), `Age dierences in noise and variability of
isometric force production', J Exp Child Psychol 80(4), 392{408.
Deutsch, K. M. & Newell, K. M. (2002), `Children's coordination of force output in a
pinch grip task', Dev Psychobiol 41(3), 253{64.
Dorfman, L. J. & Bosley, T. M. (1979), `Age?related changes in peripheral and central
nerve conduction in man', Neurology 29(1), 38.
Dum, R. P. & Strick, P. L. (2002), `Motor areas in the frontal lobe of the primate',
Physiology and Behavior 77(4-5), 677{682.
Dum, R. & Strick, P. (1991), `The origin of corticospinal projections from the premotor
areas in the frontal lobe', The Journal of Neuroscience 11(3), 667{689.
108
Duque, J., Thonnard, J.-L., Vandermeeren, Y., Sebire, G., Cosnard, G. & Olivier, E.
(2003), `Correlation between impaired dexterity and corticospinal tract dysgenesis
in congenital hemiplegia', Brain 126(3), 732{747.
Edin, B. B., Westling, G. & Johansson, R. S. (1992), `Independent control of human
nger-tip forces at individual digits during precision lifting', J Physiol 450, 547{64.
Ehrsson, H. H., Fagergren, A., Jonsson, T., Westling, G., Johansson, R. S. & Forssberg,
H. (2000), `Cortical activity in precision- versus power-grip tasks: an fmri study', J
Neurophysiol 83(1), 528{36.
Ehrsson, H. H., Fagergren, E. & Forssberg, H. (2001), `Dierential fronto-parietal activa-
tion depending on force used in a precision grip task: an fmri study', J Neurophysiol
85(6), 2613{23.
Elder, G. C. & Kakulas, B. A. (1993), `Histochemical and contractile property changes
during human muscle development', Muscle Nerve 16(11), 1246{53.
Eliasson, A. C., Forssberg, H., Hung, Y. C. & Gordon, A. M. (2006), `Development of
hand function and precision grip control in individuals with cerebral palsy: a 13-year
follow-up study', Pediatrics 118(4), e1226{36.
Eliasson, A. C., Forssberg, H., Ikuta, K., Apel, I., Westling, G. & Johansson, R. (1995),
`Development of human precision grip. v. anticipatory and triggered grip actions
during sudden loading', Exp Brain Res 106(3), 425{33.
Eliasson, A. C., Gordon, A. M. & Forssberg, H. (1991), `Basic co-ordination of manipu-
lative forces of children with cerebral palsy', Dev Med Child Neurol 33(8), 661{70.
Eliasson, A. C., Gordon, A. M. & Forssberg, H. (1992), `Impaired anticipatory control
of isometric forces during grasping by children with cerebral palsy', Dev Med Child
Neurol 34(3), 216{25.
Eliasson, A. C., Gordon, A. M. & Forssberg, H. (1995), `Tactile control of isometric
ngertip forces during grasping in children with cerebral palsy', Dev Med Child
Neurol 37(1), 72{84.
Enoka, R. M., Christou, E. A., Hunter, S. K., Kornatz, K. W., Semmler, J. G., Taylor,
A. M. & Tracy, B. L. (2003), `Mechanisms that contribute to dierences in mo-
tor performance between young and old adults', Journal of Electromyography and
Kinesiology 13(1), 1{12.
Eyre, J. A. (2007), `Corticospinal tract development and its plasticity after perinatal
injury', Neurosci Biobehav Rev 31(8), 1136{49.
Eyre, J. A., Taylor, J. P., Villagra, F., Smith, M. & Miller, S. (2001), `Evidence of
activity-dependent withdrawal of corticospinal projections during human develop-
ment', Neurology 57(9), 1543{54.
109
Fietzek, U. M., Heinen, F., Berweck, S., Maute, S., Hufschmidt, A., Schulte-Monting,
J., Lucking, C. H. & Korinthenberg, R. (2000), `Development of the corticospinal
system and hand motor function: central conduction times and motor performance
tests', Dev Med Child Neurol 42(4), 220{7.
Fleuren, J. F. M., Voerman, G. E., Erren-Wolters, C. V., Snoek, G. J., Rietman, J. S.,
Hermens, H. J. & Nene, A. V. (2010), `Stop using the ashworth scale for the assess-
ment of spasticity', Journal of Neurology, Neurosurgery & Psychiatry 81(1), 46{52.
Forssberg, H., Eliasson, A. C., Kinoshita, H., Johansson, R. S. & Westling, G. (1991),
`Development of human precision grip. i: Basic coordination of force', Exp Brain Res
85(2), 451{7.
Forssberg, H., Eliasson, A. C., Kinoshita, H., Westling, G. & Johansson, R. S. (1995),
`Development of human precision grip. iv. tactile adaptation of isometric nger forces
to the frictional condition', Exp Brain Res 104(2), 323{30.
Forssberg, H., Eliasson, A. C., Redon-Zouitenn, C., Mercuri, E. & Dubowitz, L. (1999),
`Impaired grip-lift synergy in children with unilateral brain lesions', Brain 122 ( Pt
6), 1157{68.
Forssberg, H., Kinoshita, H., Eliasson, A. C., Johansson, R. S., Westling, G. & Gor-
don, A. M. (1992), `Development of human precision grip. ii. anticipatory control of
isometric forces targeted for object's weight', Exp Brain Res 90(2), 393{8.
Friel, K. M. & Martin, J. H. (2007), `Bilateral activity-dependent interactions in the
developing corticospinal system', J Neurosci 27(41), 11083{90.
Fugl-Meyer, A. R., Jaasko, L., Leyman, I., Olsson, S. & Steglind, S. (1975), `The post-
stroke hemiplegic patient. 1. a method for evaluation of physical performance', Scand
J Rehabil Med 7(1), 13{31.
Fuglevand, A. J. (2011), `Mechanical properties and neural control of human hand motor
units', The Journal of Physiology 589(23), 5595{5602.
Gajdosik, R. L. & Bohannon, R. W. (1987), `Clinical measurement of range of motion.
review of goniometry emphasizing reliability and validity', Phys Ther 67(12), 1867{
72.
Galganski, M. E., Fuglevand, A. J. & Enoka, R. M. (1993), `Reduced control of motor
output in a human hand muscle of elderly subjects during submaximal contractions',
Journal of Neurophysiology 69(6), 2108{2115.
Gall ea, C., de Graaf, J. B., Bonnard, M. & Pailhous, J. (2005), `High level of dexterity:
dierential contributions of frontal and parietal areas', Neuroreport 16(12), 1271{
1274.
Gallus, J. & Mathiowetz, V. (2003), `Test-retest reliability of the purdue pegboard for
persons with multiple sclerosis', Am J Occup Ther 57(1), 108{11.
110
Gilmore, R., Sakzewski, L. & Boyd, R. (2009), `Upper limb activity measures for 5- to 16-
year-old children with congenital hemiplegia: a systematic review', Developmental
Medicine & Child Neurology 52(1), 14{21.
Glenmark, B., Hedberg, G., Kaijser, L. & Jansson, E. (1994), `Muscle strength from
aldolescence to adulthood relationship to muscle bre types', European Journal of
Applied Physiology and Occupational Physiology 68(1), 9{19.
Gordon, A. M., Forssberg, H., Johansson, R. S., Eliasson, A. C. & Westling, G. (1992),
`Development of human precision grip. iii. integration of visual size cues during the
programming of isometric forces', Exp Brain Res 90(2), 399{403.
Gordon, A. M., Schneider, J. A., Chinnan, A. & Charles, J. R. (2007), `Ecacy of a
hand-arm bimanual intensive therapy (habit) in children with hemiplegic cerebral
palsy: a randomized control trial', Dev Med Child Neurol 49(11), 830{8.
Greaves, S., Imms, C., Dodd, K. & Krumlinde-Sundholm, L. (2010), `Assessing bimanual
performance in young children with hemiplegic cerebral palsy: a systematic review',
Developmental Medicine & Child Neurology 52(5), 413{421.
H ager-Ross, C. & R osblad, B. (2002), `Norms for grip strength in children aged 4{16
years', Acta Pdiatrica 91(6), 617{625.
H ager-Ross, C. & Schieber, M. H. (2000), `Quantifying the independence of human nger
movements: Comparisons of digits, hands, and movement frequencies', J. Neurosci.
20(22), 8542{8550.
Hener, R. S. & Masterton, R. B. (1983), `The role of the corticospinal tract in the
evolution of human digital dexterity', Brain, Behavior and Evolution 23(3-4), 165{
183.
Hirtz, D., Thurman, D. J., Gwinn-Hardy, K., Mohamed, M., Chaudhuri, A. R. & Zalut-
sky, R. (2007), `How common are the "common" neurologic disorders?', Neurology
68(5), 326{37.
Holmefur, M., Krumlinde-Sundholm, L. & Eliasson, A. C. (2007), `Interrater and in-
trarater reliability of the assisting hand assessment', Am J Occup Ther 61(1), 79{84.
Holmstr om, L., Lennartsson, F., Eliasson, A. C., Flodmark, O., Clark, C., Tedro, K.,
Forssberg, H. & Vollmer, B. (2011), `Diusion mri in corticofugal bers correlates
with hand function in unilateral cerebral palsy', Neurology 77(8), 775{83.
Holmstr om, L., Vollmer, B., Tedro, K., Islam, M., Persson, J. K., Kits, A., Forss-
berg, H. & Eliasson, A. C. (2010), `Hand function in relation to brain lesions and
corticomotor-projection pattern in children with unilateral cerebral palsy', Dev Med
Child Neurol 52(2), 145{52.
House, J. H., Gwathmey, F. W. & Fidler, M. O. (1981), `A dynamic approach to the
thumb-in palm deformity in cerebral palsy', J Bone Joint Surg Am 63(2), 216{25.
111
Jebsen, R. H., Taylor, N., Trieschmann, R. B., Trotter, M. J. & Howard, L. A. (1969), `An
objective and standardized test of hand function', Arch Phys Med Rehabil 50(6), 311{
9.
Jesunathadas, M., Marmon, A. R., Gibb, J. M. & Enoka, R. M. (2010), `Recruitment
and derecruitment characteristics of motor units in a hand muscle of young and old
adults', Journal of Applied Physiology 108(6), 1659{1667.
Jette, A. M. (2006), `Toward a common language for function, disability, and health',
PHYS THER 86(5), 726{734.
Johansson, R. S. & Flanagan, J. R. (2009), `Coding and use of tactile signals from the
ngertips in object manipulation tasks', Nat Rev Neurosci 10(5), 345{59.
Johansson, R. S. & Westling, G. (1984), `Roles of glabrous skin receptors and sensorimotor
memory in automatic control of precision grip when lifting rougher or more slippery
objects', Exp Brain Res 56(3), 550{64.
Johansson, R. S. & Westling, G. (1988), `Coordinated isometric muscle commands ade-
quately and erroneously programmed for the weight during lifting task with precision
grip', Experimental Brain Research 71(1), 59{71.
Johnson, L. M., Randall, M. J., Reddihough, D. S., Oke, L. E., Byrt, T. A. & Bach, T. M.
(1994), `Development of a clinical assessment of quality of movement for unilateral
upper-limb function', Dev Med Child Neurol 36(11), 965{73.
Kantz, H. & Schreiber, T. (2004), Nonlinear Time Series Analysis, Cambridge.
Kellor, M., Frost, J., Silberberg, N., Iversen, I. & Cummings, R. (1971), `Hand strength
and dexterity', Am J Occup Ther 25(2), 77{83.
Kraemer, W. J., Fry, A. C., Frykman, P. N., Conroy, B. & Homan, J. (1989), `Resistance
training and youth', Pediatric Exercise Science 1(4), 336{350.
Krumlinde-Sundholm, L., Holmefur, M., Kottorp, A. & Eliasson, A. C. (2007), `The
assisting hand assessment: current evidence of validity, reliability, and responsiveness
to change', Dev Med Child Neurol 49(4), 259{64.
Kuhtz-Buschbeck, J. P., Ehrsson, H. H. & Forssberg, H. (2001), `Human brain activity
in the control of ne static precision grip forces: an fmri study', Eur J Neurosci
14(2), 382{90.
Kuhtz-Buschbeck, J. P., Sundholm, L. K., Eliasson, A. C. & Forssberg, H. (2000), `Quan-
titative assessment of mirror movements in children and adolescents with hemiplegic
cerebral palsy', Dev Med Child Neurol 42(11), 728{36.
Kurse, M. U., Lipson, H. & Valero-Cuevas, F. J. (2012), `Extrapolatable analytical func-
tions for tendon excursions and moment arms from sparse datasets', IEEE Trans
Biomed Eng 59(6), 1572{82.
112
Kutch, J. J., Kuo, A. D. & Rymer, W. Z. (2010), `Extraction of individual muscle me-
chanical action from endpoint force', Journal of Neurophysiology 103(6), 3535{3546.
Kuypers, H. G. (1960), `Central cortical projections to motor and somato-sensory cell
groups. an experimental study in the rhesus monkey', Brain 83, 161{84.
Kuypers, H. G. (1962), `Corticospinal connections: postnatal development in the rhesus
monkey', Science 138, 678{80.
Kuypers, H. G., Fleming, W. R. & Farinholt, J. W. (1960), `Descending projections
to spinal motor and sensory cell groups in the monkey: cortex versus subcortex',
Science 132, 38{40.
Kuypers, H. G. J. M. (1964), The Descending Pathways to the Spinal Cord, their Anatomy
and Function, Vol. Volume 11, Elsevier, pp. 178{202. doi: DOI: 10.1016/S0079-
6123(08)64048-0.
Lawrence, D. G. & Kuypers, H. G. (1968a), `The functional organization of the motor
system in the monkey. i. the eects of bilateral pyramidal lesions', Brain 91(1), 1{14.
Lawrence, D. G. & Kuypers, H. G. (1968b), `The functional organization of the motor
system in the monkey. ii. the eects of lesions of the descending brain-stem pathways',
Brain 91(1), 15{36.
Lebel, C. & Beaulieu, C. (2011), `Longitudinal development of human brain wiring con-
tinues from childhood into adulthood', The Journal of Neuroscience 31(30), 10937{
10947.
Lebel, C., Walker, L., Leemans, A., Phillips, L. & Beaulieu, C. (2008), `Microstruc-
tural maturation of the human brain from childhood to adulthood', Neuroimage
40(3), 1044{55.
Lemon, R. N. (1993), `The g. l. brown prize lecture. cortical control of the primate hand',
Exp Physiol 78(3), 263{301.
Lemon, R. N. (1999), `Neural control of dexterity: what has been achieved?', Exp Brain
Res 128(1-2), 6{12.
Lemon, R. N. (2008), `Descending pathways in motor control', Annu Rev Neurosci
31, 195{218.
Lemon, R. N. & Griths, J. (2005), `Comparing the function of the corticospinal system in
dierent species: organizational dierences for motor specialization?', Muscle Nerve
32(3), 261{79.
Lexell, J. & Downham, D. (1992), `What is the eect of ageing on type 2 muscle bres?',
Journal of the Neurological Sciences 107(2), 250{251.
Lexell, J., Henriksson-Lars en, K., Winblad, B. & Sj ostr om, M. (1983), `Distribution of
dierent ber types in human skeletal muscles: Eects of aging studied in whole
muscle cross sections', Muscle & Nerve 6(8), 588{595.
113
Lexell, J., Taylor, C. C. & Sj ostr om, M. (1988), `What is the cause of the ageing atrophy?:
Total number, size and proportion of dierent ber types studied in whole vastus
lateralis muscle from 15- to 83-year-old men', Journal of the Neurological Sciences
84(2{3), 275{294.
Lindberg, P., Ody, C., Feydy, A. & Maier, M. (2009), `Precision in isometric precision grip
force is reduced in middle-aged adults', Experimental Brain Research 193(2), 213{
224.
Lowe, B. D. (2001), `Precision grip force control of older and younger adults, revisited',
Journal of Occupational Rehabilitation 11(4), 267{279.
Marmon, A. R., Gould, J. R. & Enoka, R. M. (2011), `Practicing a functional task
improves steadiness with hand muscles in older adults', Medicine & Science in Sports
& Exercise 43(8), 1531{1537 10.1249/MSS.0b013e3182100439.
Marmon, A. R., Pascoe, M. A., Schwartz, R. S. & Enoka, R. M. (2011), `Associations
among strength, steadiness, and hand function across the adult life span', Medicine
& Science in Sports & Exercise 43(4), 560{567 10.1249/MSS.0b013e3181f3f3ab.
Martin, J. H. (2005), `The corticospinal system: from development to motor control',
Neuroscientist 11(2), 161{73.
Martin, J. H., Donarummo, L. & Hacking, A. (2000), `Impairments in prehension pro-
duced by early postnatal sensory motor cortex activity blockade', J Neurophysiol
83(2), 895{906.
Martin, J. H., Friel, K. M., Salimi, I. & Chakrabarty, S. (2007), `Activity- and use-
dependent plasticity of the developing corticospinal system', Neurosci Biobehav Rev
31(8), 1125{35.
Martin, J. H. & Lee, S. J. (1999), `Activity-dependent competition between developing
corticospinal terminations', Neuroreport 10(11), 2277{82.
Martin, R. J. F., Dore, E., Twisk, J., Van praagh, E., Hautier, C. A. & Bedu, M. (2004),
`Longitudinal changes of maximal short-term peak power in girls and boys during
growth', Medicine & Science in Sports & Exercise 36(3), 498{503.
Matano, S. & Hirasaki, E. (1997), `Volumetric comparisons in the cerebellar complex
of anthropoids, with special reference to locomotor types', Am J Phys Anthropol
103(2), 173{83.
Mathiowetz, V., Volland, G., Kashman, N. & Weber, K. (1985), `Adult norms for the
box and block test of manual dexterity', Am J Occup Ther 39(6), 386{91.
Mathiowetz, V., Weber, K., Volland, G. & Kashman, N. (1984), `Reliability and validity
of grip and pinch strength evaluations', J Hand Surg Am 9(2), 222{6.
Mathiowetz, V., Wiemer, D. M. & Federman, S. M. (1986), `Grip and pinch strength:
norms for 6- to 19-year-olds', Am J Occup Ther 40(10), 705{11.
114
McComas, A. J., Sica, R. E. P. & Petito, F. (1973), `Muscle strength in boys of dierent
ages', Journal of Neurology, Neurosurgery & Psychiatry 36(2), 171{173.
Milner-Brown, H. S., Stein, R. B. & Yemm, R. (1973a), `Changes in ring rate of hu-
man motor units during linearly changing voluntary contractions', The Journal of
Physiology 230(2), 371{390.
Milner-Brown, H. S., Stein, R. B. & Yemm, R. (1973b), `The contractile properties of hu-
man motor units during voluntary isometric contractions', The Journal of Physiology
228(2), 285{306.
Milner-Brown, H. S., Stein, R. B. & Yemm, R. (1973c), `The orderly recruitment of human
motor units during voluntary isometric contractions', The Journal of Physiology
230(2), 359{370.
Mosier, K. M., Lau, C., Wang, Y., Venkadesan, M. & Valero-Cuevas, F. J. (2011), `Con-
trolling instabilities in manipulation requires specic cortical-striatal-cerebellar net-
works', Journal of Neurophysiology (In Press).
Muir, R. B. & Lemon, R. N. (1983), `Corticospinal neurons with a special role in precision
grip', Brain Research 261(2), 312{316.
Muller, K., Homberg, V. & Lenard, H. G. (1991), `Magnetic stimulation of motor cor-
tex and nerve roots in children. maturation of cortico-motoneuronal projections',
Electroencephalogr Clin Neurophysiol 81(1), 63{70.
Nagi, S. Z. (1964), `A study in the evaluation of disability and rehabilitation poten-
tial: Concepts, methods, and procedures', Am J Public Health Nations Health
54(9), 1568{1579.
Nakajima, K., Maier, M. A., Kirkwood, P. A. & Lemon, R. N. (2000), `Striking dierences
in transmission of corticospinal excitation to upper limb motoneurons in two primate
species', J Neurophysiol 84(2), 698{709.
Nathan, P. W. & Smith, M. C. (1982), `The rubrospinal and central tegmental tracts in
man', Brain 105(Pt 2), 223{69.
Newell, K. M., Liu, Y. T. & Mayer-Kress, G. (2001), `Time scales in motor learning and
development', Psychol Rev 108(1), 57{82.
Nudo, R. J. & Masterton, R. B. (1990), `Descending pathways to the spinal cord, iii:
Sites of origin of the corticospinal tract', The Journal of Comparative Neurology
296(4), 559{583.
O'Brien, T. D., Reeves, N. D., Baltzopoulos, V., Jones, D. A. & Maganaris, C. N. (2010),
`In vivo measurements of muscle specic tension in adults and children', Experimental
Physiology 95(1), 202{210.
Oxford Grice, K., Vogel, K. A., Le, V., Mitchell, A., Muniz, S. & Vollmer, M. A. (2003),
`Adult norms for a commercially available nine hole peg test for nger dexterity',
Am J Occup Ther 57(5), 570{3.
115
Parikh, P. J. & Cole, K. J. (2012), `Handling objects in old age: forces and moments
acting on the object', Journal of Applied Physiology 112(7), 1095{1104.
Paus, T., Zijdenbos, A., Worsley, K., Collins, D. L., Blumenthal, J., Giedd, J. N.,
Rapoport, J. L. & Evans, A. C. (1999), `Structural maturation of neural pathways
in children and adolescents: in vivo study', Science 283(5409), 1908{11.
Peng, C.-K., Havlin, S., Stanley, H. E. & Goldberger, A. L. (1995), `Quantication of
scaling exponents and crossover phenomena in nonstationary heartbeat time series',
Chaos: An Interdisciplinary Journal of Nonlinear Science 5(1), 82{87.
Pharoah, P. O., Platt, M. J. & Cooke, T. (1996), `The changing epidemiology of cerebral
palsy', Arch Dis Child Fetal Neonatal Ed 75(3), F169{73.
Pieperho, P., H omke, L., Schneider, F., Habel, U., Shah, N. J., Zilles, K. & Amunts,
K. (2008), `Deformation eld morphometry reveals age-related structural dierences
between the brains of adults up to 51 years', The Journal of Neuroscience 28(4), 828{
842.
Pierrot-Deseilligny, E. (2002), `Propriospinal transmission of part of the corticospinal
excitation in humans', Muscle & Nerve 26(2), 155{172.
Platz, T., Pinkowski, C., van Wijck, F., Kim, I. H., di Bella, P. & Johnson, G. (2005),
`Reliability and validity of arm function assessment with standardized guidelines for
the fugl-meyer test, action research arm test and box and block test: a multicentre
study', Clin Rehabil 19(4), 404{11.
Plautz, E. J., Milliken, G. W. & Nudo, R. J. (2000), `Eects of repetitive motor training
on movement representations in adult squirrel monkeys: role of use versus learning',
Neurobiol Learn Mem 74(1), 27{55.
Poole, J. L., Burtner, P. A., Torres, T. A., McMullen, C. K., Markham, A., Marcum,
M. L., Anderson, J. B. & Qualls, C. (2005), `Measuring dexterity in children using
the nine-hole peg test', Journal of Hand Therapy 18(3), 348{351.
Porter, R. (1985), `The corticomotoneuronal component of the pyramidal tract: Cor-
ticomotoneuronal connections and functions in primates', Brain Research Reviews
10(1), 1{26.
Randall, M., Carlin, J. B., Chondros, P. & Reddihough, D. (2001), `Reliability of the
melbourne assessment of unilateral upper limb function', Dev Med Child Neurol
43(11), 761{7.
Ranganathan, V. K., Siemionow, V., Sahgal, V. & Yue, G. H. (2001), `Eects of aging
on hand function', Journal of the American Geriatrics Society 49(11), 1478{1484.
Rathelot, J. A. & Strick, P. L. (2006), `Muscle representation in the macaque motor
cortex: an anatomical perspective', Proc Natl Acad Sci U S A 103(21), 8257{62.
116
Rose, J. & McGill, K. C. (2005), `Neuromuscular activation and motor-unit ring charac-
teristics in cerebral palsy', Developmental Medicine & Child Neurology 47(5), 329{
336.
Sakzewski, L., Boyd, R. & Ziviani, J. (2007), `Clinimetric properties of participation
measures for 5- to 13-year-old children with cerebral palsy: a systematic review',
Developmental Medicine & Child Neurology 49(3), 232{240.
Schieber, M. H. & Santello, M. (2004), `Hand function: peripheral and central constraints
on performance', J Appl Physiol 96(6), 2293{2300.
Schmidt, M. & Lipson, H. (2009), `Distilling free-form natural laws from experimental
data', Science 324(5923), 81{85.
Smith, Y. A., Hong, E. & Presson, C. (2000), `Normative and validation studies of the
nine-hole peg test with children', Percept Mot Skills 90(3 Pt 1), 823{43.
Steenbergen, B. & Utley, A. (2005), `Cerebral palsy: recent insights into movement devi-
ation', Motor Control 9(4), 353{6.
Stein, R. B., French, A. S., Mannard, A. & Yemm, R. (1972), `New methods for analysing
motor function in man and animals', Brain Research 40(1), 187{192.
Stephens, J. A. & Usherwood, T. P. (1977), `The mechanical properties of human motor
units with special reference to their fatiguability and recruitment threshold', Brain
Research 125(1), 91{97.
Stucki, G., Ewert, T. & Cieza, A. (2002), `Value and application of the icf in rehabilitation
medicine', Disability & Rehabilitation 24(17), 932{938.
Sutherland, D. (1997), `The development of mature gait', Gait & Posture 6(2), 163{170.
Talati, A., Valero-Cuevas, F. J. & Hirsch, J. (2005), `Visual and tactile guidance of
dexterous manipulation tasks: an fmri study', Percept Mot Skills 101(1), 317{34.
Taylor, N., Sand, P. L. & Jebsen, R. H. (1973), `Evaluation of hand function in children',
Arch Phys Med Rehabil 54(3), 129{35.
Thomas, C. K., Bigland-Richie, B. & Johansson, R. S. (1991), `Force-frequency relation-
ships of human thenar motor units', J Neurophysiol 65(6), 1509{16.
Thomas, C. K., Ross, B. H. & Stein, R. B. (1986), `Motor-unit recruitment in human
rst dorsal interosseous muscle for static contractions in three dierent directions',
Journal of Neurophysiology 55(5), 1017{1029.
Tin, J. & Asher, E. J. (1948), `The purdue pegboard; norms and studies of reliability
and validity', J Appl Psychol 32(3), 234{47.
Valero-Cuevas, F. J., Smaby, N., Venkadesan, M., Peterson, M. & Wright, T. (2003),
`The strength-dexterity test as a measure of dynamic pinch performance', J Biomech
36(2), 265{70.
117
van de Ven-Stevens, L. A., Munneke, M., Terwee, C. B., Spauwen, P. H. & van der
Linde, H. (2009), `Clinimetric properties of instruments to assess activities in patients
with hand injury: a systematic review of the literature', Arch Phys Med Rehabil
90(1), 151{69.
Vandermeeren, Y., Sebire, G., Grandin, C. B., Thonnard, J. L., Schlogel, X. & De Volder,
A. G. (2003), `Functional reorganization of brain in children aected with congenital
hemiplegia: fmri study', Neuroimage 20(1), 289{301.
Venkadesan, M., Guckenheimer, J. & Valero-Cuevas, F. J. (2007), `Manipulating the edge
of instability', J Biomech 40(8), 1653{61.
Vollmer, B., Holmstr om, L., Forsman, L., Krumlinde-Sundholm, L., Valero-Cuevas, F. J.,
Forssberg, H. & Ull en, F. (2010), `Evidence of validity in a new method for mea-
surement of dexterity in children and adolescents', Developmental Medicine & Child
Neurology 52(10), 948{954.
Westling, G. & Johansson, R. S. (1984), `Factors in
uencing the force control during
precision grip', Exp Brain Res 53(2), 277{84.
Wolpert, D. M., Diedrichsen, J. & Flanagan, J. R. (2011), `Principles of sensorimotor
learning', Nat Rev Neurosci 12(12), 739{751.
Wolpert, D. M., Miall, R. C. & Kawato, M. (1998), `Internal models in the cerebellum',
Trends in Cognitive Sciences 2(9), 338{347.
118

Abstract (if available)

Abstract This dissertation focuses on the change in low force (< 3N) dexterous manipulation capabilities across the lifespan. A simple device based on the mechanical properties of springs allowed us to systematically test the control and the change of dexterous manipulation skills across the lifespan over 240 participants from 4-89 years of age. Dexterous manipulation capabilities improve dramatically during early childhood and adolescence, followed by gradual declines from the middle age. Here we show that the timelines of development of dexterity are much longer than previously thought and continue well into late adolescence, matching known changes in neural development. In addition, these improvements appear to be poorly predicted by changes in strength and hand anthropometrics. ❧ Muscle twitch properties of the hand muscles, specifically the time-to-peak of the first dorsal interosseous are shown to be slower in early adolescent children and peripheral changes could also be contributing to changes in dexterous manipulation. This was discovered by applying a previously developed noninvasive method, the EMG weighted average (EWA) used for the first time used in children. The parameters used by sensory modalities for dexterous manipulation appear to be different from those used for static tasks. Starting in the middle ages (45-65 years) there is a decline in both the control of fingertip force direction as well as the ability to overcome and postpone instabilities. By observing the dynamics of the whole time series we are able to show that elderly people and young children share certain aspects of control of the dynamics of manipulation.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Behavioral and neurophysiological studies of hand dexterity in health and Parkinson's disease

PDF

Demographic and clinical covariates of sensorimotor processing

PDF

Quantification of lower extremity dynamic capability: implications for anterior cruciate ligament injury and change of direction ability

PDF

Investigating the role of muscle physiology and spinal circuitry in sensorimotor control

PDF

Task-dependent modulation of corticomuscular coherence during dexterous manipulation

PDF

Neuromuscular dynamics in the context of motor redundancy

PDF

Learning reaching skills in non-disabled and post-stroke individuals

PDF

Facilitating myocontrol for children with cerebral palsy

PDF

Trunk control during dynamic balance: effects of cognitive dual-task interference and a history of recurrent low back pain

PDF

Experimental and model-based analyses of physiological determinants of force variability

PDF

The footprint of pain: investigating persistence of altered trunk control in recurrent low back pain

PDF

Relationship between brain structure and motor behavior in chronic stroke survivors

PDF

Cortical inputs to leg muscles in increasingly difficult standing tasks

PDF

Bio-inspired tendon-driven systems: computational analysis, optimization, and hardware implementation

PDF

Development of postural strategies for the control of forward momentum

PDF

The representation, learning, and control of dexterous motor skills in humans and humanoid robots

PDF

Trunk neuromechanics during turning: a window into recurrent low back pain

PDF

Using ecological momentary assessment to study the impact of social-cognitive factors on paretic hand use after stroke

PDF

insideOut: Estimating joint angles in tendon-driven robots using Artificial Neural Networks and non-collocated sensors

PDF

Manipulating cognitive intensity during aerobic exercise: clinical proof of concept

Asset Metadata

Creator Dayanidhi, Sudarshan (author)

Core Title Behavioral, muscular and dynamical changes in low force dexterous manipulation during development and aging

School School of Dentistry

Degree Doctor of Philosophy

Degree Program Biokinesiology

Publication Date 07/25/2012

Defense Date 04/17/2012

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aging,Childhood,Development,dexterity,dexterous manipulation,dynamics,Hand,low force,muscle contraction time,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Valero-Cuevas, Francisco J. (committee chair), Sanger, Terence D. (committee member), Winstein, Carolee J. (committee member)

Creator Email dayanidh@usc.edu,sududay@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-64389

Unique identifier UC11290473

Identifier usctheses-c3-64389 (legacy record id)

Legacy Identifier etd-DayanidhiS-986.pdf

Dmrecord 64389

Document Type Dissertation

Rights Dayanidhi, Sudarshan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA