VERBAL MEMORY IN IDIOPATHIC NON-DEMENTED...
Transcript of VERBAL MEMORY IN IDIOPATHIC NON-DEMENTED...
1
VERBAL MEMORY IN IDIOPATHIC NON-DEMENTED PARKINSON’S DISEASE: A STRUCTURAL MRI AND QUANTITATIVE WHITE MATTER TRACTOGRAPHY
ANALYSIS
By
JARED J. TANNER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2013
© 2013 Jared J. Tanner
3
To my wife Kristi and my children Anwyn, Lily, and Zachary
4
ACKNOWLEDGMENTS
I acknowledge my wonderful wife Kristi and children Anwyn, Lily, and Zach for
their support through the many years of my schooling. They have been with me every
step of the path, supporting with patience and love. I also acknowledge my parents who
raised me with a love of learning and education. I want to acknowledge the contributions
of my dissertation committee members – Dr. Catherine Price, Dr. Dawn Bowers, Dr.
Michael Okun, Dr. Thomas Mareci, and Dr. Thomas Kerkhoff – who provided guidance
along the way. This project would not exist without the support and guidance of my
mentor, committee chair, strong advocate, and friend, Cate Price. She supported me
through the easy and difficult times and has been the greatest mentor for which a
student could wish. Many other people contributed to the work of this dissertation
including Jade Ward, Bill Triplett, Nadine Schwab, Peter Nguyen, Sandra Mitchell,
Alana Freedland, Julianna Jaramillo, and Nicole Coronado. I would also like to
acknowledge the support of many in the department of Clinical and Health Psychology. I
am grateful for the assistance of the Alumni Fellowship, which allowed me to pursue my
dreams. Most importantly I would like to acknowledge the blessings and support of my
Heavenly Father, who blessed me with opportunities and abilities that led me to this
point.
5
TABLE OF CONTENTS
page
ACKNOWLEDGMENTS .................................................................................................. 4
LIST OF TABLES ............................................................................................................ 8
LIST OF FIGURES .......................................................................................................... 9
LIST OF OBJECTS ....................................................................................................... 11
ABSTRACT ................................................................................................................... 12
CHAPTER
1 INTRODUCTION .................................................................................................... 14
Parkinson’s Disease ............................................................................................... 14
PD Prevalence and Impact ............................................................................... 14
PD Etiology ....................................................................................................... 15
Cognitive Considerations in PD ........................................................................ 16
Memory for Verbal Information ............................................................................... 17
Cognitive and Biological Substrates of Memory ............................................... 18
Tests of Verbal Memory ................................................................................... 21
Is there a Verbal Memory Impairment in Parkinson’s Disease? .............................. 29
Parkinson’s Disease is Heterogeneous ............................................................ 31
Language and Semantic Knowledge in Parkinson’s Disease ........................... 39
Considerations for Neuroanatomical Predictors of Verbal Memory Impairment in PD ....................................................................................................................... 43
Entorhinal Cortex and Medial Temporal Lobe .................................................. 44
The Cingulate in PD ......................................................................................... 47
Acetylcholine in PD .......................................................................................... 54
2 AIMS, HYPOTHESES, AND METHODS ................................................................ 62
Study Rationale ...................................................................................................... 62
Specific Aims and Hypotheses ............................................................................... 63
6
Aim 1: Is there a Verbal Memory Deficit in PD? ............................................... 63
Aim 2: Are there Differences in Verbal Memory Brain Structures for PD? ........ 64
Aim 3: Are there Specific Structural-Cognitive Patterns? ................................. 66
Methods .................................................................................................................. 67
Study Design .................................................................................................... 67
Participants ....................................................................................................... 67
Neuropsychological Variables .......................................................................... 69
Philadelphia Repeatable Verbal Learning Test ................................................ 69
Story Memory ................................................................................................... 70
Boston Naming Test ......................................................................................... 71
Association Index ............................................................................................. 71
Imaging Protocols ............................................................................................. 72
Image Processing ............................................................................................. 74
Statistical Analyses .......................................................................................... 78
3 RESULTS ............................................................................................................... 84
Participant Characteristics ...................................................................................... 84
Aim 1: Is there a Verbal Memory Deficit in PD? ...................................................... 84
Multivariate Analyses: Group by Verbal Memory .............................................. 84
Sub-aim: Language Group Differences ............................................................ 84
PD Verbal Memory Performance by Index ....................................................... 84
Within Group Heterogeneity of Memory Impairment ........................................ 86
Aim 2: Is there a Difference in Verbal Memory Brain Structures for PD? ................ 88
Medial Temporal Lobe Structure Group Differences ........................................ 88
Within Group Correlations With Disease Severity ............................................ 88
Sub-aim: Language Track Group Differences .................................................. 89
Aim 3: Are there Specific Structural-Cognitive Patterns? ........................................ 89
7
Brain Variables and Verbal Memory Performances Regardless of Group Type .............................................................................................................. 89
Brain Variables and Semantic Performances Regardless of Group Type ........ 90
Verbal Memory Composite and Structure Relationships .................................. 90
Associations with Disease Severity .................................................................. 90
Within Group Comparisons .............................................................................. 91
Dissociation Between PrVLT Performance, Left ERC-RSC EW, and Left AF EW ................................................................................................................ 92
4 DISCUSSION ....................................................................................................... 113
Aim 1: Is there a Verbal Memory Deficit in PD? .................................................... 113
Aim 2: Are there Differences in Verbal Memory Brain Structures for PD? ............ 117
Causes of Entorhinal Volume Loss ................................................................ 118
Explanations of ERC-RSC Findings ............................................................... 120
Aim 3: Are there Specific Structural-Cognitive Patterns? ...................................... 122
Final Summary ............................................................................................... 124
Limitations and Strengths ............................................................................... 127
Future Directions ............................................................................................ 131
APPENDIX
A PARTICIPANT INCLUSION AND EXCLUSION CRITERIA .................................. 133
PD Exclusion Criteria ..................................................................................... 133
Parallel Control Group Exclusion Criteria ....................................................... 134
B MAGNETIC RESONANCE IMAGE PROCESSING WORKFLOW ....................... 135
C AIM 2 WITHIN-GROUP SUB-AIM: SIDE OF ONSET AND LATERALIZED DIFFERENCES .................................................................................................... 136
REFERENCE LIST...................................................................................................... 137
BIOGRAPHICAL SKETCH .......................................................................................... 157
8
LIST OF TABLES
Table page
3-1 Participant Baseline Characteristics ................................................................... 93
3-2 Verbal memory MANOVA group contrast ........................................................... 94
3-3 Semantic knowledge MANOVA group contrast .................................................. 95
3-4 Percent PD verbal memory impairment .............................................................. 96
3-5 Medial temporal lobe structures group contrast .................................................. 97
3-6 Left arcuate fasciculus group contrast ................................................................ 98
3-7 Brain – verbal memory and language bivariate correlations ............................... 99
3-8 Verbal memory and disease severity correlations ............................................ 100
C-1 Parkinson’s disease side of onset group contrast ............................................. 136
9
LIST OF FIGURES
Figure page
1-1 Schematic of the long-term memory system ...................................................... 57
1-2 Rendering of the anatomy of the medial temporal lobe ...................................... 58
1-3 Schematic representation of the structure and connections of the medial temporal lobe. ..................................................................................................... 59
1-4 Simplified schematic representation of the Papez circuit. ................................... 60
1-5 Fiber tracking image showing connections between the retrosplenial cortex and the entorhinal cortex. ................................................................................... 61
2-1 Schematic showing the image processing pipeline from scanner to fiber tracking and volumetric output ............................................................................ 81
2-2 Series of images demonstrating the modeling of diffusion within a single voxel in complex white matter............................................................................. 82
2-3 Series of images demonstrating the modeling of diffusion within a single voxel in simple white matter................................................................................ 83
3-1 Verbal memory index scores showing significant group difference (PD < Controls) ........................................................................................................... 101
3-2 Image showing cumulative frequency percents for PD participants across verbal memory index scores ............................................................................. 102
3-3 Chart showing PD participants’ cumulative z score impairment on 3 PrVLT index scores ..................................................................................................... 103
3-4 Chart showing all PD participants’ cumulative z score impairment on 3 Story index scores ..................................................................................................... 104
3-5 Chart showing all PD participants’ cumulative z score impairment on 2 verbal memory measures ............................................................................................ 105
3-6 Images representative of MTL volumetric and fiber tracking results ................. 106
3-7 Images showing group X brain structure scatter plots ...................................... 107
3-8 Image showing group X left AF normalized edge weight .................................. 108
3-9 A representative image showing the AF fiber tracking. ..................................... 109
10
3-10 Scatter plot showing relationship between left ERC – RSC EW and PrVLT recognition discriminability index scores ........................................................... 110
3-11 Images showing scatter plots relating left ERC volume and Story memory index scores ..................................................................................................... 111
3-12 Image showing scatter plot relating left ERC volume and verbal memory composite score ............................................................................................... 112
4-1 Representative images showing suboptimal arcuate fasciculus fiber tracking . 132
11
LIST OF OBJECTS
Object page
1-1. Supplemental video created and narrated by Jared Tanner giving an introduction to the cingulum (YouTube video) .................................................... 48
12
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
VERBAL MEMORY IN IDIOPATHIC NON-DEMENTED PARKINSON’S DISEASE: A STRUCTURAL MRI AND QUANTITATIVE WHITE MATTER TRACTOGRAPHY
ANALYSIS
By
Jared J. Tanner
August 2013
Chair: Catherine C. Price Major: Psychology
Introduction: While Parkinson’s disease (PD) is classified as a movement disorder,
cognitive declines start early in the disease process. Difficulty with verbal memory is a
common clinical complaint of non-demented individuals with idiopathic PD but research
evidence of deficits is inconclusive. Parkinson’s disease pathology affects the medial
temporal lobe (MTL) early in the disease process. While past studies have found some
evidence of structural MTL changes in idiopathic PD, none to date have incorporated
both gray and white matter of memory-related brain areas, particularly through brain
MRI fiber tracking.
Methods: 40 non-demented individuals with idiopathic PD and 40 age, education,
and gender matched peers participated in this study. Participants received brain MRI
and verbal memory testing as part of a larger comprehensive neuropsychological
evaluation. Semantic language measures were included as control variables. Magnetic
resonance images were processed and analyzed using automated, semi-automated,
and manual methods. Statistical analyses included analysis of variance and
correlations.
13
Results: PD individuals had poorer list and story memory than age matched non-
PD peers. Processing speed accounted for some variance in memory performance but
did not fully explain deficits. There were no group language differences. While there was
heterogeneity in memory performance, 20% of PD participants had significant deficits
across two memory measures. There were also changes in left entorhinal volumes,
when controlling for total intracranial volume, with PD participants having entorhinal
volumes that were 11% smaller than Controls. Group differences were not seen in the
connectivity of the cingulum between the entorhinal cortex and the retrosplenial cortex.
Smaller entorhinal volumes related with poorer story memory recall and recognition.
Decreased cingulum connectivity related with worse verbal list recognition performance.
Discussion: While a majority of non-demented individuals with PD have intact
verbal memory, this study demonstrated that amnestic verbal memory declines occur
early in the process of PD for a subset of individuals. Further, changes are seen in the
medial temporal lobes (MTL) of individuals with PD, which adds to the evidence that
amnestic deficits exist in PD and are more common than generally believed.
14
CHAPTER 1 INTRODUCTION
Parkinson’s Disease
Parkinson’s disease (PD) is classified as a movement disorder; the initial and
typical clinical symptoms are most noticeably motoric with resting tremor, rigidity,
akinesia, or postural instability. There are, however, a number of autonomic, cognitive,
memory, and mood symptoms that are sometimes overlooked or are thought to be
unrelated to PD. Cognitively and clinically, it is frequently noted that individuals
diagnosed with PD self-report with bradyphrenia (reduced speed of thinking), difficulty
retrieving words, and difficulty learning or recalling new information. Complications with
memory and cognition in PD are documented throughout the literature (Naismith et al.,
2010; Sawamoto et al., 2007; Uc et al., 2005; Cameron et al., 2010; Stepkina et al.,
2010; Park & Stacy, 2009; Rodriguez-Ferreiro et al., 2010). Collectively, these
symptoms of PD place a burden on quality of life above and beyond the burden of
motoric symptoms (Hariz & Forsgren, 2010; Chen, 2010).
PD Prevalence and Impact
It is estimated that there are about 1 million individuals in the United States who
currently have Parkinson’s disease and up to 10 million worldwide (Parkinson’s Disease
Foundation, 2013). The lifetime risk of developing Parkinson’s disease is between 1-4%
with men at greater risk than women (Elbaz et al., 2002). With access to better
healthcare and nutrition, longevity has been increasing. It is expected that the number
of individuals with Parkinson’s disease will grow as the number of older adults around
the world increases due to increasing incidence of PD with age. Parkinson’s disease is
likely caused by an amalgamation of factors. There is evidence from twin studies that
15
genetics explain a small but significant portion of the variance in developing PD
(Wirdefeldt et al., 2011). While the precise cause of PD remains unknown researchers
have made progress in understanding the etiology and course of the disease.
PD Etiology
While generally thought as a dopaminergic disorder, the hallmark pathology of
Parkinson’s disease – Lewy neurites (LN) and Lewy bodies (LB) – forms first in the
olfactory bulb and medulla oblongata. The LNs then spread upward and outward
including the substantia nigra, basal forebrain, limbic system, and, in later stages,
throughout the neocortex (Braak, Ghebremedhin, Rüb, Bratzke, & Del Tredici, 2004).
LNs and LBs increase in density as PD progresses as well as if dementia develops.
However, severity of clinical symptoms does not correlate strongly with the severity of
PD pathology; instead, worsening clinical symptoms in PD might be linked with an
interaction between more Lewy neurites and greater neuronal loss (Halliday, Lees, &
Stern, 2011). Nevertheless, relatively little is known about this interaction between
cellular PD pathology and cellular loss. While PD pathology will not be directly
addressed in the current study, cellular loss will be measured.
Some have proposed that, because the olfactory system is affected in PD,
unknown environmental pathogens entering through the olfactory system play a role in
the development of PD (Hawkes, Shephard, & Daniel, 1999; Hawkes, Del Tradici, &
Braak, 2007). These pathogens might also enter the enteric nervous system via the
stomach and progress to the lower brainstem via cell-to-cell transmission (Hawkes et
al., 2007; Halliday et al., 2011). Even if specific environmental and biological factors
resulting in PD are unknown, it is believed that a mixture of biological and environmental
factors seems to play the largest role in who develops PD (Wirdefeldt et al., 2011).
16
Differences in mixtures of biological and environmental factors might lead to differences
in presentation of Parkinson’s disease symptoms.
Cognitive Considerations in PD
In addition to the core motor, autonomic, emotional, and cognitive symptoms of
Parkinson’s disease, many patients with PD go on to develop dementia. Parkinson’s
disease patients are at up to a six fold risk compared to peers without PD for developing
a comorbid dementia, such as Alzheimer’s disease, Lewy Body dementia, vascular
dementia, or Parkinson’s disease dementia (Aarsland et al., 2001; Aarsland et al.,
2005). There are estimates that more than 35 million individuals worldwide are currently
living with dementia, with that number predicted to top 115 million in the next 40 years
(Wimo & Prince, 2010). Because individuals with PD are at an increased risk of
developing dementia, which worsens already impaired quality of life, it is imperative to
gain greater understanding of cognitive and neurobiological changes in PD and the
overlap between or comorbidity of PD and dementia.
Memory, cognitive, and functional impairments start relatively early in the disease
process and worsen if a dementia develops. This places a burden on clinicians and
researchers to detect these cognitive and memory problems early so that individuals
with PD can receive necessary treatment that will improve their quality of life. The
quality of life of PD patients is affected early by their motor symptoms, which are
functionally impairing. Much is known about the neurobiology of motor symptoms in PD,
however, not as much is known about the underlying neuroanatomical substrates of the
cognitive and memory symptoms.
The motor symptoms of PD are associated with a loss of dopaminergic neurons in
the substantia nigra pars compacta located in the mesencephalon, which is a
17
phylogenetically ancient area of the brain located in the brainstem. This loss of
dopamine affects many areas of the brain, including deep brain nuclei, such as the
caudate and putamen, which are part of the basal ganglia and are involved in motor
functions among other abilities. Cortical areas are also affected. Although dopamine
plays a dominant role in motor and cognitive slowing in PD, there are other
neurotransmitters and brain regions to consider in addition to dopamine and the basal
ganglia (i.e., the nigrostriatal pathway). Some of the other affected systems include
noradrenergic, serotonergic, and cholinergic structures and pathways (Jellinger, 1991).
A significant amount of serotonin is produced in the brainstem and much acetylcholine
is produced in the basal forebrain; both brainstem and basal forebrain are affected early
in the Parkinson’s disease process. In fact, serotonin and acetylcholine deficits can
occur earlier than or concurrently with dopamine deficits (Halliday et al., 2011).
Serotonin and acetylcholine are important in such functions as emotional regulation and
cognition. Not all individuals with PD experience the same amount of pathological
changes. There is growing evidence that Parkinson’s disease is not entirely a
homogenous disease but one with a range of affected autonomic, motor, cognitive, and
memory symptoms (Kehgia et al., 2010; Kim et al., 2009).
Memory for Verbal Information
Clinically, individuals with PD often report concern about memory and word
retrieval problems. While deficits in other cognitive functions might be interpreted by
individuals with Parkinson’s disease as “memory problems”, empirical studies show that
verbal memory impairments do occur early in the course of idiopathic PD (Baran et al.,
2009; Davidson et al., 2006); however, reasons for these difficulties remain unclear.
When assessed clinically, verbal memory difficulty in patients with PD is not
18
homogenous. Some patients do not report any significant memory problems while
others report difficulty remembering names or conversations or items on a “to-do” list.
This variability in memory difficulty has also been seen in research populations (Filoteo
et al., 1997; Weintraub et al., 2004).
Part of the reason that there can be variability in memory difficulties is the
complexity of memory. In order to remember verbal information, an individual needs to
pay attention, understand what is being said or read, and process information quickly
(particularly if the information is unstructured) while simultaneously organizing and
consolidating the information into intact storage systems. Faulty memory can occur
when there are problems in any of these areas. This also means that disentangling what
might be ‘pure’ memory difficulties from memory difficulties due to other processes is
difficult. A better understanding of the foundations and concepts of memory can
elucidate the nature of memory deficits when they occur.
Cognitive and Biological Substrates of Memory
Memory is a multi-faceted and high level component of brain function. The ability
to learn and remember starts at a basic level in the brain; each neuron changes in
response to use (Hebb, 1949). The phrase “neurons that fire together wire together”
explains that learning and memory have a biological foundation at the neuronal level;
this type of neuronal change with experience has been termed Hebbian learning after its
main proponent. When two or more neurons fire in response to a stimulus, this paired
association is strengthened, making it more likely that the next time one of them fires,
the other one will fire as well. It can thus be said that the neurons ‘learn’ and
‘remember’. This type of cellular learning is the foundation of neuroplasticity, which
simply means that the brain changes with experience. This plasticity of the brain is the
19
neurobiological reason humans and all other animals with a nervous system can learn
and remember.
In order to understand memory better, a few terms will be provided and defined.
Learning is the process of acquiring novel information whereas memory is the
persistence of that acquired information over time with the goal that it can be accessed
for further use. In order for information to be learned and become a memory, there are
three general processes worth noting: encoding, storage, and retrieval. Encoding is the
process of transferring information to memory. During the encoding phase, information
is acquired – perceived and manipulated – and consolidated – information is stored and
processed further such that its representation in storage is strengthened. There is
considerable evidence that the encoding, particularly the consolidation, of memory
occurs in the medial temporal lobe (MTL) with the entorhinal cortex and hippocampus
the chief structures of this area (Sara, 2000). There is, however, evidence that other
brain regions – such as the frontal lobes – are active during and important for memory
encoding (Stebbins et al., 2002).
Storage is thus the ‘library’ that holds all of the encoded information for permanent
retention. Storage of memories is believed to be widely distributed throughout the
cerebral cortex including the frontal lobes but is still largely dependent on the medial
temporal lobe (Smith & Squire, 2009).
The last part of the process of the memory system is retrieval, which is acquiring
information from storage for active use (Gazzaniga, Ivry, & Mangun, 2002). While
diverse brain structures are involved in retrieval, including the medial temporal lobe,
prefrontal cortex, and parietal regions (Clement, Belleville, & Mellah, 2010), memory
20
retrieval is mediated by and most strongly dependent upon the hippocampus and
entorhinal cortex (Muzzio et al., 2009; Meulenbroek et al., 2010; Hoscheidt et al., 2010).
Memory is often further reduced into three temporal types of memory: sensory,
short-term, and long-term (Anderson, 1976). Sensory memory is essentially everything
that the senses (e.g., vision or hearing) register in any given instant; this type of
memory lasts milliseconds to seconds. Information from sensory memory that is
attended to and selected for retention then moves into short-term memory, which lasts
seconds to minutes. Information can then be further consolidated into long-term
memory, which lasts minutes to an indefinite period. While important parts of the overall
memory process, sensory and short-term memory will not be discussed further because
they are not directly applicable to this project. It should be noted that there are
competing views and contradictory evidence for this model of the memory system but
addressing controversy in memory models is beyond the scope of the current project
(review for more information on models of memory: Aggleton & Brown, 2006;
Markowitsch et al., 1999; Cave & Squire, 1992).
Long-term memory (see Figure 1-1) is complex with at least two main distinctions
– declarative and non-declarative. These memory systems are separate and can be
separately affected by pathology or other damage (e.g., Roediger, 1990). Declarative
memory is memory for facts, figures, people, faces, names, events, and objects, among
other information. This is information that is usually explicitly learned and can thus be
actively and intentionally recalled. Even though there are both verbal (names, shopping
lists; Shallice & Warrington, 1970) and non-verbal (object or face recognition)
components of declarative memory, even non-verbal memory can be expressed, or
21
translated, into language and thus declared. For example, memory for faces is non-
verbal but an individual could describe a face using words, hence its classification under
declarative memory. Non-declarative memory is, among other things, memory for motor
skills such as playing tennis or the piano (procedural memory). Usually it is tasks and
skills that are implicitly learned and to which we generally do not have conscious
access. Components of this memory system usually cannot be expressed in words.
While various memory systems were briefly reviewed, the focus of this study is on
verbal memory so other types will not be explored further. Because memory is multi-
faceted, memory deficits can develop in a number of ways. In order to assess verbal
memory deficits, verbal memory tests were developed by researchers and clinicians.
Two common types of verbal memory tests are list-learning and story memory.
Tests of Verbal Memory
Verbal list-learning measures. Verbal list learning tests, such as the Hopkins
Verbal Learning Test – Revised (HVLT-R; Brandt & Benedict, 2001), the California
Verbal Learning Test – II (CVLT-II; Delis et al., 2000), or the Philadelphia (repeatable)
Verbal Learning Test (PrVLT; Price et al., 2009) are used both clinically and in research
settings. The tests are complex in that they include a set of indices used to decipher
whether learning problems are associated with working memory, encoding and/or
retrieval difficulties. Regardless of the specific test name (HVLT, CVLT, PrVLT) each
list-learning test includes three main components: a set of list learning trials (immediate
recall), a delayed recall trial, and a recognition component.
List learning trials allow for researchers and clinicians to assess encoding of the
test information. Specifically, researchers can look at the number of words retained over
the learning trials, or they can assess improvement over the trials to see if individuals
22
benefit from repeated exposure. Damage to the MTL has been associated with poorer
learning and encoding. In one study, researchers found that atrophy of the medial
temporal lobe, including the hippocampus, was correlated with total learning
performance even within a group of Alzheimer’s disease patients; those who had
smaller hippocampi and more medial temporal lobe atrophy consistently learned fewer
total words on a list-learning test (Wolk et al., 2011). Testing for information after a
delay (delayed recall) allows researchers to evaluate the consolidation and persistence
of learned information. The number of words recalled after a delay can be assessed; it
can also be compared to the number of words learned on any given learning trial. Doing
so provides a measure of the persistence of information in storage. Delis and
colleagues (2005) demonstrated that by looking at delayed recall performance, taking
into account how many extra words a person gives, it is possible to reliably differentiate
between people with more MTL and cortical atrophy (e.g., Alzheimer’s disease) and
those with subcortical atrophy (e.g., Huntington’s disease). People with better memory
not only remember more information but they are also able to inhibit incorrect responses
and thus not produce as many extra words (intrusions) during recall. In other words,
people with better memory not only recall more words but also make fewer errors during
a recall task. This is an ability that is heavily dependent on the MTL.
It has been demonstrated that the thickness of the cortex of the MTL is positively
correlated with delayed memory performance in Alzheimer’s patients. While the cortical
thickness of other temporal and frontal areas relates with working memory performance,
the thickness of cortical areas other than the MTL do not reliably correlate with delayed
memory performance (Wolk et al., 2011). This provides additional evidence that the
23
MTL is the key area of the brain responsible for the transfer of information into storage
as well as the later retrieval of that information from storage. Other researchers have
found that Alzheimer’s patients who have smaller left entorhinal cortices do worse on
delayed list-learning recall (Di Paola et al., 2007). Smaller hippocampal volumes have
also been associated with poorer delayed recall in non-demented, healthy older adults
(Price et al., 2010; Tupler et al., 2007). This is important evidence that gross volume
changes help explain variability in memory performance; that variability is linked with
gross volume changes is true within groups of healthy individuals or within groups of
neurologically compromised individuals.
The third main component of list learning measures is recognition testing. For the
recognition test, a larger group of words is read to participants, from which they need to
select the words they heard previously. Recognition tests are also heavily dependent on
the MTL; individuals who have Alzheimer’s disease and who have thinner MTL cortex
(with the implication of more atrophy) perform more poorly on recognition testing (Wolk
et al., 2011). There is, however, some controversy over the role the MTL, and
specifically the hippocampus, plays in recognition testing (Kramer et al., 2005) but the
preponderance of evidence (Brown, Warburton, & Aggleton, 2010; Gold & Squire, 2006)
supports the role of the hippocampus and related structures, such as the entorhinal
cortex, as necessary (but not necessarily sufficient) for the encoding and consolidation
(and retrieval) of information.
While the MTL structures are necessary for verbal memory, they are not sufficient
to explain all memory functioning. Brain areas, such as the frontal lobes, play important
roles in verbal memory. Damage to the frontal lobes or frontal connections often result
24
in working memory and executive (planning, organization, inhibition) deficits. On list-
learning tests, performance on initial learning trials is strongly affected by working
memory whereas later ones are more affected by long-term memory (Vakil &
Blachstein, 1993). Thus, even though the MTL is vital for memory performance, damage
or disruption to other brain areas will also affect memory. Individuals experiencing
memory deficits due mainly to MTL (and closely connected brain regions) damage (e.g.,
encoding and consolidation deficits) are said to have an amnestic disorder. In contrast,
individuals who experience memory problems due to frontal-subcortical areas and
connections (e.g., processing speed, working memory, and organizational deficits) are
described as having dysexecutive memory problems. These memory profiles will be
discussed more in depth later.
With typical clinical scores such as total learning, delayed recall, and recognition, it
is possible to start to investigate and differentiate amnestic and dysexecutive profiles.
Delayed recall and recognition scores are particularly sensitive to changes in the MTL
(Wolk et al., 2011; Brown, Warburton, & Aggleton, 2010; Gold & Squire, 2006). There is
evidence that initial learning is not as affected by hippocampal atrophy as delayed recall
and recognition are in individuals with dementia (Tanner et al., 2011) because it is more
dependent on working memory abilities (Vakil & Blachstein, 1993), which are widely
distributed throughout the brain.
Some verbal memory measures like the CVLT and PrVLT allow for in-depth
analysis of amnestic patterns. Specifically, both the CVLT and PrVLT contain five
learning trials, a separate interference trial where a new list of words is presented, a
short delayed free recall trial (e.g., “Tell me all the words you can remember from the
25
first list; the one I read several times to you.”), a short delayed cued recall trial, long
delayed (after 20-30 minutes) free and cued recall trials, and a recognition component
that includes foils from the interference trial. Whereas the CVLT was designed for broad
clinical and research use, the PrVLT was developed specifically for use with older adults
(Price et al., 2009).
Other list-learning indices have been associated with MTL changes. One index
has been referred to as a savings index (Lamberty, Kennedy, & Flashman, 1995; Welsh
et al., 1994). The savings index is a measure of information retained from the immediate
learning to the delayed time period (Lamberty et al., 1995). It is a measure of an
individual’s ability to retain learned information when factoring in how much the
individual initially learned (Filoteo et al., 2009). This savings score has been shown to
be sensitive to early memory changes in dementia (Welsh et al., 1994). Individuals who
‘save’ less learned information have a relatively rapid rate of forgetting. This quick
forgetting is largely affected by MTL pathology (Squire, Stark, & Clark, 2004) but might
also be affected by other brain regions (Filoteo et al., 2009). Overall though, how much
information is retained, or ‘saved’, is affected by the integrity of the MTL and as such,
the savings index is sensitive to MTL atrophy.
Scores from individual list-learning trials or indices can be used to differentiate
between MTL and frontal effects on memory. While the following index scores will not
be used for the main analyses of the present study, they are presented as examples of
the utility of using various list-learning index scores for differentiating between types of
memory impairment. For example, two of the trials on the CLVT and PrVLT can be
particularly useful for assessing for the presence or absence of amnestic difficulties:
26
cued recall and an interference trial. Following a free (non-cued) delayed recall trial, on
the CVLT and PrVLT, the participant is given a cued recall trial. Many list-learning tests
include words from different semantic categories, for example, places, animals, or
vegetables. On the cued recall trial, an individual is thus asked to tell the examiner all
the words that were in a particular semantic category. This serves as a cue that
provides context and structure to help organize memory recall. This can enhance recall
in situations where information is stored but there is difficulty retrieving it (Craik &
Tulving, 1975).
The interference trial is a learning trial with a novel list of words. It is usually given
just after the five learning trials for the first list and before a short delayed recall. Giving
this trial allows one to probe for the effects of working memory and the ability to inhibit
inappropriate responses. Thus structuring a list-learning test to include cued recall, an
interference trial, and improved recognition testing can be particularly helpful and allow
researchers and clinicians to better assess for the presence or absence of amnestic
difficulties (Davis et al., 2002). Specifically, individuals with medial temporal lobe
pathology make more intrusion errors on free and cued recall as well as during
recognition testing (Delis et al., 1991). This means that they have more of an
indiscriminant response set – they cannot differentiate between words from the learning
trials and distracter words. Assessing participants’ recognition discriminability is
particularly salient to the current study. This type of memory pattern (an indiscriminant
response set) occurs when there is more disruption or damage to the MTL. Price and
colleagues (2009) found that individuals with dementia with little evidence of subcortical
vascular disease (i.e., their dementia is thought to be mainly cortical and
27
neurodegenerative in nature) produced more cued recall intrusions as well as false
positives on recognition testing. While more is known about memory patterns in
dementia, particularly Alzheimer’s disease, within PD, relatively little is known about the
medial temporal lobe, particularly the entorhinal cortex, and associated white matter
pathways as they relate with verbal memory performance. A discussion of this will occur
later in this manuscript.
List-learning tests are sensitive to memory changes because they are generally
free from contextual and other associative information that can serve as semantic
support from which to pull information during recall (Lezak et al. 2004). There are,
however, at least two prominent problems with list-learning verbal memory tests for
assessing memory deficits in PD. One problem is that they are affected by speed of
processing (see for example, Diamond, Johnson, Kaufman, & Graves, 2008). Because
Parkinson’s disease typically results in bradyphrenia (Huber, Shuttleworth, &
Freidenberg, 1989; Rogers et al., 1987) – cognitive slowing – what looks like a verbal
memory encoding deficit might simply be caused by reduced processing speed and
working memory (Foster et al., 2010). This means that on memory measures that are
strongly affected by processing speed, such as list-learning tests, if there are
processing speed deficits, ‘pure’ memory difficulties might be overstated. One proposed
solution is to use multiple memory measures, especially ones that might not be as
affected by processing speed, such as story memory measures (Zahodne et al., 2011).
Another problem with using list-learning tests – and any neuropsychological test
for that matter – is that norms developed for the tests are not usually targeted to a
particular population, such as PD. This is an issue because the extent to which findings
28
can be generalized from one population to another can be limited by the nature of the
normative sample (Greenaway et al., 2009). When applying normative data from
broader samples of individuals who are not necessarily similar to the targeted sample, it
is possible to either overestimate or underestimate impairment. This is a problem that is
easily solved. A useful solution to alleviate this problem that minimizes unexplained
variance between groups is using matched control groups for the derivation of
normative data.
Story learning measures. In addition to list-learning measures, there are other
ways to assess verbal memory. Another common way to assess verbal memory is by
use of a story memory test. Briefly, a paragraph of prose is read to an individual who is
then asked to tell it immediately back to the examiner and then again after a 20-30
minute delay. Story memory tasks also include a recognition component where the
individual is asked to answer a series of simple yes/no questions about the story.
Performance on story memory tests is thought to be less affected by processing speed
and executive deficits than list-learning tests are; it has also been linked to the MTL with
more atrophy and pathology of the MTL associated with poorer story memory
performance on all indices of the tests – immediate recall (learning), delayed recall
(retention), and recognition (Jokinen et al., 2004; Berlingeri et al., 2008; see also Price
et al., 2010). Some have argued that it is important to have multiple measures of verbal
memory when assessing for memory impairment (Loewenstein et al., 2009). Having
impairment on both list-learning and story tasks provides stronger evidence of actual
memory difficulty than does having impairment on either alone (Zahodne et al., 2011;
29
Janecek et al., 2011). Both list-learning and story memory tests are useful for assessing
memory of individuals with Parkinson’s disease, particularly when combined.
Is there a Verbal Memory Impairment in Parkinson’s Disease?
It is believed that many verbal memory deficits in PD are associated with
impairments involving frontal system functions (attention, processing speed, and/or
working memory). These deficits are linked to dysfunction of circuits between the basal
ganglia and the frontal lobes (Tinaz et al., 2008; Saint-Cyr, 2003). These circuits were
first described in depth by Alexander, DeLong, and Strick (1986) as connecting the
basal ganglia with different parts of the frontal lobes with an organization dependent on
the function of the circuit; the different circuits are involved in motor, cognitive, and
emotional processes. It is believed that dysfunction in these circuits reduces memory
performance due to a deficit of processing speed and executive abilities (e.g., set
shifting, ordering; Zahodne et al., 2011).
When these frontal-subcortical circuits are disrupted, which occurs in PD, this can
affect the encoding and storage of information from an organizational level. Retrieval of
information from storage is also reduced; however, because information was encoded
(not necessarily as well as in an individual without PD but also not necessarily at an
impaired level) the information was processed and stored. With difficulty freely retrieving
information, an individual with frontal-subcortical dysfunction benefits from cuing
because the information was stored but the processes of searching and retrieving from
storage are not working flawlessly. In other words, when prompted with a cue (e.g.,
“Which of the following words did I previously read to you?” – a recognition task),
someone with frontal-subcortical dysfunction will typically benefit from that cue because,
in part, it reduces the effort needed to retrieve information. This has been interpreted as
30
indicative of a primary self-retrieval problem rather than one of primary encoding (Stone
et al., 1998; Kramer et al., 2005; Dobbins, Simons, & Schacter, 2004). In other words,
frontal-subcortical dysfunction disrupts memory more by reducing the ease of retrieval
than it does by reducing the ability to store and retain information, although the
efficiency of such processes are likely reduced.
On formal clinical testing such as with word list-learning, which is devoid of
significant structure, being able to impose structure through grouping similar words
together (semantic clustering) or applying mnemonic devices (memory strategies)
during the encoding phase improves performance (Weintraub et al., 2004). Individuals
with PD can learn information, just not as efficiently or as quickly as people without PD.
This pattern of memory deficit is a dysexecutive memory profile, which is believed to be
due to frontal network (frontal lobes plus connected subcortical structures) dysfunction,
slowed speed of processing, or executive control shortcomings (Baddeley & Wilson,
1988; Baddeley, Della Sala, & Spinnler, 1991; Stuss & Alexander, 2007).
Conversely, while frontal systems are involved in verbal memory performance of
individuals with PD, it is possible that frontal system dysfunction does not fully explain
memory deficits seen in individuals with PD; there is an argument that some of the
verbal memory deficits in PD patients might be related to compromised storage of
verbal information (Lee et al., 2010). This means, in part, that there is a breakdown in
the encoding process; thus, some information is not transferred to storage. This
breakdown in encoding goes beyond a reduction in the efficiency of the encoding
process – some information is simply not processed and maintained in storage. This is
what happens in disorders such as Alzheimer’s disease, which initially and primarily
31
affects the MTL (entorhinal cortex, hippocampus, and surrounding areas). People with
pure encoding deficits will have difficulty retrieving information from storage because the
information never was stored and so there is nothing to retrieve, or at least not as much
information as with healthy individuals. Additionally, learned information can rapidly
leave storage, which reduces the amount of information in storage that can be later
retrieved. With memory deficits, when cued with a recognition task there is little to no
benefit and performance is at or near chance level. This type of memory profile – poor
encoding, storage, and hence, retrieval because there is little information to draw from
storage – has been termed an amnestic profile (Delis et al., 1991).
Parkinson’s Disease is Heterogeneous
There is evidence that PD patients experience amnestic memory difficulties as
well as dysexecutive memory deficits or even no memory difficulties. When looking at
the progression of PD, there are PD patients who never develop dementia, some who
develop Parkinson’s disease dementia, Lewy body dementia, small vessel vascular
disease dementia, Alzheimer’s disease, or some other variety or mixture of dementia
pathologies and etiologies. While PD is not a dementia, individuals with PD are at an
increased risk of developing dementia (Aarsland et al., 2001). Even those who are non-
demented can start exhibiting cognitive and memory deficits early in the disease
process (Baran et al., 2009). It follows that because memory is multi-faceted and
dementias are varied, those with PD who will progress and eventually develop some
form of dementia will have cognitive deficits that progress along different paths and with
different trajectories.
Thus, it is not likely that all Parkinson’s patients, even ones who will not develop
dementia at some point, will exhibit the same memory deficits or even any significant
32
memory deficits, at least until late in the progression of PD. There is evidence that
supports this heterogeneity in memory; not all PD patients exhibit a primary retrieval
(dysexecutive) memory deficit pattern. There is evidence that a subset of individuals
with PD show evidence of an amnestic profile, such as is typical in damage to or
pathology of the MTL (Weintraub et al., 2004; Filoteo et al., 1997). Other idiopathic PD
patients also might have a mixture of amnestic and dysexecutive difficulties while others
might not have any obvious verbal memory deficits.
In one study where researchers used a list-learning test to classify types of
memory deficit in individuals with PD, the researchers found that 50% of PD patients did
not have significant memory deficits, 26% of PD patients had a dysexecutive profile
similar to that of Huntington’s disease and 23% had memory problems more similar to
the amnestic profile that patients with Alzheimer’s disease typically display (Filoteo et
al., 1997). These memory profile classifications were based on performance on the 16
item list-learning test (California Verbal Learning Test {CVLT}). Specifically, the
researchers looked at the number of words individuals learned, the number of errors
they made during recall (stating words not on the list), and how well individuals did on
recognition relative to free recall (to see if participants benefited from the recognition
trial). These particular scores were shown previously to be sensitive to memory deficit
differences between patients with Alzheimer’s disease and those with a prototypical
subcortical disease – Huntington’s disease. Individuals with Alzheimer’s disease do not
learn many words, they quickly forget the words they do learn, “recall” a number of
words that were not on the list, and do no better on recognition than they did on free
33
recall of words. Patients with Huntington’s disease, however, while doing poorly in total
learning, benefit from the recognition trial (Massman et al., 1992).
In the study by Filoteo et al. (1997), the researchers investigated whether or not
PD patients with amnestic (23% of their PD group) versus dysexecutive (26% of their
PD group) memory profiles also differed on other neuropsychological measures. There
was no between-group difference in language, verbal fluency, or frontal organizational
functions. The only test for which there was a difference was on a test commonly
associated with degraded semantic memory (semantic fluency), which is often impaired
in Alzheimer’s disease; PD patients with an amnestic memory profile produced fewer
words than PD patients with a dysexecutive profile produced. Even though there was a
reduction in semantic fluency in the amnestic group, there were no differences in
language, verbal (non-semantic) fluency, or frontal organizational skills. The
researchers interpreted their results as providing evidence that the amnestic PD
patients did not also have Alzheimer’s pathology. Thus, memory dysfunction within
idiopathic PD possibly varies independently of comorbid Alzheimer’s pathology.
However, this research has some limitations. The authors did not produce enough
evidence to entirely rule out early Alzheimer’s disease processes in the amnestic PD
patients. This is not necessarily a problem but it is unclear whether or not this
heterogeneity in memory performance in PD is due to PD pathology or concomitant
Alzheimer’s or other pathology. However, this is an issue that can only be addressed
through brain tissue biopsy. An additional limitation is that the authors did not include
story memory performance in their analyses, which would have helped their argument if
the amnestic PD group also had story memory deficits. In spite of these limitations,
34
these results provide evidence that there might be heterogeneity of memory deficits in
PD.
Another study revealed a similar memory pattern in PD patients with 48% with
unimpaired memory, 33% with impaired retrieval (i.e., dysexecutive memory), and 17%
with impaired encoding (i.e., amnestic deficits; Weintraub et al., 2004). Weintraub and
colleagues found that in a group of non-demented PD patients, those with dysexecutive
and amnestic memory profiles had similar free recall performance on the HVLT-R but
the dysexecutive group had significantly higher recognition scores and fewer intrusion
errors than the amnestic group had. Additionally, Weintraub and colleagues found a
significant association between a measure of executive function and impaired retrieval –
meaning that those who benefitted from the recognition component of the verbal
memory task tended to have more problems with their ability to organize, shift set, and
perform other executive tasks.
While verbal memory for word lists is thought to be more affected in patients with
PD, there is evidence that story memory performance is also affected in non-demented
PD participants (Beatty et al., 2003). Beatty and colleagues showed that non-demented
PD participants performed as poorly on a story memory test as they did a list-learning
test. Also in this study, the poor performance on story memory was found even when
controlling for a possible floor effect in test performance. Deficits in story memory are
more typical of amnestic verbal memory difficulties than of dysexecutive ones.
There is further evidence for amnestic memory problems in PD patients:
researchers have found recognition impairments in PD patients relative to controls
(Beatty et al., 2003; Beatty et al., 1989; Helkala et al., 1988). This means that verbal
35
memory deficits in non-demented PD patients might not be solely related to problems
with retrieval or executive dysfunction. That is a conclusion supported by the findings of
Filoteo et al. (1997) and Weintraub et al. (2004). Thus, while executive deficits affect
verbal memory performance in a proportion of PD participants, some verbal memory
deficits might also be attributed to other neuronal or neurodegenerative processes.
This concept of multiple memory profiles in non-demented idiopathic PD patients
is not without controversy but there is building evidence of heterogeneity of motor,
cognitive, and emotional symptoms in Parkinson’s disease (see for example, Hardy,
2010). Verbal memory is only one of the cognitive functions affected by Parkinson’s
disease. A look at some broader areas of cognition will provide additional evidence of
heterogeneity in Parkinson’s disease.
A Growing Concern: Mild Cognitive Impairment and Parkinson’s disease.
There is growing concern for the development of dementia in PD. It is widely accepted
that mild cognitive impairment (MCI) places individuals at greater risk of developing
dementia (Gauthier et al., 2006). Additionally, MCI is a heterogeneous condition with
some individuals having more memory (amnestic) difficulties, some having more
executive difficulties, some converting to dementia over time, others staying with mild
impairment, and others converting back to normal cognition (Douaud et al., 2011). It is
generally believed that those with amnestic MCI are more likely to develop Alzheimer’s
disease, those with dysexecutive MCI are more likely to develop a subcortical or
vascular dementia, and those with mixed presentations could develop either dementia
or a mixed presentation of dementias (Gauthier et al., 2006). There continues to be a
debate regarding the significance or relevance of mild cognitive impairment in PD.
36
In Alzheimer’s disease, there is evidence that memory difficulties and disruptions
to language are related. Individuals who have Alzheimer’s disease often have word-
finding difficulties, which difficulties are known as a fluent, anomic aphasia (Price et al.,
1993). While it is possible to have language difficulties without memory impairment, it is
common to have disruption to the language as well as the memory system with the
cortical atrophy common in pathologies such as Alzheimer’s disease or other amnestic
disorders (Cummings, 1986; Galton et al., 2000). It is thus reasonable to expect
changes to language (e.g., difficulty naming or reductions in semantic abilities) to be
comorbid with amnestic memory difficulties. While there can be disruptions to the
access of the semantic storage system with damage to subcortical systems such as is
common in Huntington’s disease and Parkinson’s disease, individuals with more cortical
changes have a breakdown in the semantic storage system itself, rather than difficulty
with accessing it (Randolph et al., 1993). This is due to the changes to the cortex,
including the temporal lobe. There can be, however, heterogeneity in cognitive deficits
in populations with neurological disorders.
There have been attempts to classify subgroups of PD with mild cognitive
impairment (MCI). Caviness and colleagues (2007) used a cut point of 1.5 standard
deviations when compared to age matched peers to classify PD participants as having
MCI. A deficit of 1.5 standard deviations on a neuropsychological test using a normally
distributed (Gaussian) curve, places an individual at about the 7th percentile. This
means that 93% of age-matched peers perform as well as or better than someone with
a score at least a 1.5 standard deviation deficit. A 1.5 standard deviation deficit is also a
typical clinical cutoff for classifying impairment.
37
Using this cutoff, Caviness and colleagues (2007) looked at five broad cognitive
domains: frontal/executive, amnestic (memory), visuospatial, attention, and language.
They found that 62% of participants had normal cognition in all domains, 21% had mild
cognitive impairment, and 17% had dementia. A clinical classification of dementia, from
a neuropsychological standpoint, usually requires impairments in activities of daily living
as well as cognitive and memory impairments. Of the 21% classified as PD-MCI, 22% of
those had only an amnestic deficit and 11% had multiple domains including amnestic
deficit (39% had only frontal/executive dysfunction and an additional 22% had multiple
domains but no amnestic memory difficulties). Thus, the authors found that there is
variability in cognitive and memory profiles of individuals with PD.
However, there are a number of potential criticisms of Caviness and colleagues’
study. First, there were only 18 individuals in the PD-MCI group (total sample N = 86),
which resulted in just 4 being diagnosed as having amnestic MCI (with an additional 2
who had amnestic MCI plus difficulty in one or more other domains). This works out to
7% of the broader sample being diagnosed with amnestic MCI (or amnestic MCI plus
other deficits). It is difficult to draw firm conclusions with a limited number of PD patients
having MCI and a smaller sample of those with amnestic MCI. It is possible that this
estimate is low because the authors assessed verbal memory using the auditory verbal
learning test (AVLT), which was not designed specifically for use with older adults
(Lezak et al., 2004) and might not be a particularly sensitive measure to assess for an
amnestic memory profile. Nevertheless, there is compelling evidence that there are
some PD patients who have amnestic memory problems (Janecek et al., 2011);
however, the extent to which amnestic deficits occur in PD is not fully clear.
38
Another limitation is the test indices used to examine MCI in PD. Caviness and
colleagues (2007) investigated only immediate and delayed recall, which might provide
a limited view of list learning (Price et al., 2009). They did not include recognition
memory, which might have changed some of the classifications away from amnestic
(i.e., if there were substantial improvements in recognition from recall or if there were no
deficits in recognition compared to controls). Further, the researchers used literature-
based normative values, which they state might offer better reliability and
generalizability but conversely, using literature-based norms rather than a closely
matched normative sample can limit the sensitivity and specificity of findings and
introduce uncontrolled variance into the results. In addition to this, Caviness et al.
(2007) did not include a story memory task, which when combined with list-learning
tasks, can improve classification of mild cognitive impairment (Rabin et al., 2009;
Loewenstein et al., 2009). In short, the study potentially has serious limitations but does
support the idea of memory deficit heterogeneity in PD.
Even given these limitations, we can tentatively conclude that in non-demented PD
patients, there is variability in cognitive deficits; this is clear when we compare the
results of the study by Caviness et al. (2007) to other similar studies (see for example,
Foltynie et al., 2004; Aarsland et al., 2010). There is further evidence that a significant
portion of PD patients might have amnestic memory difficulties. Other researchers
found in a large and well-controlled study that 14% of individuals with PD could be
classified as having amnestic MCI (Aarsland et al., 2010). Overall, while the precise
proportion remains unknown, this review of the literature establishes that it is likely that
between 7-22% of non-demented individuals with PD experience amnestic memory
39
deficits with somewhat higher proportions of individuals experiencing dysexecutive MCI
(e.g., Caviness et al., 2007; Filoteo et al., 1997). The variability in that range could be
due to different memory measures used, different cut-off points and criteria for defining
types of memory difficulties, different sample sizes, or differences in study inclusion and
exclusion criteria. Using more specific analyses targeted on drawing out amnestic
difficulties should allow for stronger conclusions to be made about how disruption to the
MTL affects memory in PD patients. In summary, there has been considerable and
building evidence in recent years to suggest heterogeneity in how Parkinson’s disease
affects cognition and memory (Jellinger, 2010).
Language and Semantic Knowledge in Parkinson’s Disease
Semantic knowledge is one component of language that has been shown to be
relatively intact in old age and even in some neurological disorders such as Parkinson’s
disease (Portin, Laatu, Revonsuo, & Rinne, 2000). Semantic knowledge, however, is
impaired in Alzheimer’s disease (Della Sala et al., 2004; Carew, Lamar, Cloud,
Grossman, & Libon, 1997). It has also been shown to be impaired in cognitively
compromised individuals with Parkinson’s disease (Portin et al., 2000; Bayles et al.,
2000). Wide networks of brain regions subserve semantic abilities (Catani & Jones,
2005; Friederici, 2009). For most individuals, such language functions are largely
lateralized to the left cerebral hemisphere, particularly to the perisylvian region (Catani
& Jones, 2005). Damage to any of these diverse brain regions can result in reduced
semantic functioning.
Focus on localized cortical regions started in the 1800s with case studies
published by Broca and Wernicke demonstrating deficits in language resulting from left
hemisphere brain damage (Geschwind, 1972; Geschwind, 1971). From these case
40
studies we learned that damage to left inferior frontal and temporal-parietal regions
resulted in language deficiencies of differing presentation. In brief, it is believed that
language comprehension is mediated primarily by the temporal lobe whereas language
production is more frontally mediated. However, that summary understates the
complexity of language and the brain networks that support language functions
including semantic knowledge.
While many aspects of language are preserved in the early stages of PD, there is
evidence that some language functions are affected in PD. Holtgraves and McNamara
(2010) found a deficit in pragmatic comprehension, which is language use in social
contexts. They found that these deficits (specifically, a deficit in speech act
comprehension, implying greater difficulty understanding the speech of others in social
contexts) were related to poorer performance on an executive function measure (Stroop
Color-Word interference task) and could not be fully explained by slowed processing
speed. Additionally, these social language difficulties were correlated with disease
severity. Pragmatic comprehension, however, is a complex form of language, involving
a number of other cognitive functions. As such, pragmatic comprehension is not
considered a pure language task (Zurif, 1980), especially compared to a simple
semantic retrieval task.
Grossman (1999) found that sentence comprehension deficits occur in PD. These
deficits were linked with selective attention difficulties, which are believed to rely heavily
on affected frontostriatal circuitry. Sentence comprehension requires holding in mind a
phrase including the connection between a subject, verb, object, and location. Thus,
sentence comprehension has a large demand on working memory. Sentence
41
comprehension is thus a fairly complicated form of language, affected by other cognitive
functions. Simpler forms of language, however, might also be affected in PD.
There is evidence of impaired confrontation naming in PD. Matison and colleagues
(1982) found a slight deficit on the Boston Naming Test in a group of 22 PD patients.
They called this a “tip of the tongue” syndrome or a “word production anomia”. Cooper
and colleagues (1991) found language, including confrontation naming, deficits in
individuals with PD; however, these only occurred in individuals with global cognitive
deficits. In other studies, however, researchers have not found impairments on semantic
knowledge tasks, such as confrontation naming, in idiopathic, non-demented PD
(Tröster, Stalp, Paolo, Fields, & Koller, 1995; Pagonabarraga et al., 2008).
Generally, the accepted view about PD is that any language difficulties observed
in PD are not similar to aphasias seen with focal or diffuse left hemisphere lesions
(Grossman, 1999) or those that occur with cortical dementias, such as semantic
dementia or Alzheimer’s disease. Thus, it is believed that most language difficulties in
non-demented PD are due to frontal-striatal network deficiencies, which result in slowed
processing speed, attention deficits, working memory difficulties, and executive
dysfunction (see Bastiaanse & Leenders, 2009 for a review). This implies that brain
regions and networks directly supporting language functions including semantic
knowledge are likely not affected in PD, at least early in the disease process.
Language and semantic brain networks. Production of spoken words, or even a
single vowel or consonant sound, requires a broad network within the brain (Sörös et
al., 2009). This network relies heavily on cortical and white matter areas surrounding the
left Sylvian fissure for most right-handed individuals (Catani, Jones, & ffytche, 2005;
42
Turken & Dronkers, 2011). It involves frontal, parietal, and temporal cortical regions with
underlying white matter as well as subcortical structures, such as the thalamus, also
playing a role (Crosson, 1999). The long temporal to frontal white matter circuit that
serves as a major language pathway is called the superior longitudinal fasciculus (SLF)
/ arcuate fasciculus (AF). There is some controversy whether or not the arcuate
fasciculus is merely part of the SLF or if it is a separate, but closely related pathway
(Duffau, 2008; Bernal & Ardila, 2009; Catani & Thiebaut de Schotten, 2008). For
simplicity, the SLF and AF will be grouped together for this paper and referred simply as
the arcuate fasciculus because the term fits the curved nature of the white matter
bundle in the perisylvian region.
The AF can be separated into three regions – an anterior, which connects the
frontal (Broca’s territory) with parietal regions (Geschwind’s area), a posterior segment,
which connects parietal (Geschwind’s territory) and temporal cortex (Wernicke’s
territory), and a long segment, connecting frontal cortex with temporal cortex (Catani et
al., 2005). The AF is believed to be important for aspects of language functioning and
disruptions to any part of the network can result in changes to language functions. It is,
however, a complex fiber pathway with bidirectional connections between frontal –
parietal and frontal – temporal regions (Duffau, 2008). While gross changes to the AF
are not reported in Parkinson’s disease, there is evidence of microstructural white
matter changes in the superior longitudinal fasciculus relatively early in the Parkinson’s
disease process (Gattellaro et al., 2009), although these changes were not specifically
localized to the AF portion of the superior longitudinal fasciculus. The AF and its
function will be discussed more in depth later in this document.
43
Considerations for Neuroanatomical Predictors of Verbal Memory Impairment in PD
Idiopathic Parkinson’s disease (PD) affects not just the functions (e.g., cognition,
memory, movement, and emotion) produced by the brain but also the structure of the
brain. Neuroimaging, especially magnetic resonance imaging (MRI), allows for the
investigation of the structure of the gray and white matter of the brain. Magnetic
resonance image analyses allow us to better understand the neuroanatomical
correlates or predictors of cognitive and memory processes. One type of magnetic
resonance imaging that has grown considerably in research and clinical use is diffusion
weighted imaging. Diffusion MRI data can be used for elegant visualization and
quantification of the white matter in the brain. This permits better understanding of not
just the discrete and localized neuronal areas involved in memory and cognitive
processes but also the distributed networks connecting these areas. In other words, no
part of the human brain is an island just as no function of the brain is completely
independent of other functions.
The focus of this study is on how memory performance relates to the structure of
both gray and while matter regions of the brain. Gray matter regions implicated in
memory include the entorhinal cortex, which is part of the medial temporal lobe. White
matter regions involved in memory processes are widely distributed but include, among
other areas, the cingulum bundle. The focus in the present discussion will first be on the
entorhinal cortex and then on the cingulum, particularly the posterior portion of the
cingulum that connects to the entorhinal region.
44
Entorhinal Cortex and Medial Temporal Lobe
In idiopathic, non-demented Parkinson’s disease, there is evidence for
hippocampal atrophy relative to controls (Laakso et al., 1996; Jokinen et al., 2009). The
hippocampus is heavily connected with the entorhinal cortex, which has been shown to
atrophy prior to the hippocampus in PD and Alzheimer’s patients (Jokinen et al., 2009;
Choo et al., 2010). In general, the entorhinal cortex is the site of the earliest disruption
in MTL atrophy and is possibly the cause of most of the memory changes seen in
amnestic disorders because it disconnects the hippocampus from the neocortex
(Insausti, 1993). This makes the entorhinal cortex an important structure for detecting
early changes that can result in memory deficits. Both structures are part of the MTL
(see Figure 1-2).
Historically, little was known about the functions of the MTL until James Papez
(1937) proposed that the MTL was involved in emotion. This created additional interest
in the MTL but relatively little was discovered about its role in cognitive and memory (or
emotion, for that matter) processes until the 1950s. The medial temporal lobe became a
focus for memory research following the seminal article by Scoville and Milner (1957)
about Scoville’s patient Henry Molaison (H.M.), who had his most of his hippocampi and
amygdala plus surrounding parahippocampal gyri removed bilaterally as part of an
experimental treatment for epilepsy. Following the procedure he experienced severe
anterograde amnesia. This study and numerous subsequent studies demonstrated the
importance of the MTL for the formation of semantic memory (Wolk et al., 2010), which
is a form of declarative (explicit) memory. Patients with focal MTL damage have intact
attention, working memory, visoperceptual skills, implicit memory, language, semantic
knowledge, and global intelligence (Squire et al., 2006).
45
The medial temporal lobe is comprised of the hippocampal complex – the
subiculum (the main output of the hippocampus), dentate gyrus, CA1 and CA3 fields of
the hippocampus (regions CA2 and CA4 also exist but those are generally not included
in connectional mappings of the hippocampus) – as well as the entorhinal, perirhinal,
and parahippocampal cortices.
As is shown in Figure 1-3, the entorhinal cortex is the main afferent and efferent
cortex of the MTL. It essentially controls the flow of information to and from the
hippocampus (Schwarcz & Witter, 2002). The main white matter pathway from the
entorhinal cortex into the body of the hippocampus (mainly the molecular layer of the
dentate gyrus; Insausti, 1993) is called the perforant path, which has been shown to
show degradation in older individuals with and without dementia (Yassa, Muftuler, &
Stark, 2010). Disruption to any part of these circuits has the potential to affect all parts
of the circuit.
Medial temporal lobe neuropathology in PD. Braak et al. (2004) proposed a
series of stages of PD pathology. They stated that Lewy neurites are initially found in
the medulla oblongata and olfactory bulb. From the medulla, the Lewy neurites spread
upward into the substantia nigra and then to the allocortical areas of the medial
temporal lobe and basal forebrain, disrupting both the dopaminergic and
acetylcholinergic systems of the brain. Most of these pathological changes occur prior to
phenotypical and clinical expression of motor symptoms. As symptom severity
progresses, Lewy neurites and Lewy bodies continue to aggregate in affected areas
while also spreading throughout the rest of the cortex, focusing particularly on lightly
myelinated or unmyelinated neurons with narrow axons (Braak et al., 2004).
46
A recent neuropathological study was focused on the presence of -synuclein
(Syn), tau, and amyloid (A) peptide in the brains of PD patients with and without
dementia. All PD brains showed Syn and A in the entorhinal cortex as well as a
number of other brain regions; these concentrations of pathologic proteins increased in
those who had dementia but were significant in non-demented PD. These results
indicate that even in cognitively intact PD patients there is widespread cortical pathology
in addition to the prevalent brainstem and subcortical changes (Kalaitzakis et al., 2009).
Micrograph investigations of pathological changes in the brains of PD patients
have found extensive Lewy neurites and Lewy bodies in the transentorhinal region,
which is the primary immediate entry into the main body of the entorhinal cortex. This
disruption to the entorhinal cortex is greater than that seen in the anterior cingulate
cortex, which also shows significant changes in PD patients. If pathology is related with
cognitive and behavioral functioning, this pattern of pathology seems to indicate that
memory disturbances might be at least as common as the well-recognized affective
(e.g., apathy) disturbances. Additionally, there are Lewy neurites prevalent in the cornu
ammonis (CA) of the hippocampus in PD patients (Braak & Braak, 2000). These
neurites are clusters of proteins (primarily -synuclein) that are common in other
neurodegenerative disorders, including Alzheimer’s disease (Marui et al., 2004). In
Alzheimer’s disease, like in PD, Lewy neurites are common in the cornu ammonis of the
hippocampus. This overlap in pathology between PD and AD might result in somewhat
similar cognitive deficits.
These results demonstrate considerable progressive limbic system changes that
occur in PD that could disrupt both cognitive and emotional functions. Specifically, the
47
entorhinal cortex is a sensitive predictor of early memory changes (Stoub, Rogalski,
Leurgans, Bennett, & deToledo-Morrell, 2010; Braak & Braak, 1991). In a recent fMRI
study, Brickman, Stern, and Small (2010) sought to associate distinct aspects of
memory with blood flow (i.e., cerebral blood volume) in different components of the
MTL. The researchers found a dissociation between the entorhinal cortex and the
dentate gyrus for delayed recall and delayed recognition. Specifically, on a verbal list
learning test, the authors found that cerebral blood volume in the entorhinal cortex was
related to delayed free recall and retention performance.
There is additional evidence that memory tasks utilize the hippocampal region of
the brain. Performance on a visual recognition task (part of the Benton Visual Memory
Test) was correlated with blood flow in the dentate gyrus. The authors suggest that their
results contribute to evidence that the entorhinal cortex is involved in maintaining
representations in memory over a delay. This means that disruptions to the entorhinal
cortex should result in faster decay of learned information. The entorhinal cortex is also
involved in encoding during memory tasks as well as delayed cued recall of information
(Fernandez, Brewer, Zhao, Glover & Gabrieli, 1999). Other research demonstrates a
similar involvement of the entorhinal cortex in memory processes (Coutureau & Di
Scala, 2009; Martin et al., 2010) and that the entorhinal cortex might be the primary site
of hippocampal region dysfunction in Alzheimer’s disease (Reitz et al., 2009; Insausti,
1993).
The Cingulate in PD
The hippocampus and associated medial temporal lobe structures (entorhinal,
perirhinal, and parahippocampal cortices) are merely one group and part of a distributed
network. Another part of this network is the cingulum, a bundle of white matter fibers
48
that travels anterior and posterior in the brain just dorsal to the corpus callosum and just
beneath the cingulate cortex (view Object 1-1 for a video introduction to the cingulum).
Object 1-1. Supplemental video created and narrated by Jared Tanner giving an introduction to the cingulum; work supported by NINDSK23-060660 (YouTube link: http://www.youtube.com/watch?v=8TAmyOAkCz8)
The cingulum travels from the basal forebrain above the cribiform plate of the skull
to curve around and travel to the pole of the temporal lobe. The cingulum roughly
parallels the path of the fornix – another pathway important in memory processes – and
in fact, a portion of the cingulum and the crus of the fornix meet near the splenium of the
corpus callosum where together they travel to and along the inferior hippocampus – the
pes hippocampi (Schmahmann & Pandya, 2006). Further, the cingulum is
interconnected with the uncinate fasciculus in both the frontal and temporal lobes,
making what researchers have called a “limbic ring” travelling the circumference of the
entire limbic system (Yakovlev & Locke, 1961). This limbic ring allows information from
distributed brain regions to travel to and from the MTL in order to affect the creation and
utilization of stored information, among other cognitive functions. While there are other
pathways involved in memory, this limbic ring and associated pathways and structures
is vital for normal memory functioning.
Cingulate anatomy. The cingulate (cingulate cortex and cingulum) is involved in
multiple functions, including motivation, emotion, and memory; the cingulate is the major
functional area of the limbic system. One function of the cingulum is to connect the
cingulate cortex and septal nuclei with the parahippocampal gyri (Stadlbauer et al.,
2008; Yasmin et al., 2009), of which the entorhinal cortex is a part, as well as to connect
various brain regions to the MTL in general (Choo et al., 2010; Stadlbauer et al., 2008;
Yasmin et al., 2009). Many of the fibers in the cingulum are bidirectional, creating loops
49
in connectivity; this means that the cingulum is not just carrying information to the MTL
but also from it to other brain regions. The cingulate is part of network that James
Papez initially proposed as an emotion circuit (Papez, 1937) but later research has
shown that it is necessary for intact episodic memory (Zola-Morgan & Squire, 1993).
Damage anywhere along the circuit can disrupt normal memory performance (Budson &
Price, 2005). The Papez circuit (Figure 1-4) is roughly the following: hippocampus –
fornix – mamillary bodies – mamillothalamic tract – anterior nucleus of the thalamus –
cingulate cortex – cingulum (Papez, 1937).
Retrosplenial cortex. The cingulate cortex also receives input from diverse
association cortices, including the prefrontal cortex and the caudate nucleus. One area
of the cingulate implicated in memory is the retrosplenial cortex (and adjacent posterior
cingulate cortex). The retrosplenial cortex is one of the main areas of cortex providing
and receiving connections to and from the entorhinal cortex (Kobayashi & Amaral,
2003). Up to 20% of the connections to the entorhinal cortex in a monkey are from the
retrosplenial cortex (Insausti, Amaral, & Cowan, 1987). In the entorhinal cortex, the
retrosplenial cortex connects via the cingulum (see Figure 1-5) mainly to the caudal
portion in an overall topographical arrangement (e.g., lateral retrosplenial cortex
connecting more laterally in the entorhinal cortex and the caudomedial aspects of
retrosplenial cortex connecting more medially in the entorhinal cortex; Kobayashi &
Amaral, 2003). It is believed that these areas are similarly connected in humans.
Additionally, the retrosplenial cortex merges with the parahippocampal gyrus ventral to
the corpus callosum. These close integrations with MTL structures imply a close
functional relationship with areas pertinent to memory functioning.
50
Damage to the retrosplenial cortex can result in severe anterograde and
retrograde amnesia (Valenstein et al., 1987). In the case study by Valenstein et al.
(1987), the authors reported on a 39-year-old male who experienced an intracerebral
hemorrhage just lateral to the left splenium. He additionally had an arteriovenous
malformation surrounding the splenium in the pericollosal cistern. This hemorrhage
resulted in the destruction of the retrosplenial cortex as well and the underlying
cingulum. He experienced retrograde amnesia with impairments in episodic events of
up to 4 years prior to his hemorrhage. He also had severely affected anterograde
amnesia, especially for verbal information; nonverbal memory was relatively spared.
There is further evidence of the importance of the retrosplenial region in the
formation and retrieval of memories. The retrosplenial cortex is heavily connected with
the anterior and anterodorsal nuclei of the thalamus, which are key structures for
memory processes (Vogt et al., 1987; Papez, 1937). Additionally, the retrosplenial
cortex is a major termination site for the outflow of acetylcholine from the basal forebrain
via the cingulum (Selden et al., 1998). The basal forebrain is affected by PD pathology,
relatively early in the disease process (Braak et al., 2004), suggesting a potential
disruption of acetylcholine input into the retrosplenial area in individuals with PD. The
ties between acetylcholine, memory, and cognition are well-documented if not always
straightforward for clinical treatment (Mesulam, 2004; Levin & Simon, 1998; Calabresi,
Picconi, Terry & Buccafusco, 2003; Parnetti, & Di Filippo, 2006; Trinh, Hoblyn, Mohanty,
& Yaffe, 2003).
There is evidence that some of the earliest signs of metabolic decline in
Alzheimer’s disease occur in the retrosplenial cortex, even prior to metabolic changes in
51
the parahippocampal region (Minoshima et al., 1997; Villain et al., 2008; Vann,
Aggleton, & Maguire, 2009). These metabolic changes are separate from the gray
matter atrophy that early on and primarily affect the MTL in individuals who later
develop amnestic memory disorders, such as Alzheimer’s disease (Chételat et al.,
2003). It is believed that local structural and biochemical as well as distal changes result
in the early retrosplenial and posterior cingulate hypometabolism. It is clear that the
retrosplenial cortex, underlying cingulum, and hippocampal region all play important
supportive roles in memory processes in the normally functioning brain as well as in
dysfunction.
Cingulum. The broader cingulate area is involved in multiple functions as it spans
much of the anterior-posterior length of the human brain. The cingulum carries the
connections to and from areas of the cingulate cortex. As a result of its diverse functions
and connections, it has fibers entering from and exiting to various areas of association
cortices along its course (Schmahmann & Pandya, 2006). Thus, damage to the broader
cingulate and underlying cingulum might indirectly as well as directly affect memory,
especially if damage is in the posterior portion of the cingulate (Valenstein et al., 1987).
The cingulum is further important in this memory circuit because, as mentioned
earlier, it serves as one of the primary afferent pathways into the entorhinal cortex and
associated MTL structures. Papez (1937) reported that in a monkey brain, the posterior
cingulum fibers are the prominent bundle that enters the hippocampus. While there are
fibers that connect directly to the hippocampal formation via the cingulum, fibers
connect to the pre-subiculum, subiculum, and, more importantly, to the entorhinal
cortex; these fibers then enter the hippocampus via the perforant path. The entorhinal
52
and perirhinal cortices are immediately connected to the dentate gyrus of the
hippocampus and thus serve as the primary input into the hippocampus. Thus, damage
to the entorhinal cortex disconnects the hippocampus from limbic structures, such as
the cingulate, as well as other association cortices (Salat et al., 2010).
Cingulate changes in neurodegenerative disorders. Choo and colleagues
(2010) demonstrated posterior cingulate changes seen on diffusion MRI in MCI and AD
patients. Karagulle Kendi et al. (2008) revealed decreased fractional anisotropy (FA) in
the anterior cingulum of PD patients relative to controls. However, Gattellaro et al.
(2009) found only altered mean diffusivity (MD) and not decreased FA in the cingulum
of PD patients relative to controls. Thus, there is tentative evidence for altered integrity
of the cingulum in PD patients. In addition to gross white matter changes there are
changes to the gray matter of the cingulate.
Cingulate cortex changes in PD. There are clearly widespread neurobiological
changes in PD patients even without dementia. However, research focused on
structural brain changes in PD participants is sparse. There is evidence of gross
cingulate change in participants with mild cognitive impairment; such impairment might
be seen in Parkinson’s disease. Structural MRI studies of participants with amnestic
MCI as well as those in the early stages of AD have shown reductions in both PCC and
ACC volumes. Reduced cingulate volumes are useful in discriminating between MCI
patients who do and do not go on to develop AD (Salmon & Laureys, 2009). While
changes to the white matter can occur prior to gray matter changes, loss of white matter
will cause changes to the connected gray matter and loss of gray matter will result in
loss of connected white matter (Agosta et al., 2011). Thus, with loss of cortex, there
53
likely is also a concordant white matter loss, although the relationship is not necessarily
direct or strong. It can then be assumed that with cingulate cortex atrophy, there will be
a partial reduction in the integrity of the underlying cingulum bundle. Whether or not a
similar pattern of atrophy and changes in non-demented idiopathic Parkinson’s disease
patients can be seen is one of the goals of this present investigation. So far there
appear to be neuroanatomic changes to the cingulate in PD but adding to this is
evidence of neurochemical changes.
Dopamine and the cingulate. The cingulate cortex receives considerable
dopaminergic input from the substantia nigra and ventral tegmentum. Reductions in
dopamine (DA) in Parkinson’s disease directly and indirectly affect the functioning and
structure of the cingulate; however, much of the DA input to the cingulate flows to the
rostral cingulate premotor area, so the DA reductions might lead to DA-related motor
symptoms more than DA-related cognitive symptoms (Vogt, Vogt, Purohit, & Hof, 2009).
Dopamine depletion is only one of a number of neurobiological changes to the cingulate
in Parkinson’s disease and other neurologic diseases. Like the hippocampus, the
cingulate is also affected by Lewy bodies in idiopathic Parkinson’s disease, Alzheimer’s
disease, Parkinson’s disease dementia (PDD), dementia with Lewy bodies, and even
‘normal’ aging (Mattila et al., 2000; Vogt et al., 2009). In fact, the cingulate gyrus seems
to be particularly vulnerable to Lewy body aggregation, especially in patients who are
progressing towards or have PDD or dementia with Lewy bodies (Braak et al., 2004). In
PD patients, there is evidence that the number of Lewy bodies in the cingulate gyrus
correlates with cognitive impairment as measured by criteria set by the Consortium to
Establish a Registry for Alzheimer’s Disease (CERAD) criteria (Mattila et al., 2000). One
54
caveat is that this relationship was based on a non-specific global scale of cognitive
impairment and thus might have poor specificity for memory, although memory
generally declines with global cognitive decline. Another caveat about applying those
findings to the present study is that DA and Lewy body changes are not directly
measurable with current structural MRI technology and thus might not result in gross
structural changes. However, what is important about Mattila and colleagues’ (2000)
findings is that they provide additional evidence for a link between cingulate changes
and cognitive impairments.
Acetylcholine in PD
A group of neurons in the basal forebrain called the nucleus basalis of Meynert
(nbM) contains an aggregation of cholinergic cells that project widely to the neocortex.
The basal forebrain also projects directly to the hippocampus through the fornix (Selden
et al., 1998). Because of this, acetylcholine is thought to be involved in memory
processes. Additionally, patients with dementias, including Alzheimer’s disease and
Lewy body dementia, are given acetylcholinesterase inhibitors, which have been shown
to be modestly effective in temporarily reversing some of the memory deficits (Pepeu &
Giovannini, 2010). Patients with Alzheimer’s disease show reduced activity of choline
acetyltransferase (ChAT) throughout the brain, including the cortex, striatum, and
hippocampus (Bartus et al., 1982; Zola-Morgan & Squire, 1993). In addition to
Alzheimer’s disease patients, choline acetyltransferase activity is also reduced in
patients with PD both without and with dementia (Mattila et al., 2001; Dubois et al.,
1983; Whitehouse et al., 1988) and even in the absence of AD pathology (Perry et al.,
1985). Parkinson’s disease patients without dementia had reductions in choline
acetyltransferase in parietal and occipital cortices. Parkinson’s disease patients with
55
cognitive impairments had greater loss of ChAT activity throughout the cortex except
occipital cortex compared to non-cognitively impaired PD patients. Parkinson’s disease
patients with concurrent pathologically-confirmed Alzheimer’s disease had more ChAT
loss only in the entorhinal cortex compared to cognitively impaired PD patients even
though their degree of cognitive impairment was greater than cognitively impaired PD
patients without AD. Demented (not AD) and non-demented PD patients also had
reductions of acetylcholinesterase throughout the cortex with greater reductions seen in
PD patients with dementia than those without dementia (Perry et al., 1985).
Using positron emission tomography (PET) and single photon emission computed
tomography (SPECT), clinicians and researchers are able to study neurochemical
changes in vivo. In non-demented idiopathic PD patients, there were reductions in the
SPECT vesicular acetylcholine transporter (VAChT) ligand [I-123]-IBVM in the parietal
and occipital cortices; however, in demented PD patients, there were greater and more
widespread deficits. Additionally, acetylcholine esterase (AChE) activity and levels are
significantly reduced in PD patients without dementia with further reductions in
demented PD patients (Bohnen & Albin, 2010). Loss of cholinergic neurons in the basal
forebrain as well as reduced ACh activity in the cortex can serve as markers for the
severity of cognitive disturbances in PD.
With much of the production of ACh in the basal forebrain, neuronal degeneration
of the basal forebrain will reduce ACh levels and affect related proteins. In Parkinson’s
disease there is evidence of degeneration of the nucleus basalis of Meynert (particularly
marked by an accumulation of -synuclein in the basal forebrain) early in the disease
process and even in the absence of dementia (Bohnen & Albin, 2010; Braak et al.,
56
2004). Kalaitzakis and colleagues (2009) found high concentrations of Syn pathology
in the nucleus basalis of Meynert in both demented and non-demented PD participants.
The authors believe these findings point to a universal cholinergic deficit in PD. Thus, it
is possible that some cognitive deficits – especially memory dysfunction – in PD are
associated more closely with basal forebrain degeneration (and concurrent depletion of
acetylcholine) than with substantia nigra deterioration and dopamine deficits. The
implication from this is that memory difficulties in PD appear to be at least partially
‘cortical’ rather than completely ‘subcortical’.
One area of the brain ACh deficits affect is the cingulate gyrus. The medial
cholinergic pathway travels out from the nbM primarily through the cingulum with
terminations throughout cerebral cortex, including the cingulate gyrus and retrosplenial
cortex (Selden et al., 1998). Thus, degeneration of the nbM might lead to reductions in
the integrity of the cingulum. Whether or not these changes can be visualized using
imaging methods such as those produced from diffusion-weighted scans remains to be
determined. Also, it is possible to disrupt the cholinergic pathways without damaging the
basal forebrain directly (Selden et al., 1998); disruption of the cingulum, whether by
lesion or loss of axonal integrity, could result in cholinergic denervation ‘downstream’
from the site of disruption. This basal forebrain-retrospenial ACh pathway through the
cingulum is also important in light of research demonstrating profound anterograde
amnesia with damage to the retrosplenial cortex (Valenstein et al., 1987).
57
Figure 1-1. Schematic of the long-term memory system.
58
Figure 1-2. Rendering of the anatomy of the medial temporal lobe. The entorhinal cortex is not labeled in this image but is part of what is labeled ‘Parahippocampal gyrus’.1
1 Image (http://commons.wikimedia.org/wiki/File:Hippocampus_(brain).jpg) created by Frank Gaillard and available for use under the GNU Free Documentation License (http://www.gnu.org/copyleft/fdl.html).
59
Figure 1-3. Schematic representation of the structure and connections of the medial temporal lobe.
60
Figure 1-4. Simplified schematic representation of the Papez circuit.
61
Figure 1-5. Fiber tracking image showing the ventroposterior cingulum connecting between the retrosplenial cortex and the entorhinal cortex. Image visualization using TrackVis (Wang, Benner, Sorensen, & Wedeen, 2007)
62
CHAPTER 2 AIMS, HYPOTHESES, AND METHODS
Study Rationale
It is not well understood what extent of verbal memory deficits in PD patients can
be explained by damage to the medial temporal lobes, closely connected cortical
regions, and associated white matter areas, such as the posterior cingulum. What is
clear is that many PD patients experience verbal memory difficulties, even early in the
disease process. There is also evidence that, for some PD patients, a portion of their
memory difficulties is amnestic in nature (Filoteo et al., 1997) but the few studies
assessing verbal memory heterogeneity in PD were heterogeneous in results with
reports of 7% to 23% of non-demented individuals with PD experiencing amnestic
memory deficits. The current study will help clarify the proportion with amnestic memory
deficits in early PD.
Memory for verbal information is heavily dependent on the MTL, which
experiences pathological changes early in the course of idiopathic PD. There are also
other changes that occur in related brain areas, such as the cingulum; the posterior
section of the cingulum carries the bulk of the connections between entorhinal and
retrosplenial areas, both of which are involved in memory processes. The purpose of
this study is thus to elucidate the relationship between verbal memory and the MTL and
posterior cingulum in non-demented individuals with PD. This is not ignoring or
minimizing the contributions of changes in subcortical structures, such as the basal
ganglia that are heavily dependent on dopamine for functioning; rather, the purpose of
this study is to investigate the relative contributions of the MTL and a related white
matter pathway to the verbal memory performance of individuals with PD.
63
Specific Aims and Hypotheses
Aim 1: Is there a Verbal Memory Deficit in PD?
To examine verbal memory abilities using clinical tools (list learning, story
memory) in non-demented individuals diagnosed with PD (n=40) relative to age and
education matched non-PD peers (i.e., Controls, n=40).
Sub-aim. Semantic knowledge of the same individuals will also be assessed in
order to serve as a dissociation with verbal memory.
Hypothesis. Due to potential disruption of the medial temporal lobe and
associated fiber pathways in PD patients, it is hypothesized that individuals with PD will
score lower than non-PD peers on a combined set of memory indices assessing
savings, retention, and discrimination between true and false positives on recognition
testing (i.e., learned information versus foils). This will be assessed with a multivariate
analysis of variance (MANOVA). Follow-up univariate analyses will seek to confirm
hypotheses that individuals with PD perform more poorly than controls on individual
indices of savings, delayed recall, and recognition discriminability from the PrVLT as
well as Story memory savings index (percent retained), delayed recall, and recognition
discriminability scores. These relationships between group and verbal memory will be
partially explained by processing speed but variance in scores between groups will not
be fully explained by processing speed.
Between group sub-aim. Conversely, it is predicted that there will be no group
difference in scores on the Boston Naming Test or Association Index (AI) – a scale
derived from a test of semantic fluency – because the semantic knowledge system is
thought to be largely intact in PD. This will serve as a dissociation with verbal memory
because memory should be impaired but language should not be impaired.
64
Within group memory sub-aim. Although it is predicted that verbal memory
performance will be worse in individuals with PD compared to non-PD peers, it is further
predicted that there will be heterogeneity in verbal memory profiles for individuals with
PD when using control data as normative values. Additionally, because of prominent
and general executive and processing speed deficits in PD patients, for the PrVLT the
delay memory and recognition discriminability indices will be more compromised relative
to the savings index, which is less affected by executive and processing functions.
However, processing speed deficits will not fully explain memory deficits.
As part of this analysis, I will seek to relate other cognitive measures with the
components of list-learning and story memory indices. Measures of processing speed
(the processing speed index) will have a positive relationship with PrVLT delayed recall
and Story recall but a weaker relationship with PrVLT recognition discriminability or
Story recognition. This relationship is expected because while processing speed is an
important substrate of working memory (Dujardin et al., 2007) and thus encoding of
information, the effects of processing speed are minimized on a recognition task. In
other words, recognition is not as dependent on effortful retrieval as free recall is
(Filoteo et al., 2009). It is also predicted that within the PD group, there will be an
inverse relationship between disease severity and verbal memory performance.
Aim 2: Are there Differences in Verbal Memory Brain Structures for PD?
To investigate differences in verbal memory neuroanatomical structures
(entorhinal cortex and entorhinal to retrosplenial connections) for individuals with PD
and healthy non-PD peers. As a sub-aim, group differences in the integrity of semantic
knowledge white matter pathways (specifically, part of the left superior
longitudinal/arcuate fasciculus) will also be assessed.
65
Hypothesis. It is predicted that individuals with idiopathic non-demented PD
(n=40) will have smaller entorhinal cortex volumes and decreased integrity of the
connection between the entorhinal and retrosplenial cortices (which travels through the
cingulum), relative to non-PD age and education-matched peers (n=40). This prediction
is based on evidence of both hippocampal and entorhinal atrophy in non-demented
individuals with PD (Laakso et al., 1996; Jokinen et al., 2009) as well as evidence of
cingulate changes, particularly in the posterior region of the cingulate, in individuals with
PD (Braak & Braak, 2000). Additionally, with entorhinal changes, verbal memory
changes, reductions in acetylcholine, concentrations of Lewy bodies, and the number of
connections between the entorhinal cortex and the retrosplenial cortex, it is expected
that there will be connectivity changes between the retrosplenial and entorhinal cortices.
Between-group Sub-aim. Conversely, it is predicted that there will not be a
difference in the integrity of the arcuate fasciculus (AF), specifically, connections
between the left middle temporal gyrus (MTG) and ventrolateral frontal regions (i.e.,
pars opercularis). This area of the left frontal lobe has been implicated in producing
spoken words and broader language functioning (Indefrey & Levelt, 2004). Object-
related semantic knowledge is particularly dependent on the posterior temporal lobe
(Damasio & Tranel, 1993). In older adults, when producing word sounds and words, the
middle temporal gyrus seems to be more involved than in younger adults (Sörös et al.,
2009), making connections to this region potentially important for understanding
language. Language, of which semantic knowledge is a subset, is dependent upon a
wide network of brain areas including the frontal, parietal, and temporal lobes,
particularly in the left hemisphere (Broca, 1861; Wernicke, 1874; Catani, Jones, &
66
ffytche, 2005). Given that these regions are considered rather unaffected early in the
disease process, it is not anticipated these regions would be differentially compromised
in PD relative to non-PD age matched peers.
Aim 3: Are there Specific Structural-Cognitive Patterns?
To investigate the relative contribution of retrosplenial to entorhinal connectivity
and the entorhinal cortex volume to verbal memory performance in PD and Control
participants. Also to investigate the contribution of AF connectivity to semantic
knowledge.
Hypothesis. It is predicted that there will be a positive correlation between left
entorhinal cortex volume and recall and recognition performances on verbal memory
measures across both groups. That is, regardless of group (PD, Control), participants
with smaller left entorhinal cortices will have lower scores for both recall and recognition
components of verbal memory tests. A similar pattern will exist for the relationship
between left entorhinal to retrosplenial (ERC – RSC) connectivity and verbal memory. It
is specifically predicted that lower integrity of the left ERC – RSC connections will relate
with worse delayed recall and recognition verbal memory performance. Analyses will
also be run assessing the associations of disease severity with brain and verbal
measures. It is predicted that verbal memory measures will associate with disease
severity.
Sub-aim. To investigate the relationships between arcuate fasciculus (AF)
connectivity and measures of language across both groups. It is predicted that there will
be a positive correlation between language (BNT and the AI scale from semantic
fluency) and AF connectivity.
67
Sub-aim. To investigate the relationships between 1) ERC – RSC connectivity and
semantic knowledge and 2) AF connectivity and verbal memory. It is predicted that
there will not be a significant relationship between the cingulum (ERC – RSC) and
semantic knowledge neither will there be a significant relationship between the arcuate
fasciculus and verbal memory, providing evidence of dissociation between functions of
the tracts.
Within-group sub-aim. For both PD and Control groups, the associations
between brain structure and verbal memory will be calculated in order to assess the
hypothesis that worse memory performance is related with smaller left entorhinal
volumes and lower ERC – RSC connectivity. These relationships will additionally be
controlled for disease severity; it is predicted that within each group, controlling for
disease severity measures will not change the associations between brain structure and
memory performance.
Methods
Study Design
This was a prospective use of structural MRI and neuropsychological assessment
data from a recently ended NINDS funded investigation (NINDS: K23NS060660; Price).
Participants
Two age-matched participant groups were used to address Aims 1 and 2. The
groups were combined in order to address Aim 3.
Idiopathic Non-demented Parkinson’s Disease
Age and education matched non-PD peers
68
The final participant pool included 80 participants (40 per group). Inclusion and
exclusion criteria were based on the criteria for the NINDS funded study from which
these data were acquired (see APPENDIX A).
Recruitment. The current study acquired participants recruited as part of a
National Institute of Neurological Disorders and Stroke funded study that examined
white matter in Parkinson’s disease (NINDS; Primary Investigator = Catherine Price,
Ph.D.). Male and female PD participants over the age of 60 were recruited from the
community through referrals from the University of Florida Center for Movement
Disorders and Neurorestoration as well as through community flyers. Parkinson’s
disease participants had consensus-confirmed diagnosis of Parkinson’s disease based
on the United Kingdom PD Society Brain Research Center criteria (Gibb & Lees, 1988).
All individuals with PD were non-demented and with Hoehn and Yahr scale range of 1-3
(Hoehn & Yahr, 1967). Individuals with other neurodegenerative disorders, significant
medical disease that could limit lifespan, or major psychiatric disorders were excluded
from participation. Included PD participants were thus representative of healthy
community-dwelling older adults with Parkinson’s disease. Some members of the
Control group were family of PD participants but most were recruited from flyers and
direct mailings to the community. Control group participants had similar exclusionary
criteria and closely matched PD participants other than not having symptoms of PD.
While participants were recruited from the general community and recruitment was as
inclusive as possible, the final participant pool did not include members of ethnic
minority groups.
69
Neuropsychological Variables
To rule out dementia, participants were administered a comprehensive testing
protocol covering major domains of cognition and memory: attention, processing speed,
language, visuoperceptual abilities, intelligence, verbal and visual memory, and
executive function.
Primary dependent measures for the present investigation include: the
Philadelphia (repeatable) List Learning Test (PrVLT) and the Logical Memory (Story)
test from the Wechsler Memory Scale 3rd Revision (WMS-III) are of primary interest.
Secondary dependent measures for the present investigation include: Association Index
from Animal Fluency and Boston Naming Test.
Philadelphia Repeatable Verbal Learning Test
Even though the Philadelphia (repeatable) Verbal Learning Test was previously
briefly described, it will be described again for the methodology of this study. The PrVLT
is a list learning and memory test designed for use with older research participants that
has been shown to be sensitive to different patterns (e.g., cortical or subcortical) of
verbal memory impairment (Price et al., 2009; Libon et al., 2010; Tanner et al., 2011). It
has a similar design to the California Verbal Learning Test (CVLT) but with lists of either
9 or 12 words. One other notable difference from the CVLT is that the PrVLT includes
words pulled specifically from a corpus of words produced by healthy older adults.
There are 5 learning trials (list A), one interference trial of a novel list of 12 words (list
B), short and long delay cued and free recall, and a recognition trial. The recognition
trial includes all 12 words from list A, all 12 from list B, 12 unrelated foils, and 12
semantically related foils. The PrVLT allows for fine-tuned analyses of learning and
70
memory, especially for distinguishing between memory impairment profiles (Price et al.,
2009).
PrVLT scores considered sensitive to MTL changes were created for targeted
analysis for Aim 1. The following scores from the PrVLT were used as dependent
variables in analyses: long delayed free recall, a savings index (Lamberty, Kennedy, &
Flashman, 1995; Welsh et al., 1994), and recognition discriminability.
Long delayed free recall. The total number of words correctly and freely recalled
(i.e., without cuing) after a 25 minute filled delay were recorded as the long delayed free
recall score.
Savings index. The savings index is a measure of information retained from the
immediate learning to the delayed time period. It was calculated as free delayed recall /
learning trial 5 (Filoteo et al., 2009; see also for comparison, Lamberty et al., 1995).
This savings score has been shown to be sensitive to early memory changes in
dementia (Welsh et al., 1994).
Recognition discriminability. Recognition discriminability is the number of words
correctly endorsed on recognition testing is corrected for the ‘noise’ of foils endorsed.
This was calculated using d’ from signal detection theory.1
Story Memory
Logical Memory from the WMS-III is a commonly used clinical and research story
memory test. For this test a brief, paragraph passage of prose was read. The individual
was then asked to recall as much of the story using the same words and in the same
1 Calculated in Excel using the formula NORMSINV(prob), where prob is P(h) or P(fa). This gives the area under the curve of the normal distribution: d' = NORMSINV( P(h) ) - NORMSINV( P(fa) )
71
order as was presented, if possible. After this a different story was presented and the
individual asked to recall the second story. Then this process was repeated for the
second story (i.e., the second story was presented and recalled twice). After a 20-30
minute delay, the individual was asked to recall each of the two stories. Lastly, a
recognition component was presented where a series of yes/no questions were asked
about the two stories.
The story memory index scores of interest for use as dependent variables were
delayed recall (total number of story elements freely recalled after a filled 25 minute
delay), savings (retention percent – ratio of delayed recall to immediate {learning}
recall), and the recognition score.
Boston Naming Test
The Boston Naming Test (BNT) is a 60 item test of receptive language (Kaplan,
Goodglass, & Weintraub, 1983). Participants were shown simple line drawings of
objects and asked to produce the names of the objects. The total items correctly named
without cuing was the score of interest for analysis.
Association Index
Category (semantic) fluency tests are sensitive to cortical disruption, primarily of
the temporal lobe (Rodriguez-Ferreiro et al., 2011). Using characteristics of words, it is
possible to classify the cohesion between responses on semantic fluency tests. Animal
fluency responses can be classified using various taxonomic and zoological categories:
size, geographic location, habitat, zoological class, zoological orders and families, and
diet (Carew et al., 1997). Using these categories, the shared attributes between
successive words can be classified and scored as the sum of shared attributes across
dyads of words for all responses, divided by the total number of responses minus one.
72
For example, if an individual gave 16 responses, starting with horse and cow, the
association between that particular dyad would be 5 because those animals match on 5
of 6 categories: big, local, herbivore, mammal, and farm. Doing this for each successive
dyad results in a total association score. For 16 responses, if the total of association
scores was 58, the association index (AI) score would be 58 / (16 – 1) = 3.87.
It has been shown that individuals with ischemic vascular dementia have relative
preservation of semantic networks compared to individuals with Alzheimer’s disease,
even though the number of generated words is similar to that by individuals with AD
(Carew et al., 1997). This is interpreted to mean that even though those with subcortical
dementias have reduced output, there is still an understanding of the subordinate
connections between words (animals). This means that those with subcortical disorders
are still able to cluster words semantically and thus any deficits are not likely due to
semantic problems but rather to processing speed deficits or executive dysfunction.
Even though non-demented individuals with PD should have considerable and growing
temporal lobe PD pathology (Braak et al., 2004) that could potentially affect
performance on the Association Index, due to their lack of general cognitive deficits, it is
not likely that individuals in early stage PD would experience deficits in AI scores. The
primary variable of interest for this analysis was total AI score.
Imaging Protocols
Magnetic resonance imaging and Parkinson’s disease. Magnetic resonance
imaging (MRI) is a type of imaging that allows researchers and clinicians to acquire high
resolution in vivo images of the human brain. Magnetic resonance imaging is based on
the fact that various types of tissues or chemicals have different physical (cellular,
molecular, and atomic) properties that react differently to manipulations within a strong
73
magnetic field. Basically, protons, when exposed to the strong magnetic field in an MRI
machine, tend to align with the magnetic field. Using pulsed gradients of radio waves,
the MRI machine shifts protons out of alignment, and in the process, measures changes
in the radio frequency signal as the protons return to their aligned resting states. By
varying signal gradients and measurements, different components of a brain, for
example, can be visualized with some clarity. For the present study there were two
types of MRI scans that were pertinent – T1 weighted structural scans and diffusion
weighted structural scans.
T1 imaging. T1 MR images are commonly produced clinically and for research to
provide clear images of the structure of the brain. At magnetic field strengths of 3 Tesla
(3T), it is possible to obtain high resolution (1x1x1mm voxels) 3D images within a
relatively brief period of time (5-10 minutes). T1 scans offer good contrast between gray
and white matter and cerebrospinal fluid (CSF). In T1 scans, gray matter (cell bodies)
appears darker than white matter (myelinated axons), and CSF appears even darker
than gray matter. Using T1 scans, researchers can differentiate between various cortical
and subcortical areas of the brain. Consequently, it is possible to obtain reliable
estimates of the volumes of brain regions; these volume estimates allow for statistical
analyses to be performed with and clinical judgments made about various brain
structures.
Diffusion weighted imaging. Diffusion weighted MRI (DWI) measures the
displacement of water molecules over time. In cell bodies and CSF, the displacement is
relatively quick and isotropic but in the white matter of the brain, the displacement is
more anisotropic, that is, it tends to move in a restricted, directional manner rather than
74
equally in all directions. Because water molecule displacement is directional in white
matter, diffusion weighted imaging is particularly useful for investigating the structure
and integrity of the white matter. DWI thus can be considered as providing a measure of
the microstructural integrity of brain cells (Le Bihan, 1995) and is useful for investigating
altered white matter microstructure (Rosas et al., 2010). Investigations of
microstructural alterations can be done between groups or within groups of individuals.
For example, researchers found changes in the white matter of the corpus callosum in
Huntington’s disease (Rosas et al., 2010), altered diffusivity in the frontal lobes of
Parkinson’s patients (Karagulle Kendi et al., 2008), and changes in the white matter
near the substantia nigra in Parkinson’s patients (Yoshikawa et al., 2004). Diffusion
weighted imaging thus is a tool that can be used to see disruption of circuits in the brain.
In order to perform these analyses, MRI requires time spent with automated and
manual processing of the images.
Image Processing
Figure 2-1 provides an overview of the workflow that was utilized for the fiber
tracking analysis of the diffusion weighted images. The major components of the
pipeline will be briefly described.
Data acquisition. Data were acquired with a Siemens 3T Verio scanner. Single-
shot EPI diffusion weighted images were acquired with diffusion gradients applied along
6 directions (b = 100 s/mm2) and 64 directions (b = 1000 s/mm2). These diffusion scans
were then combined and motion corrected in order to increase the signal to noise ratio.
Imaging parameters were 73 contiguous axial slices with a slice thickness of 2mm, and
TR/TE = 17300/81ms. Two T1-weighted sequences were acquired with the following
parameters: 176 contiguous slices, 1mm3 voxels, TR/TE = 2500/3.77ms. Multiple T1-
75
weighted volumetric sequences were acquired in order to average the T1 scans to
optimize the signal-to-noise ratio by reducing the effects of motion as well as by
increasing gray-white contrast.
Diffusion image processing. Diffusion weighted MRI data from all participants
were processed using in-house software based on an advanced fiber tracking analysis
called mixture of Wishart (MOW) that allows for the visualization and quantification of
white matter. This method is an improvement over diffusion tensor analysis (DTI) by
enhancing the parameterization of complex fibers within voxels (Jian et al., 2007). See
Figure 2-2 for examples of the modeling improvements of MOW over DTI for fiber
tracking.
In less complex areas without ‘kissing’ or ‘crossing’ fibers, such as medial portions
of the corpus callosum or core regions of the cingulum, the difference between DTI and
MOW might be marginal; however, given that the MOW modeling includes DTI as a
subset, there is no detriment to using MOW instead of DTI. See Figure 2-3 for examples
of glyphs from the cingulum. This area was chosen for visual comparison between DTI
and MOW methods because it is highly directional and thus, relatively simple
structurally. Other areas of the cingulum are more complex but this area was chosen to
provide contrast with the frontal area in Figure 2-2.
Diffusion variables of interest. A measure of connectivity strength was used to
quantify the cingulum between the entorhinal cortex and retrosplenial cortex. This
connection between regions of interest (nodes) was quantified in the following manner.
Any two nodes (Na,b) are connected with an edge (Ea,b) if there is at least one fiber (f)
that starts and stops in the two nodes (Na,b). Each edge (E) has both a length (lE) and
76
weight (wE). Weight (wE) was defined as: . This mathematical model
corrected for the bias that the lengths of fibers have on the weight wE (Hagmann et al.,
2007). Next wE was corrected for voxel volume (V) and fiber seed density (SD):
. In the present study, V = 8 and SD = 64. Then, to account for differences in
ROI surface areas, the corrected edge weight (wEc) was multiplied by two and divided by
the sums of the surface areas SAa,b: . In order to control for potential
connectivity differences caused by differences in head size (see Yan et al., 2011), EWC
was then divided by intracranial volume (ICV) in mm3, giving a final EWC / ICV variable
of interest. Some variables were scaled linearly in order to increase apparent value for
ease of interpretation.
Entorhinal volumetrics. Structural T1 scans were processed using FreeSurfer,
which is a set of MRI analysis tools. These tools allow for automated processing of T1
MRI data (Segonne et al., 2004; Fischl et al., 2002), ideally with little initial input from
the user; however, quality checking was performed in order to assess reliability of
results. From the FreeSurfer processing, an averaged brain (across both acquired T1
images) with enhanced gray-white contrast and increased signal-to-noise was aligned
to the MNI152 template brain (Fonov et al., 2011; Mazziotta et al., 2001) using a linear,
non-destructive registration technique with 6 degrees of freedom (FLIRT; Jenkinson et
al., 2002; Jenkinson et al., 2001; Greve et al., 2009). This was done in order to correct
for head tilt and to align participants’ brains along the anterior commissure – posterior
commissure axis. Entorhinal cortices were then manually traced by an expert rater
wE 1
l ff FE
wE V
SD
EWC 2wE c
SAa SA
b
77
according to published methods of identifying the entorhinal cortex on MR scans (refer
to Insausti et al., 1998 for an in-depth presentation of this process).
Briefly, starting 2mm posterior to the appearance of the temporal stem, the lateral
wall of the parahippocampal gyrus was traced between the sulcus semiannularis and
the collateral sulcus, descending into the collateral sulcus at varying depths depending
on the overall depth of the collateral sulcus. This method closely matches the
cytoarchitecture of the entorhinal cortex and allows for localization on T1 MRI (Insausti
et al., 1998).
This method is reliable (intra-rater DSC > 0.8; inter-rater reliability DSC > 0.8) and
has been shown to relate with verbal memory performance in older adults (Price et al.,
2010).
Retrosplenial region of interest. Structural T1 scans were processed through
the FreeSurfer pipeline. From the cortical parcellation (labeling of different areas of the
cortex based on template atlases while accounting for individual sulcal and gyral
variability), the isthmus of the cingulate was exported, inflated by 1mm in 3 dimensions,
and imported into ITK-SNAP for manual cleaning (e.g., to remove overlap with corpus
callosum) to localize the ROI to the perisplenial region (retropsplenial cortex plus
underlying white matter in order to improve tracking results).
Final imaging variables of interest. Entorhinal cortex - the volumes of the left
entorhinal cortices (as calculated above) were entered into statistical analyses. These
volumes were adjusted for intracranial volume. Entorhinal cortex to retrosplenial cortex
connectivity strengths (ERC – RSC EWC; as calculated above) was entered into
statistical analyses. These values (EWC) were also corrected for intracranial volume as
78
described above. This value is hereafter abbreviated to ‘ERC – RSC EWC’. Connectivity
between the left MTG and frontal regions (AF EWC) was also entered into the analyses.
Statistical Analyses
Data that were not normally distributed were normalized using the Box-Cox
procedure (Osborne, 2010), which allows for more robust normalization of value
distributions than more traditional normalization methods; these traditional methods
(e.g., square root, log, square root) are variations of power transformations. The Box-
Cox procedure normalizes by allowing for an optimal transformation to be chosen from
a range of power transformations. The Box-Cox procedure incorporates other traditional
power transformations but is more robust than any of them individually.
Analyses were Bonferroni corrected where appropriate to control for multiple
comparisons.
Aim 1. Index scores from the PrVLT and Story memory were used in a between-
group (PD and Control) MANOVA model to assess if PD participants overall across all
included index scores have lower verbal memory scores than Control participants.
Additionally, the Processing Speed Index (PSI) was used as a covariate in a one-way
MANCOVA analysis in order to control for the effects of speed of processing on
memory. All multivariate analyses were followed up by univariate analyses in order to
assess individual index score group differences.
Sub-aim. The total score on the Boston Naming Test and the Association Index
(AI) from animal fluency were used in a between-group one-way MANOVA model to
assess if PD and Control participants had similar semantic abilities.
Within PD group analysis. Scores on the PrVLT and Story memory indices were
compared with the score on the recognition discriminability index in order to investigate
79
the heterogeneity of levels impairment. Z scores established using data from Controls
were used to create frequency counts of impairment in each verbal memory index score
(scores above z = -1.0 were classified as ‘Not Impaired’; scores between z < -1.0 and >
-1.5 were classified as ‘Mild Impairment’, scores between z < -1.5 and > -2.0 were
classified as ‘Moderate Impairment’, and scores below z = -2.0 were classified as
‘Severe Impairment’). Given how closely matched the PD and Control groups were, this
classification system was an appropriate compromise between minimizing Type I and
Type II errors.
Further, to assess the level of and variability in overall memory impairment for
each PD participant, a cumulative sum of ‘z score impairment’ was used to assess
individual memory performance variability. Z scores < 0 were added together to create
this summed impairment for each individual.
Additionally, for the PD and Control groups separately the PrVLT scores were
correlated with cognitive measures of processing speed (Processing Speed Index) to
assess the role processing speed plays within groups.
Aim 2. Left entorhinal cortex volumes were quantified, as was ERC – RSC EWC.
These values were used as the variables of interest. All volumetric analyses were then
controlled for total intracranial volume to reduce structure volume variability that results
from head size differences. These quantifications of the entorhinal cortices and ERC –
RSC EWC were then entered into Student’s t-test analyses to determine group
differences between PD and Control participants.
Within group correlations. Bivariate correlations between measures of disease
severity (total l-Dopa intake and UPDRS total score) and neuroanatomical variables
80
were run in order to assess the relationship between disease severity and
neuroanatomy.
Sub-aim. Arcuate fasciculus EWC was entered into an independent samples
Mann-Whitney U Test to assess the hypothesis that there are no group differences in
the structural integrity of this white matter pathway.
Aim 3. Using bivariate correlations relationships between verbal memory
performance on the index scores from Aim 1 and neuroanatomical variables (left
entorhinal to retrosplenial cortex connectivity and left entorhinal volume variables) were
calculated. This allowed for the investigation of the relationships of specific
neuroanatomical regions with list learning and story memory.
In order to simplify comparisons, a verbal memory composite score was calculated
to assess the relationships between overall memory performance and brain structure.
Additionally, bivariate correlations were calculated between verbal memory index scores
and disease severity as well as neuroanatomy and disease severity. Further, within
group comparisons were run in order to assess relationships between neuroanatomy
and verbal memory with and without controlling for disease severity.
To assess dissociations in the relationships between neuroanatomy and cognition,
bivariate correlations were also calculated between left AF EWC, the left ERC – RSC
EWC, left ERC volume, measures of language (BNT and the AI scale from semantic
fluency), and verbal memory index scores.
81
Figure 2-1. Schematic showing the image processing pipeline from scanner to fiber tracking and volumetric output. A full-size image is attached in Appendix B.
82
A
B C
Figure 2-2. Series of images demonstrating the modeling of diffusion within a single voxel. A) color fractional anisotropy image with crosshairs centered on a voxel in the frontal forceps, which is an area with many crossing axonal fibers, B) DTI glyph showing a single voxel with complex (crossing) fibers, in this instance an area close to the frontal forceps in the white matter of the frontal lobe, C) MOW glyph showing the same voxel as in B. Note the ability to resolve the complexity of the fibers better than DTI.
83
A
B C
Figure 2-3. Series of images demonstrating the modeling of diffusion within a single voxel. A) color fractional anisotropy image with crosshairs centered on a voxel in the cingulum, which is an area with few crossing axonal fibers, B) DTI glyph showing a single voxel with relatively simple (directional) fibers, in this instance an area in the core of the cingulum, C) MOW glyph showing the same voxel as in B. The differences between MOW and DTI in this instance are likely marginal.
84
CHAPTER 3 RESULTS
Participant Characteristics
A set of 40 PD participants and 40 Controls matched on demographic variables,
intelligence estimates, and general cognition (all p > 0.05; see Table 3-1). Groups
differed significantly on measures assessing PD symptom severity, disease duration,
and processing speed (see Table 3-1).
Aim 1: Is there a Verbal Memory Deficit in PD?
Multivariate Analyses: Group by Verbal Memory
As hypothesized, PD participants obtained significantly lower scores than Control
participants when combining the six index scores in a one-way MANOVA (see Table 3-
2). The multivariate analysis revealed a significant effect for group (Wilks’ lambda F =
3.83, p < 0.01, partial eta squared = 0.24; power to detect the effect was 0.95).
When covarying for processing speed (PSI), there was no longer a significant
multivariate group difference across all six verbal memory index scores (MANCOVA;
Wilk’s lambda F = 1.84, p = 0.10).
Sub-aim: Language Group Differences
As hypothesized, the one-way MANOVA with language measures was not
significant (Wilks’ lambda F = 0.91, p = 0.50; partial eta squared 0.02; see Table 3-3).
Neither individual measure significantly differed between groups.
PD Verbal Memory Performance by Index
Univariate analyses showed that PrVLT recognition discriminability and Story
recognition discriminability scores had the strongest between group effect sizes.
85
Savings scores from both the PrVLT and the Story memory task did not differ between
groups.
PrVLT long delay free recall. See Figure 3-1. PD < Controls. Parkinson’s disease
participants recalled an average of 8.13 words; Control participants recalled an average
of 9.65 words (F = 9.69, p < 0.01, eta2 = 0.11).1 In a follow-up analysis, when controlling
for PSI, PD < Controls (F = 4.09, p < 0.05).
PrVLT savings. Parkinson’s disease participants were not significantly reduced
compared to Controls. Parkinson’s disease mean = 84.18%; Control mean = 88.67% (F
= 1.66, p = 0.20, eta2 = 0.02). In a follow-up analysis, there was not a significant group
difference when controlling for PSI (F = 1.33, p = 0.25).
PrVLT recognition discriminability. See Figure 3-1. PD < Controls. PD mean =
3.01; Control mean = 3.49. F = 12.58, p < 0.01, eta2 = 0.14. Follow-up analyses
revealed that PD participants endorsed more false negatives (PrVLT recognition hits;
PD mean = 11.13, Control mean = 11.48, p = 0.05) and false positives (PrVLT
recognition errors; PD mean = 2.78, Control mean = 1.69, Mann-Whitney U p < 0.01)
than Controls.2 In a follow-up analysis, there was a significant group difference (PD <
Controls) when controlling for PSI (F = 9.55, p < 0.01).
1 A follow-up analysis of intrusions on delayed recall revealed that PD did not differ from Controls in the number of long delay free recall intrusions produced but PD > Controls in the number of long delay cued recall intrusions. Long delay free recall intrusions: PD free recall intrusions (mean = 1.08) > Control free recall intrusions (mean = 0.63); Mann-Whitney U p = 0.08. Long delay cued recall intrusions: PD recall intrusions (mean = 3.00) > control participants (mean = 1.83); (Mann-Whitney U p = 0.03).
2 A sub-analysis of the false positive errors on recognition testing revealed that the group differences were driven by semantically-related novel distracter words (PD mean = 1.90, Control mean = 0.88; Mann-Whitney U p < 0.01) but not distracter words from the interference trial (List B; PD mean = 0.30, Control mean = 0.18, Mann-Whitney U p = 0.67) or unrelated novel distracters (PD mean = 0.08, Control mean = 0.05, Mann-Whitney U p = 0.55).
86
Story delayed recall. PD < Controls. Parkinson’s disease participants recalled an
average of 24.80 story elements; Control participants recalled an average of 28.60 story
elements (F = 6.22, p = 0.02, eta2 = 0.07). See Figure 3-1. In a follow-up analysis, after
controlling for PSI, there were no group differences in Story delayed recall (F = 2.90, p =
0.09).
Story savings. Parkinson’s disease participants’ scores were not significantly
reduced compared to Controls. PD = 81.72%, Controls = 87.90% (F = 3.46, p = 0.07,
eta2 = 0.04). In a follow-up analysis, there was not a significant group difference when
controlling for PSI (F = 1.48, p = 0.23).
Story recognition discriminability. See Figure 3-1. PD < Controls. PD mean =
84.08, Control = 90.74 (F = 11.92, p < 0.01, eta2 = 0.14). In a follow-up analysis, after
controlling for PSI, there were no group differences in story recognition discriminability
(F = 3.59, p = 0.06).
Within Group Heterogeneity of Memory Impairment
Within the PD group there was variability in memory performance. Using scores
from the matched Controls, norms were created for the PD participants. Four levels of
impairment were created based on z scores: Not Impaired (z score > -1.0); Mild
Impairment (z score > -1.5 & < -1.0); Moderate Impairment (z score > -2.0 & < -1.5); and
Severe Impairment (z score < -2.0). Using these classifications for the six verbal
memory index scores, as many as 66% of PD participants exhibited impairment (see
Table 3-4 and Figure 3-2).
PD participants demonstrated greater compromise of recognition discriminability
scores for both verbal memory measures than for both delayed recall and savings.
87
Verbal memory associations with processing speed and disease severity.
The Processing Speed Index had a positive relationship with PrVLT LDFR (r = 0.33, p <
0.05) but did not relate with any other verbal memory index score (all r values < 0.15, all
p values > 0.35). Within the PD group, there are no correlations between disease
severity (UPDRS total score), levodopa intake (LED: levodopa equivalence dosage),
and verbal memory index scores (all p values > 0.06).
Individual verbal memory variability. When cumulative z impairment scores
(any z score < 0) were summed for PD participants, individual heterogeneity in verbal
memory performance was seen (see Figures 3-3, 3-4, and 3-5).
Of 39 PD participants (one PD individual was excluded for missing PrVLT data):
14/39 (35.90%) averaged at least 1 z score below the Control mean across 3 PrVLT
index scores; 5/39 (12.82%) averaged at least 2 z scores below the mean compared to
Controls across 3 index scores; and 7/39 (17.95%) individuals scored at or above the
mean for all 3 index scores of the PrVLT compared to Controls (see Figure 3-3).
For Story memory, 17/40 (42.5%) PD participants averaged at least 1 z score
below the Control means across the 3 index scores while 5/40 (12.5%) averaged at
least 2 standard deviations below the Control means. 6/40 (15%) PD participants
scored at or above the mean of Controls across all 3 story index scores (see Figure 3-
4).
When combining both verbal memory measures and all 6 index scores: 18/39
(46.15%) individuals averaged at least 1 standard deviation below the mean of Control
participants; 2/39 (5.13%) averaged 2 standard deviations below the mean; and 2/39
88
(5.13%) individuals performed at or above the mean for all 6 index scores (3 PrVLT, 3
Story) compared to Controls (see Figure 3-5).
Aim 2: Is there a Difference in Verbal Memory Brain Structures for PD?
Medial Temporal Lobe Structure Group Differences
The manual segmentation method used to acquire entorhinal volumes is reliable
(intra-rater DSC > 0.8; inter-rater reliability DSC > 0.8). Left entorhinal volumes were
normally distributed. Left ERC – RSC EWs were not normally distributed, thus Box-Cox
normalization was run to fit the data to a Gaussian distribution. See Figure 3-6 for
representative images of entorhinal volumetric and ERC – RSC tracking results.
As hypothesized, using an independent samples t-test analysis, PD participants
showed reduced left entorhinal volumes when corrected for TICV compared to Control
participants (t = 2.71, p < 0.01; see Table 3-5 and Figure 3-7). Left entorhinal volumes
were 11% smaller in PD participants than in Controls (Cohen’s d = 3.82, observed
power = 0.93). Contrary to what was predicted, however, PD participants did not
significantly differ from Controls in entorhinal to retrosplenial edge connectivity (t = 0.88,
p = 0.38; Cohen’s d = 0.21; see Table 3-5 and Figure 3-7). An independent samples
Mann-Whitney U test was also performed on the original, uncorrected left ERC-RSC
EW data; as with the t-test, there were no significant differences in the distribution of the
EW values between groups (p = 0.24).
Within Group Correlations With Disease Severity
Within the Control group, there were no significant relationships between l-Dopa
intake3, PD symptom severity, and neuroanatomical variables (all p values > 0.12).
3 One Control participant was on a dopamine agonist medication for restless leg syndrome.
89
Within the PD group, there were no significant relationships between l-Dopa intake, PD
symptom severity, and neuroanatomical variables (all p values > 0.51).4
Sub-aim: Language Track Group Differences
Non-normality resulted in an independent samples Mann-Whitney U Test being
performed with the data. As hypothesized, there were no group differences in the
distributions of edge weight of the white matter of the arcuate fasciculus (AF)
connecting the middle temporal gyrus with the pars opercularis (Mann-Whitney U p =
0.47; refer to Table 3-6 for group means; see Figure 3-9 for an example of the AF fiber
tracking output). See Figure 3-8 for a scatterplot of AF Edge Weight values by group.
Aim 3: Are there Specific Structural-Cognitive Patterns?
Brain Variables and Verbal Memory Performances Regardless of Group Type
As expected, there were relationships between neuroanatomical regions of
interest and verbal memory. For all individuals (PrVLT: n = 79; Story: n = 80), one-way
bivariate correlations yielded positive associations between ERC variables and verbal
memory index scores (see Table 3-7, Figure 3-10, and Figure 3-11), but the strength of
the associations varied by test index. The strongest associations were identified for
PrVLT recognition discriminability, story recognition discriminability, and story savings.
While the left ERC volume had a positive association with Story recognition
discriminability and Story savings, contrary to what was predicted it did not relate to
PrVLT index scores or Story delayed recall. While the left ERC-RSC EW was positively
associated with PrVLT recognition discriminability, it did not relate with other verbal
4 Supplemental within-group analyses were performed to assess if there was a relationship between PD side of onset and brain structure. Results are shown in Appendix C.
90
memory index scores. Additionally, a positive relationship was found between the edge
weight of the left AF between the pars opercularis and the middle temporal gyrus and
PrVLT savings scores (see Table 3-7).
Brain Variables and Semantic Performances Regardless of Group Type
Table 3-7 displays correlations between semantic measures and brain variables.
The left arcuate fasciculus did not relate with either semantic measure. The left ERC
and the left ERC-RSC EWC also did not relate with either semantic measure.
Verbal Memory Composite and Structure Relationships
A composite score was calculated to pool verbal memory test variance and control
effects of multiple comparisons. Using the composite of the six verbal memory scores, a
mild but significant association was found between left ERC volume, corrected for ICV,
and verbal memory across all subjects (r = 0.22, p < 0.05; see Figure 3-12). There were
no significant associations between the verbal memory composite and left ERC – RSC
EW or left AF EW (p values > 0.39).
Associations with Disease Severity
Disease severity and verbal memory. When combining all participants, there are
significant negative associations between UPDRS total score and verbal memory index
scores. Total levodopa dose did not relate with verbal memory except for Story
recognition discriminability (see Table 3-8).
Disease severity and neuroanatomy. When combining PD and Control
participants there were no relationships between l-Dopa intake, UPDRS score, and
neuroanatomical variables except for a mild negative association between UPDRS total
score and left entorhinal volume (r = -0.30, p < 0.01; all other p values > 0.09).
91
Within Group Comparisons
PD. Within the PD group there were significant correlations between verbal
memory performance and brain structure: Story savings and left ERC volume (r = 0.29,
p = 0.04); PrVLT recognition discriminability and left ERC – RSC EW (r = 0.33, p =
0.02). All other relationships are non-significant. There was a trend for a relationship
between left AF and PrVLT savings (r = 0.24, p = 0.07).
Controlling for disease severity. When removing the effects of total UPDRS
score, significant correlations were found between left ERC volume and Story savings (r
= 0.28, p < 0.05) and left ERC – RSC EW and PrVLT recognition discriminability (r =
0.32, p = 0.02).
When removing the effects of levodopa intake, significant correlations were found
between left ERC – RSC EW and PrVLT long delay free recall (r = 0.29, p = 0.04) and
left ERC – RSC EW and PrVLT recognition discrimination (r = 0.42, p < 0.01). There
was a significant correlation between left AF EW and PrVLT savings (r = 0.29, p = 0.04).
There was also a trend for left ERC volume and Story savings (r = 0.27, p = 0.06).
After for covarying for processing speed, significant correlations were found
between left ERC – RSC EW and PrVLT recognition discriminability (r = 0.33, p = 0.02)
as well as left ERC volume and Story savings (r = 0.23, p = 0.04).
Controls. Within the Control group there also were significant correlations
between brain structure and verbal memory performance: Story savings and left ERC
volume (r = 0.41, p < 0.01).
Controlling for disease severity. When controlling for the effects of processing
speed, the relationship between Story savings and left ERC volume remained (r = 0.41,
p < 0.01).
92
Dissociation Between PrVLT Performance, Left ERC-RSC EW, and Left AF EW
Left ERC – RSC EWC was moderately and positively associated with PrVLT
recognition discriminability (Spearman’s rho = 0.29, p = 0.01). No relationship was
identified for the left AF EWC to the PrVLT recognition discriminability score
(Spearman’s rho = 0.03, p = 0.82). However, conversely, left AF EWC correlated with
Story savings but left ERC – RSC EWC did not; this dissociation was not predicted.
93
Table 3-1. Participant baseline characteristics
Measure PD (n=40) Control (n=40) P Value
Age 67.85 (5.47) 67.95 (4.65) 0.93
Education 16.23 (2.94) 16.55 (2.49) 0.60
MMSE 28.93 (1.25) 29.25 (0.93) 0.19
DRS-2 Total 139.45 (3.22) 140.23 (2.54) 0.24
WTAR 107.55 (7.47) 108.80 (8.76) 0.49
UPDRS Total 30.78 (15.86) 3.63 (3.67) < 0.01
Disease Duration 7.58 (5.09) N.A. < 0.01
l-Dopa Equiv. Score 694.34 (364.88) 1.00 (6.33) < 0.01
Processing Speed Index 99.44 (10.50) 112.45 (11.14) < 0.01
MMSE = Mini-Mental State Examination; DRS-2 = Dementia Rating Scale – 2nd Version; WTAR = Wechsler Test of Adult Reading; UPDRS Total = United Parkinson’s Disease Rating Scale Total score; l-Dopa Equiv. Score = Levodopa Equivalent Score = Total Daily levodopa dosage intake in milligrams. One controls was on levodopa for restless leg syndrome. Processing Speed Index = Composite index score from the Wechsler Adult Scale of Intelligence – III.
94
Table 3-2. Verbal memory MANOVA group contrast
Measure PD (n=39) Control (n=40) P Value Eta Squared
PrVLT LDFR 8.13 (0.35) 9.65 (0.34) 0.003 0.11**
PrVLT Savings 84.18 (2.49) 88.67 (2.45) 0.202 0.02*
PrVLT R. Discr. 3.01 (0.10) 3.49 (0.09) 0.001 0.14***
Story Delay Free 24.80 (1.09) 28.60 (1.07) 0.015 0.07**
Story Savings 81.72 (2.37) 87.90 (2.34) 0.067 0.04*
Story R. Discr. 84.08 (1.37) 90.74 (1.35) 0.001 0.14***
* Small effect size, ** Medium effect size, ***Large effect size. LDFR = long delay free recall; R. Discr. = recognition discriminability. Note: one PD participant was excluded because of missing PrVLT data as a result of administration error. An additional analysis was conducted assessing side of onset of PD symptoms and relationships with verbal memory; all relationships were non-significant (p > 0.05).
95
Table 3-3. Semantic knowledge MANOVA group contrast
Measure PD (n=40) Control (n=40) P Value Eta Squared
BNT 56.78 (3.32) 57.58 (2.19) 0.282 0.01**
AI Total 3.30 (0.48) 3.24 (0.53) 0.583 0.004*
* <Small effect size; ** Small effect size. Note: BNT = Boston Naming Test; AI = Association Index (Carew et al., 1997), which assesses consistency of semantic features for word exemplars. BNT and AI scores were normalized using the Box-Cox procedure (Osborne, 2010) due to significant skew and kurtosis. Mean raw scores are presented in the table for ease of interpretation.
96
Table 3-4. Percent PD verbal memory impairment
Measure N.I. Mild Moderate Severe
PrVLT LDFR 61.53% 15.38% 12.82% 10.26%
PrVLT Savings 71.79% 7.69% 15.38% 5.13%
PrVLT R. Discr. 43.59% 25.64% 7.69% 23.08%
Story Delay Free 60.00% 2.50% 20.00% 17.50%
Story Savings 72.50% 7.50% 12.50% 7.50%
Story R. Discr. 57.50% 7.50% 12.50% 22.50%
Note: LDFR = long delay free recall; R. Discr. = recognition discriminability. N.I. = Not impaired: z score > -1.0; Mild = z score > -1.5 & < -1.0; Moderate = z score > -2.0 & < -1.5; Severe = z score < -2.0.
97
Table 3-5. Medial temporal lobe structures group contrast
Measure PD (n=40) Control (n=40) P Value
L ERC/ICV 7.04 (0.22) 7.88 (0.22) < 0.01
L E-R/ICV EW 4.7 x 10-3 (2.8 x 10-3) 5.3 x 10-3 (2.9 x 10-3) 0.38
Note: ERC/ICV = Entorhinal cortex volume corrected for intracranial volume X 10000; E-R/ICV EW = Entorhinal cortex to retrosplenial cortex edge weight normalized and adjusted for intracranial volume (values also normalized using Box-Cox procedure).
98
Table 3-6. Left arcuate fasciculus group contrast
Measure PD (n=40) Control (n=40) P Value
Left AF/ICV EW 9.00 (9.12) 10.56 (9.14) 0.47
Note: AF/ICV EW = Arcuate Fasciculus edge weight normalized and adjusted for intracranial volume.
99
Table 3-7. Brain – verbal memory and language bivariate correlations
Measure Left ERC/ICV Left E-R/ICV EW Left AF/ICV EW
PrVLT LDFR 0.12 0.11 0.15
PrVLT Savings 0.08 0.01 0.22
PrVLT R. Discr. 0.15 0.24* 0.13
Story Delay Free 0.06 0.08 -0.03
Story Savings 0.39** 0.05 -0.01
Story R. Discr. 0.22* 0.09 0.09
BNT 0.09 0.05 -0.07
AI -0.05 -0.05 0.11
Note: *p<0.05, **p<0.01; ERC/ICV = Entorhinal cortex volume corrected for intracranial volume X 10000; E-R/ICV EW = Entorhinal cortex to retrosplenial cortex edge weight normalized and adjusted for intracranial volume; AF/ICV EW = arcuate fasciculus edge weight normalized and adjusted for intracranial volume; LDFR = long delay free recall; Recog. Discr. = Recognition discriminability. BNT = Boston Naming Test; AI = Association Index from semantic fluency. Note: Running two-way bivariate correlations did not change the significance of the results.
100
Table 3-8. Verbal memory and disease severity correlations
Measure UPDRS Total L.E.D.
PrVLT LDFR -0.30** -0.22
PrVLT Savings -0.09 0.00
PrVLT R. Discr. -0.39** -0.22
Story Delay Free -0.31** -0.17
Story Savings -0.27* -0.14
Story R. Discr. -0.36** -0.25*
* p < 0.05, ** p < 0.01. LDFR = long delay free recall; R. Discr. = recognition discriminability; UPDRS Total = total score on the United Parkinson’s disease Rating Scale; L.E.D. = Levodopa Equivalence Dosage score.
101
A B
C D Figure 3-1. Verbal memory index scores showing significant group difference (PD <
Controls). A) Mean PrVLT Long Delay Free Recall (PD = 8.13 [0.35]; Control = 9.65 [0.34]), B) Mean PrVLT Recognition Discriminability (PD = 3.01 [0.10]; Control = 3.49 [0.09]), C) Mean Story Delayed Free Recall (PD = 24.80 [1.09]; Control = 28.60 [1.07]), and D) Mean Story Recognition Discriminability (PD = 84.08 [1.37]; Control = 90.74 [1.35]).
102
Figure 3-2. Image showing cumulative frequency percents for PD participants across six verbal memory index scores. Blue = frequency of Not Impaired, Green = Mild Impairment, Orange = Moderate Impairment, and Red = Severe Impairment.
103
Figure 3-3. Chart showing all PD participants’ (n = 40) cumulative z score impairment on 3 PrVLT index scores. Higher bars indicate more impairment. Z scores were only included if they were < 0. The absolute values of negative z scores were taken and z scores summed across 3 PrVLT indices. *cepk005 was missing PrVLT data.
104
Figure 3-4. Chart showing all PD participants’ (n = 40) cumulative z score impairment on 3 Story index scores. Higher bars indicate more impairment. Z scores were only included if they were < 0. The absolute values of negative z scores were taken and z scores summed across 3 Story indices.
105
Figure 3-5. Chart showing all PD participants’ (n = 40; *cepk005 was missing PrVLT data but is displayed above) cumulative z score impairment on 2 verbal memory measures (across 6 total index scores). Higher bars indicate more impairment. Z scores were only included if they were < 0. The absolute values of negative z scores were taken and z scores summed across both tests.
106
A
B C Figure 3-6. Images representative of MTL volumetric and fiber tracking results. A) A
representative image of the left entorhinal cortex, B) Image showing cingulum fibers between the entorhinal cortex and retrosplenial cortex; C) Alternate image showing fibers between entorhinal cortex and retrosplenial cortex.
107
A
B Figure 3-7. Images showing group X brain structure scatter plots. A) Group difference in
left entorhinal cortex volumes, B) Group difference in left ERC – RSC EW values. Note: The highest neuroanatomic variable values in A and B are not the same individual.
108
Figure 3-8. Image showing group X left AF normalized edge weight. Edge weight values were scaled by a factor of 1 X 106.
109
Figure 3-9. A representative image showing the AF fiber tracking.
110
Figure 3-10. Scatter plot showing relationship between left ERC – RSC EW and PrVLT recognition discriminability index scores. Trend line and 95% confidence interval depicted.
111
A
B Figure 3-11. Images showing scatter plots with trend line and 95% confidence intervals
relating left ERC volume and Story memory index scores. A) Left ERC / ICV and Story savings, B) Left ERC / ICV and Story recognition discriminability.
112
Figure 3-12. Image showing scatter plot with trend line and 95% confidence intervals
relating left ERC volume and verbal memory composite score.
113
CHAPTER 4 DISCUSSION
Aim 1: Is there a Verbal Memory Deficit in PD?
The current study involving a group of non-demented individuals with idiopathic
Parkinson’s disease indicates that there is reason for the concern many individuals with
PD have about memory; memory complaints in PD are not solely subjective.
Participants with Parkinson’s disease, on average, recalled fewer words from a learned
list after a 25-minute delay compared to matched Control participants. They also
endorsed fewer words and made more false positives than matched peers during a
delayed recognition trial. On a separate verbal memory task, PD participants also
recalled fewer elements from short stories after a delay than matched Controls and had
greater difficulty correctly recognizing what they previously heard.
Processing speed was a contributor to the memory differences and accounted for
significant variance in the overall multivariate model but univariate results demonstrate
that slowed processing speed cannot fully account for the changes in memory. These
findings of verbal memory deficits match with previous work focused on memory in PD
(e.g., Filoteo et al., 1997) but extend past research by demonstrating deficits on multiple
verbal memory measures, which provides stronger evidence of significant memory
deficits (Zahodne et al., 2011). In contrast to verbal memory deficits, there were no
differences in semantic abilities between PD and Control participants. Thus, mild to
moderate verbal memory deficits occur relatively early in the PD disease process and in
the absence of deficits in language or global cognition; these memory deficits cannot be
fully explained by processing speed declines.
114
Verbal memory differences. Individuals with PD recalled fewer words (average
of 1.5) after a delay. Disruptions at different stages along the learning and memory
process can each affect performance. It is possible that the group difference in
remembering a list of words could be explained in part by PD participants’ greater
difficulty in learning words compared to Controls.1 If PD participants struggle more when
learning information, these difficulties might be due to either amnestic or dysexecutive
memory deficits or a combination of both. Individuals with amnestic memory deficits
display less learning, worse recall, and poor recognition of previously seen items. In the
present study, individuals with PD recalled fewer words after a delay and had poorer
recognition of words relative to their peers.
Individuals with amnestic memory deficits also experience rapid forgetting of
learned words. However, in the present study there was no evidence of differences in
the ability to retain learned words between groups. Even though there were no
statistically significant group differences, a significant subset of PD participants (see
Figure 3-2) had difficulty retaining learned information on both list-learning and story
measures.2 Therefore, at least a sub-group of PD patients appears to have difficulty
retaining learned information over a delay. This idea of subgroups of PD participants
with amnestic difficulties will be explored later in this discussion.3
1 PD participants learned an average of 1 fewer words per trial; e.g., Trial 5 Control mean: 10.78, Trial 5 PD mean: 9.70, p < 0.01
2 Using a typical clinical cut-off score of z = -1.5 (calculated using the matched controls as norms) 20.51% and 20% of individuals with PD had increased rates of forgetting of learned words and stories, respectively.
3 It should be noted that the level of impairment and rates of impairment on Story Savings are higher than what is seen using traditional norms such as from the Wechsler Memory Scale. Using norms from the Wechsler Memory Scale – III, no PD participants scored more than 1.5 standard deviations below the mean and only 5% (2/40) had Story Savings greater than 1 standard deviation below the mean. Thus,
115
Even though there might be amnestic memory deficits among the PD patients, the
results so far could be explained largely by dysexecutive/frontal difficulties in PD. There
is, however, further evidence that the memory deficits in PD are not entirely explained
by dysexecutive difficulties. Individuals with PD have significant reductions in
processing speed compared to controls, yet when controlling for processing speed,
group differences in individual verbal memory index scores remain. Processing speed
and executive function are closely related and affected similarly by pathology (see for
example, Prins et al., 2005). Processing speed is a major component of executive
function and is thought to be dependent upon the frontostriatal system and dopamine,
which are disrupted in Parkinson’s disease (Gabrieli, Singh, Stebbins, & Goetz, 1996).
However, controlling for deficits in processing speed does not explain all reductions
seen in verbal memory in PD.
Individuals with amnestic memory deficits also typically produce extra, unlearned
words after a delay (Canolle et al., 2008, Delis et al., 1991). While not part of the main
analyses for this study, a follow-up analysis (see Chapter 3 Footnote 1) demonstrated
that PD participants produce extra unlearned words on list recall and endorse more
semantically-related false positives during recognition testing than their matched peers
make and endorse, which is suggestive of amnestic-type difficulties in PD.
Adding to the case for amnestic difficulties in PD is the finding that individuals with
PD had more difficulty than Controls on a story memory task, a test that is believed to
without using carefully matched Controls, deficits are underestimated (Type II error). These findings seem to demonstrate the benefit of using normative data based on a closely matched control group rather than using population-based norms, particularly in a sample such as the one for the present study that includes individuals who are on the whole more educated than the general population. Using population-based norms can result in severe underestimates of impairment.
116
be relatively resistant to executive deficits (Lezak, Howieson, & Loring, 2004; Price et
al., in revision). Parkinson’s disease participants were impaired recalling the stories
after a delay as well as correctly answering yes/no questions about the stories. As
previously mentioned, 20% of PD participants had rapid forgetting of the stories relative
to Control participants. While not statistically significant, a subset of individuals with PD
did not retain as much story information. This reduction could have significant clinical,
ecological, and subjective implications (e.g., tracking conversations and remembering
them afterward might be more difficult for individuals with PD). Not retaining information
is common in amnestic memory deficits (Delis et al., 1991).
The results from the present study and from past research provide evidence of
amnestic difficulties affecting individuals with PD. One recent study found that while
there appear to be deficits in both familiarity and recollection in PD that are linked with
executive aspects of memory, the researchers also found evidence of recollection and
familiarity deficits that were MTL-linked and thus ‘pure’ memory difficulties (Cohn,
Moscovitch, & Davidson, 2010). On the other hand, while not specific to PD, individuals
with focal frontal lobe lesions have been shown to have poorer delayed recall, increased
intrusion rates, and impaired recognition memory on a list-learning test (Baldo, Delis,
Kramer, & Shimamura, 2002), which generally matches the memory profile expected
following damage to the medial temporal lobe. While this potentially makes it difficult to
classify memory difficulties as amnestic or dysexecutive, the findings of Baldo et al.
point to the importance of demonstrating MTL changes when making conclusions about
the nature or type of memory deficits.
117
The results from the present study and others demonstrate that there likely exist
subtypes of PD-MCI where some individuals have primarily memory or a combination of
memory and executive difficulties, others have mainly executive deficits, and still others
are relatively intact cognitively. On a positive note, the findings indicate that at an
average of 7 years after diagnosis, significant numbers of individuals with PD have
relatively unaffected memory.
There are no apparent relationships between disease severity and verbal memory
performance in individuals with PD so impairments in verbal memory cannot be entirely
explained by disruptions to dopaminergic systems that are believed to be driving many
of the motor and related symptoms of Parkinson’s disease. In the context of intact
global cognition, PD participants have deficits on multiple verbal memory measures.
This multi-measure impairment indicates that not only is memory an early concern in PD
but also that memory difficulties might not be wholly attributable to executive deficits.
Conversely, it might demonstrate that the Story memory task is more affected by
processing speed and executive processes than popularly believed. Even if that is the
case, the results of the present study contribute to the building evidence that a
subgroup – around 20% – of individuals with PD appear to experience amnestic
memory deficits (Cohn et al., 2010; Filoteo et al., 1997). Findings of smaller entorhinal
volumes in PD support this conclusion.
Aim 2: Are there Differences in Verbal Memory Brain Structures for PD?
Left entorhinal volumes were 11% smaller in PD participants than in matched
Controls when correcting for head size. A corresponding difference, however, in the
strength of white matter connections between the entorhinal cortex and retrosplenial
cortex area was not found. As expected, there were also no group differences in the
118
strength of the connection between the left middle temporal gyrus and the pars
operculus. While reductions in entorhinal cortex volumes in non-demented PD have
been found previously (see for example, Goldman et al., 2012), this study adds to the
evidence of structural changes to the MTL in non-demented individuals with Parkinson’s
disease. These medial temporal lobe changes can be interpreted as providing indirect
evidence of PD Lewy body, Lewy neurite, amyloid-Beta pathology, and α-synuclein
pathology. Reductions in entorhinal volume might even substitute as an in vivo
pathological burden. However, without direct pathological confirmation, this is only
speculative.
Causes of Entorhinal Volume Loss
According to the Braak staging hypothesis (Braak et al., 2004) the MTL is a region
of the brain that is affected early by PD pathology. Some have proposed that PD
pathology might have two originating regions – olfactory bulb and lower medulla
oblongata. Hawkes, Del Tredici, and Braak (2007) formulated this into a dual-stage
hypothesis for PD – namely that PD pathology starts at both olfactory and medullary
regions, spreading to adjacent regions until the whole brain is affected. Researchers
have shown that early clinical symptoms can be explained by these sites of pathological
disruption. For example, olfactory deficits are common in early Parkinson’s disease as
are sleep disorders, autonomic disorders, and gastrointestinal dysfunction (Hawkes et
al., 2007). These early non-motor clinical symptoms, which pre-date motor symptoms,
match the sites in PD most affected early by Braak stage pathology. Once clinical motor
symptoms of PD occur (typically between Braak stages 3 and 4), the MTL is already
affected by PD pathology, possibly even doubly so as a result of being a potential
convergence site between the olfactory bulb and the medulla oblongata. Neurons in the
119
entorhinal cortex, Ammon’s horn of the hippocampus, and related components of the
limbic system (e.g., amygdala) are particularly vulnerable to PD pathology (Braak &
Braak, 2000; Braak et al., 2004). It is also possible that individuals with PD and
cognitive and memory impairments have more limbic and neocortical pathology than
individuals without clinically significant non-motor symptoms (Jellinger, 2009); this
indicates that potential motor and cognitive subtypes of PD might have pathological
causes.
With these pathological changes, it is possible that there are concomitant
structural changes; however, pathology can result in severe functional impairment long
before considerable neuronal death and atrophy occur (Braak et al., 2000). Even so,
while the present study does not include direct pathological measurements, there is
evidence of volume loss in the entorhinal cortex. Futher, Lewy body aggregation in the
entorhinal cortex is predictive of cognitive functioning and dementia within Parkinson’s
disease (Kovari et al., 2003). Individuals with PD who have smaller entorhinal volumes
are also more likely to have dementia (Goldman et al., 2012). Thus, within the present
sample it is possible that individuals with reductions in ERC volumes might be at greater
risk for developing dementia in the future. Quantifying entorhinal and other MTL
volumes early in the disease process could help inform clinical diagnosis and treatment
and help increase accuracy of predicting future cognitive decline.
While there is considerable evidence to support the Braak stages of PD pathology,
there is controversy over the clinical utility of the stages (for a review, see Jellinger,
2009). In short, Braak-proposed PD pathology stages do not relate to clinical
manifestations of PD; in other words, severity of Braak stage PD pathology is not
120
predictive of severity of clinical symptoms (Jellinger, 2008; Jellinger, 2009; Beach et al.,
2009). Additionally, 30-55% of older adults exhibit widespread Lewy body and Lewy
neurite pathology without meaningful clinical phenotypical expression (Jellinger, 2009).
What might be most helpful in predicting severity of clinical symptoms is a combination
of Lewy body, Lewy neurite, α-synuclein, and neuronal cell loss analyses; in other
words, quantifying structural and regional neuronal loss in addition to pathology should
increase sensitivity and specificity to clinical stages of PD and cognitive impairments
within PD. This seems to imply that when pathological studies are not possible (e.g., in
living subjects), decreases in neuronal integrity and neuronal loss might be the most
sensitive measure of clinical dysfunction. This gives further support to the idea of
consistently quantifying structural brain changes for clinical use.
Explanations of ERC-RSC Findings
While it was expected that there would be reductions in PD, there were no
significant group differences in the connectivity between the entorhinal cortex and the
retrosplenial cortex. It might be the case that there are no gross disruptions in the
temporal portion of the cingulum connecting entorhinal and retrosplenial cortices but
there are a number of possible explanations and confounding factors that need to be
addressed.
If there are no group differences in edge weight it might be that this particular
white matter pathway is, in early stages of non-demented PD, not strongly affected by
the PD disease process. While there is good rationale for the disruption of this pathway
in PD (Mattila et al., 2001; Park & Stacy, 2009; Metzler-Baddeley et al., 2012; Lee et al.,
2010; Kobayashi & Amaral, 2003), it is possible that such disruptions have not yet
translated into gross structural changes of the white matter. In other words, PD
121
pathology, functional (neurotransmitter) changes, and loss of entorhinal cortex volume
might not have disrupted the white matter connecting the entorhinal cortex and the
retrosplenial cortex.
It is also possible that only a subgroup of individuals with PD has reduced ERC –
RSC connectivity. There are, however, no individuals with PD who have ERC – RSC
edge weights that are more than 1 standard deviation lower than the mean EW of
Controls. This relatively tight distribution of values does not necessarily mean that edge
weights are unimpaired – a small reduction in edge weight might be significant as a
proxy for pathology and significant clinically. This issue will be addressed as part of Aim
3.
From a qualitative standpoint, tracking results match known anatomical
connections. This method of fiber tracking is novel and involves a number of technical
and complex steps for analysis. Steps were taken along each part of the processing of
data to minimize error. However, because quantitative fiber tracking is a relatively new
method and because it relies on a complex interplay between biology and technology,
there are a number of technical difficulties in fiber tracking that potentially affected the
results. Such potential difficulties will be discussed later.
In summary, left entorhinal volumes were 11% smaller in PD participants than in
matched controls when correcting for head size. A difference, however, in the strength
of white matter connections between the entorhinal cortex and retrosplenial cortex was
not found. This provides evidence of MTL cortical loss in non-demented individuals with
idiopathic PD but relatively intact white matter connecting to the entorhinal cortex.
122
Aim 3: Are there Specific Structural-Cognitive Patterns?
As expected, across all subjects there was a positive association between
increasing left ERC volume and the ability to retain learned story information and
correctly answer questions about the stories. These findings replicate previous research
demonstrating associations between entorhinal volume and Story memory (Price et al.,
2011). Generally, past research has shown a relationship between the volume of the
entorhinal cortex (Killany et al., 2002; Rodrigue & Raz, 2004) or activity of the entorhinal
cortex (Goto et al., 2011; Cabeza, Dolcos, Graham, & Nyberg, 2002) and performance
on verbal memory measures (Price et al., 2010). MTL structures are necessary for the
process of making information usable at a later time (Squire, Stark, & Clark, 2004).
While the functions of the individual structures (e.g., entorhinal cortex, hippocampus, or
perirhinal cortex) can be dissociated from each other to some extent, there are not
always clear distinctions between what each structure does (Squire et al., 2004) so
finding clear associations between structure and cognitive function is not always
possible.
While the volume of the entorhinal cortex was important in remembering a
passage of prose, the integrity of the white matter between the entorhinal and
retroplenial cortices associated with the ability of older adults to correctly recognize
previously heard words from a list. This suggests distinct differences in gray versus
white matter function for verbal memory performance in the PD and Control samples.
When relationships between structure and memory were assessed for both groups
separately, for both groups left ERC volume was a significant predictor of the ability to
retain learned story information. When groups were analyzed separately, the positive
association between the integrity of the white matter between the left entorhinal and
123
retrosplenial cortices and list recognition remained only for the PD group. The
involvement of this temporal portion of the cingulum in recognition memory
demonstrates the role of brain networks in memory processes.
The recognition score analyzed was taken from signal detection theory. In order to
perform well, an individual has to not only correctly recognize target words (true
positives) but also correctly reject distractor words (true negatives). There is evidence of
reduced parietal glucose metabolism early in Alzheimer’s disease. Some have
speculated that the interface between temporal and parietal lobes is important for
normal memory functioning (Buckner, 2004). It has been hypothesized that the
retrosplenial cortex serves as an interface between working memory, which is highly
dependent on frontal and parietal regions and long-term memory; this is in turn, highly
dependent on the medial temporal lobe (Kobayashi & Amaral, 2003; Buckner, 2004).
Given past research demonstrating frontal lobe lesions resulting in reduced
memory performance, including increased intrusions (i.e., “false memories”; Baldo et al.,
2002), and given the close connections the retrosplenial cortex (and cingulum) has with
the frontal lobes, it is not surprising the long association pathway of the cingulum plays
a role in recognition – being able to accurately choose target words and suppress
distractions.
Whereas there was a significant relationship between the left ERC – RSC
connectivity and performance on PrVLT recognition discriminability, there was no
relationship between edge weight of the left arcuate fasciculus and PrVLT recognition
discriminability. These results show a dissociation between fiber pathways and their
relationship to a component of verbal memory.
124
In summary, individuals with larger entorhinal cortex volumes – relative to
intracranial volume – retain and recognize more elements from a prose passage than
do those who have smaller entorhinal cortex volumes. In short, the entorhinal cortex is
involved in preventing decay of learned prose information. Further, individuals who have
stronger white matter connections between the entorhinal cortex and retrosplenial
cortex perform better on a verbal recognition memory task. The current project was
designed as a start to better understand the role that structures and networks play in
verbal memory processes in individuals with Parkinson’s disease.
Final Summary
The results from the present study provide conclusive evidence of both early
memory disruptions and entorhinal volume loss in early stage, idiopathic, non-demented
PD. The verbal memory changes are partially due to deficits in processing speed and
executive functions but the results of the present study lend support to the idea that mild
but pure amnestic deficits, at least for a subset of individuals with Parkinson’s disease,
exist early in the Parkinson’s disease process. Further, these verbal memory deficits
occur without concomitant semantic system degradation as measured by category
fluency and naming tests in the context of intact global cognition. The present study has
five major implications.
Study design and verbal memory. First, these memory deficits in PD were
discovered in part due to the well-controlled nature of the study. In other words, the
present study demonstrated the need for good comparison groups. It is likely that
without a carefully matched control group, verbal memory impairment would be
understated. An example of this is looking at rates of Story memory impairment using
test norms rather than Control group norms. Using demographically-corrected test
125
norms, no individuals with PD scored less than 1.5 standard deviations below the mean
for Story savings (percent retention), yet deriving normative data from the Control group
yielded 20% of individuals with PD having Story savings scores 1.5 standard deviations
or more below the Control mean. This means that without using carefully matched
Control groups, Type II error rates are likely higher than desired.
Second is that memory impairment in non-demented PD is marked across two
verbal memory measures – list-learning and Story memory. While both measures
assess verbal memory, both are differentially affected by other cognitive processes.
Having deficits on multiple verbal memory measures provides stronger evidence of real
memory deficits (Janecek et al., 2011; Zahodne et al., 2011) by reducing spurious
effects related to multiple comparisons. Thus, controlling for multiple comparisons was
done at the model level by including a specific set of test indices hypothesized to
associate with neuroanatomical variables. Statistical corrections for multiple
comparisons were also conducted but did not significantly change the results and were
not reported.
Third, even though included individuals with Parkinson’s disease were cognitively
intact, scoring similarly to Controls on global measures of cognition and orientation,
individuals with Parkinson’s disease had smaller left entorhinal cortex volumes when
correcting for intracranial volume. While the data are cross-sectional and volumetric
changes over time cannot be measured, assuming similar entorhinal volumes before
PD and other neurological pathology took hold, individuals with PD lost on average 11%
of the volume of the entorhinal cortex. While a number of individuals with PD had intact
entorhinal volumes, overall, Control individuals had the largest ERC volumes and
126
individuals with PD had the smallest ERC volumes. These results seem to provide
indirect evidence of Lewy body and other pathological changes occurring in the medial
temporal lobe before significant cognitive declines occur. Structural changes are
important to quantify because individuals with and without PD who have smaller ERC
volumes could be at risk of developing significant memory problems as they age
(Rodrigue & Raz, 2004).
Fourth, and related to the previous point is that for memory measures there were
individuals with PD who scored and measured similarly to their Control peers (15-18%
of PD individuals scored at or above the Control mean for verbal memory measures). In
fact, the majority of individuals with PD in this study did not have significant memory
deficits. On the other hand, there was a significant subgroup who had verbal memory
deficits. 20-40% of individuals with PD demonstrated at least moderate impairment on
one or more verbal memory index score and 20% (8/40) had moderate verbal memory
impairment across both list learning and story memory measures (17.5% with a stricter
definition of impairment across both measures). This demonstrates that at least
moderate memory deficits are common early in PD without any global cognitive deficits.
It also demonstrates, on the other hand, that in the absence of global cognitive changes
significant proportions of individuals with PD do not have significant memory difficulties.
Fifth, including both gray and white matter is important for understanding the
distributed nature of cognition and the effects that pathological changes of neurological
disorders have on the brain. That group differences were seen in entorhinal volume but
not the cingulum between entorhinal and retrosplenial cortices might reflect pathological
progression (i.e., staging). In other words, significant PD pathological changes might not
127
have spread beyond the brainstem and MTL, at least to the point of significantly
affecting the structure of this section of the cingulum. These results might also signify
different levels of α-synuclein and Lewy bodies or neurites by region.
Limitations and Strengths
Verbal memory and Parkinson’s disease. Participants in the present study were
well-matched between groups. They averaged 67.9 years old, tended to be highly
educated (16.39 years average), had intact global cognition (MMSE = 29.09; DRS-2 =
139.84), had similar estimated premorbid vocabulary skills (WTAR = 108.18), and were
without significant health concerns (other than Parkinson’s disease). The lack of
significant health concerns was due to strict inclusion and exclusion criteria. Because
the groups are well-matched and participants are generally healthy, it is likely that this
resulted in an increased sensitivity to memory disturbances in PD (i.e., reduction in
Type II error rates) without overestimating memory difficulties. This is evident in that a
portion of PD participants only had significant deficits on the Story memory task when
using scores from the matched Controls as norms. When compared to the general age-
matched population using test publisher norms (Wechsler, 1997), none of the
participants with PD experienced significant impairments on the Story memory test. In
other words, impairments in Story memory were only seen when using a carefully-
matched Control group.
While the groups were well-matched, a potential limitation of the findings is that
the participants are not necessarily representative of the broader population and so the
findings of this study might not be generalizable to the general PD population. However,
in general, the careful matching of individuals with PD with Control participants is a
great strength of this project. This is particularly true because rates of memory problems
128
clinically and scientifically might be under-reported when relying on published norms.
This is not to question the validity or usefulness of established norms, rather it is to
argue for the utility of using norms based on closely-matched groups.
An additional limitation is that the included individuals were all Caucasian, thus
limiting further generalizability. However, because both groups were well-matched, this
controls for demographic effects in between-group analyses. In other words, because
groups were well-matched, between-group differences are broadly generalizable but
within-group findings will have limited applications to other groups of people.
Another potential limitation is that the included individuals with Parkinson’s disease
were generally healthy, as the study had strict inclusion and exclusion criteria. While
this results in ‘clean’ data, individuals in general society are not always as healthy. This
limitation, however, should not affect between-group results. What having well-
controlled groups does is increase the probability that deficits seen in individuals with
PD are due to the disease process and not extraneous comorbidities. Having ‘clean’
groups does limit the potential variability in verbal memory scores. This variability is also
limited because individuals were recruited only if they had intact global cognition at
baseline evaluation. While this allowed for a greater understanding of verbal memory
relatively early in the Parkinson’s disease process, it possibly limited the relationships
with neuroanatomical variables. This speaks to performing larger and longitudinal
studies or ones that include wider ranges of cognitive variability.
Related to this, while it was expected that there would be stronger relationships
between the volume of the entorhinal cortex and individual scores from the included
verbal memory tests, it is possible that other structures are more important for the
129
measured components of verbal memory than the entorhinal cortex is in this particular
sample. In other words, it is possible that other brain regions or functions become more
salient in the memory process in a mixed population of PD and healthy older adults
(Cabeza, 2002). This might be the case because as one area (e.g., entorhinal)
dysfunctions, other areas might start to compensate for that dysfunction. Researchers
assessing memory and other cognitive functions in cognitively intact older adults have
demonstrated that there appears to be less functional asymmetry in older adults
compared to younger adults (termed Hemispheric Asymmetry Reduction in Older Adults
{HAROLD}). It is beyond the scope of this dissertation, however, to address this
hypothesis but it is interesting to speculate whether older adults in general (particularly
neurologically compromised individuals) might not rely on broader brain networks for
cognitive and memory functions, thus reducing relationships between individual
components of networks (e.g., entorhinal cortex or entorhinal to retrosplenial white
matter) and memory and cognition. This hypothesis gives support for continuing to
measure and quantitatively track more and more complex brain areas and networks.
Neuroimaging. The following potential shortcomings are largely true for any
diffusion MRI analysis and are not unique to this project but are discussed briefly in
order to underscore the complexity of diffusion analyses.
Diffusion MRI is susceptible to artifacts as the result of subject motion (Leemans &
Jones, 2009; Rohde et al., 2003), head position in the scanner, Echo Planar Imaging
(EPI) distortion (Bammer et al., 2002), and partial volume averaging as the result of
voxel size (Vos et al., 2011). Voxel size is particularly important when seeking to
visualize fiber pathways that are similar in width to or smaller than the in-plane
130
resolution of the diffusion weighted imaging voxels. Whereas in the present analysis 64
streamlines (‘fibers’) were calculated for each 8 mm3 cube of brain matter, there are
thousands to tens of thousands axons in the same space. Because of methodological
limitations to visualizing and analyzing the white matter, there is a need for fiber tracking
at higher spatial resolutions (which will be discussed briefly at a later point) in order to
reduce partial volume effects. Higher resolutions should increase sensitivity to change
in the white matter; it does, however, come at a cost of time, signal-to-noise, movement,
and other factors (Pfefferbaum et al., 2004). For a concise review of difficulties of white
matter diffusion analyses refer to Jones and Cercignani’s (2010) recent article.
Fiber tracking is a relatively new method of brain imaging analysis. While the
underlying mathematical model and method of tracking used in this analysis is
established (Jian et al., 2007), mapping of the fiber networks and connections between
regions is an area under development and not without challenges (see Jones, 2010, for
a review of challenges faced by researchers performing fiber tracking analyses). Refer
to Figure 4-1 for examples of fiber tracking with sub-optimal results (compare to Figure
3-9). The fibers match anatomic structures but the number of calculated streamlines is
limited.
In spite of potential limitations to diffusion imaging in general and fiber tracking
specifically, the tracking results match known anatomy; further, reasonable steps were
taken in order to minimize error so tracking results are thought to be valid. Further
reliability and validity studies are planned in future work in order to continue to improve
methodology.
131
Future Directions
It will be important to look at other parts of the memory network in order to start
to build a more complete picture of the relationships between segments of memory
networks and components of verbal memory. Future work can include tracking between
entorhinal cortex and the hippocampus (perforant pathway; see Augustinack et al.,
2010; Christidi et al., 2011, for example), tracking the fornix, tracking the uncinate
fasciculus, and other brain regions. It is technically possible to track the entire Papez
circuit in order to see relationships between each set of connection diodes (e.g., ERC to
RSC, ERC to hippocampus, hippocampus to mammillary bodies, etc.) and verbal
memory index scores. It could also be important to measure the integrity of the entire
system to assess for relationships with verbal memory. These analyses are important
for detecting early brain changes that might predict future memory decline, or at least
discover individuals at greater risk of developing memory deficits in the future and as
such have significant clinical relevance.
Future directions also speak to the need to better understand the relationship
between PD pathology progression, verbal memory, and structural brain changes. Such
efforts will require close multidisciplinary efforts. All are important research areas in
order to better understand the in vivo progression of Parkinson’s disease and other
neurodegenerative disorders with the goal of earlier and more effective interventions.
132
A
B Figure 4-1. Representative images showing suboptimal arcuate fasciculus fiber tracking.
A) participant with lowest AF edge weight – note the few streamlines calculated; B) participant with low AF edge weight – also has few streamlines.
133
APPENDIX A PARTICIPANT INCLUSION AND EXCLUSION CRITERIA
Diagnosis
o Non-demented Idiopathic PD ‘on’ medication
o Diagnosis based on United Kingdom PD Society Brain Research Center criteria (Gibb & Lees, 1988)
o Hoehn and Yahr scale range 1-3 (Hoehn and Yahr, 1967).
Patients are excluded if they match criteria for comorbid neurodegenerative disorders (see below)
‘Normal’ Age/Education Matched Adults
See exclusion criteria below
Ethnicity/race: all ethnic and racial groups.
Sex/gender: men and women
Age: 60 and up. There is no upper age limit imposed.
Handedness: Right handed
PD Exclusion Criteria
Underlying medical disease likely to limit lifespan or confound outcome analyses.
Other neurodegenerative disorders.
Patients are excluded at baseline if they present with signs of a dementia as indicated by DSM-IV criteria and a Dementia Rating Scale-2nd Edition age and education corrected scale score < 8.
Psychiatric Exclusions: A major psychiatric disorder. Also, patients who meet criteria for Major Depression or experience a Major Depressive Episode within three months prior to study recruitment will be excluded. We did not exclude patients reporting mild depression or anxiety for many PD patients report such symptoms.
Conditions or behaviors (e.g., claustrophobia) likely to affect imaging or cognitive testing.
134
Parallel Control Group Exclusion Criteria
Exclusion criteria for this group match that of the PD patients except that it will be
required that they do not have symptoms of PD. Scores on the Dementia Rating Scale-
Revised will be within the average range [age and education scale score > 8; (Jurica et
al., 2001)] and scores on the Mini Mental State Exam will be in the normal range [total
score > 27; (Folstein et al., 1975; Lezak, 1995)].
135
APPENDIX B MAGNETIC RESONANCE IMAGE PROCESSING WORKFLOW
136
APPENDIX C AIM 2 WITHIN-GROUP SUB-AIM: SIDE OF ONSET AND LATERALIZED
DIFFERENCES
25/40 PD participants had right side symptom onset, 14/40 had left side onset,
and 1/40 experienced axial/gait changes first. When comparing left and right side onset
PD participants, there were no significant differences in brain structure volumes or fiber
integrity (see Table C-1).
Table C-1. Parkinson’s disease side of onset group contrast
Measure PD-R (n = 25) PD-L (n=14) P Value
Left ERC/ICV 6.94 (0.94) 7.30 (1.50) 0.36
Left ERC-RSC EW 0.55 (0.46) 0.33 (0.33) 0.12
Left AF EW 8.16 (7.08) 8.06 (8.15) 0.97
Note: ERC/ICV = Entorhinal cortex volume corrected for intracranial volume X 10000; ERC-RSC EW = Entorhinal cortex to retrosplenial cortex edge weight normalized and adjusted for intracranial volume; AF EW = Arcuate Fasciculus edge weight normalized and adjusted for intracranial volume.
137
REFERENCE LIST
Aarsland, D., Zaccai, J., & Brayne, C. (2005). A systematic review of prevalence studies of dementia in Parkinson's disease. Movement Disorders: Official Journal of the Movement Disorder Society, 20(10), 1255–1263.
Aarsland, D., Bronnick, K., Williams-Gray, C., Weintraub, D., Marder, K., Kulisevsky, J.,…Emre, M. (2010). Mild cognitive impairment in Parkinson disease: A multicenter pooled analysis. Neurology, 75(12), 1062–1069. doi:10.1212/WNL.0b013e3181f39d0e
Aarsland, D., Andersen, K., Larsen, J., Lolk, A., Nielsen, H., & Kragh-Sorensen, P. (2001). Risk of dementia in Parkinson's disease: A community-based, prospective study. Neurology, 56(6), 730.
Aggleton, J. P., & Brown, M. W. (1999). Episodic memory, amnesia, and the hippocampal-anterior thalamic axis. The Behavioral and Brain Sciences, 22(3), 425–44.
Aggleton, J. P., & Brown, M. W. (2006). Interleaving brain systems for episodic and recognition memory. Trends in Cognitive Sciences, 10(10), 455–463. doi:10.1016/j.tics.2006.08.003
Agosta, F., Pievani, M., Sala, S., Geroldi, C., Galluzzi, S., Frisoni, G. B., & Filippi, M. (2011). White matter damage in Alzheimer disease and its relationship to gray matter atrophy. Radiology, 258(3), 853–863. doi:10.1148/radiol.10101284
Alexander, G., DeLong, M., & Strick, P. (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience, 9(1), 357–381.
Anderson, J. R. (1976). Language, memory, and thought. Mahwah, NJ: Lawrence Erlbaum.
Augustinack, J. C., Helmer, K., Huber, K. E., Kakunoori, S., Zöllei, L., & Fischl, B. (2010). Direct visualization of the perforant pathway in the human brain with ex vivo diffusion tensor imaging. Frontiers in Human Neuroscience, 4, 42. doi:10.3389/fnhum.2010.00042
Baddeley, A., Della Sala, S., & Spinnler, H. (1991). The two-component hypothesis of memory deficit in Alzheimer's disease. Journal of Clinical and Experimental Neuropsychology, 13(2), 372–380.
Baddeley, A., & Wilson, B. (1988). Frontal amnesia and the dysexecutive syndrome. Brain and Cognition, 7(2), 212–230.
138
Baldo, J. V., Delis, D., Kramer, J., & Shimamura, A. P. (2002). Memory performance on the California Verbal Learning Test-II: Findings from patients with focal frontal lesions. Journal of the International Neuropsychological Society, 8(4), 539–546.
Bammer, R., Auer, M., Keeling, S. L., Augustin, M., Stables, L. A., Prokesch, R. W.,…
Fazekas, F. (2002). Diffusion tensor imaging using single‐shot SENSE‐EPI. Magnetic Resonance in Medicine, 48(1), 128–136.
Baran, B., Tekcan, A. I., Gürvit, H., & Boduroglu, A. (2009). Episodic memory and metamemory in Parkinson's disease patients. Neuropsychology, 23(6), 736–745. doi:10.1037/a0016631
Bartus, R. T., Dean, R. L., Beer, B., & Lippa, A. S. (1982). The cholinergic hypothesis of geriatric memory dysfunction. Science, 217(4558), 408–414.
Bastiaanse, R., & Leenders, K. L. (2009). Language and Parkinson's disease. Cortex, 45(8), 912.
Bayles, K. A., Tomoeda, C. K., Montgomery, E. B., Cruz, R. F., & Azuma, T. (2000). The relation of mental status to performance on lexical-semantic tasks in Parkinson's disease. International Journal of Speech Language Pathology, 2(2), 67–75. doi: 10.3109/14417040008996792
Beatty, W. W., Ryder, K. A., Gontkovsky, S. T., Scott, J. G., McSwan, K. L., & Bharucha, K. J. (2003). Analyzing the subcortical dementia syndrome of Parkinson's disease using the RBANS. Archives of Clinical Neuropsychology, 18(5), 509–520.
Beatty, W. W., Staton, R. D., Weir, W. S., Monson, N., & Whitaker, H. A. (1989). Cognitive disturbances in Parkinson's disease. Journal of Geriatric Psychiatry and Neurology, 2(1), 22–33. doi:10.1177/089198878900200106
Berlingeri, M., Bottini, G., Basilico, S., Silani, G., Zanardi, G., Sberna, M.,…Paulesu, E. (2008). Anatomy of the episodic buffer: A voxel-based morphometry study in patients with dementia. Behavioural Neurology, 19(1-2), 29–34.
Bernal, B., & Ardila, A. (2009). The role of the arcuate fasciculus in conduction aphasia. Brain: A Journal of Neurology, 132(9), 2309–2316. doi:10.1093/brain/awp206
Bohnen, N. I., & Albin, R. L. (2011). The cholinergic system and Parkinson disease. Behavioural Brain Research, 221(2), 564–573. doi:10.1016/j.bbr.2009.12.048
Braak, H., & Braak, E. (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathologica, 82(4), 239–259.
Braak, H., & Braak, E. (2000). Pathoanatomy of Parkinson's disease. Journal of Neurology, 247(2), II/3–II/10.
139
Braak, H., Ghebremedhin, E., Rüb, U., Bratzke, H., & Del Tredici, K. (2004). Stages in the development of Parkinson's disease-related pathology. Cell and Tissue Research, 318(1), 121–134. doi:10.1007/s00441-004-0956-9
Brandt, J., & Benedict, R. H. B. (2001). Hopkins Verbal Learning Test–Revised: Professional Manual. Lutz, FL: Psychological Assessment Resources.
Brickman, A. M., Stern, Y., & Small, S. A. (2010). Hippocampal subregions differentially associate with standardized memory tests. Hippocampus. doi:10.1002/hipo.20840
Broca, P. (1861). Remarques sur le siège de la faculté du langage articulé, suivies d’une observation d’aphémie (perte de la parole). Bulletin De La Société Anatomique, 36, 330–356.
Brown, M. W., Warburton, E. C., & Aggleton, J. P. (2010). Recognition memory: Material, processes, and substrates. Hippocampus, 20(11), 1228–1244. doi:10.1002/hipo.20858
Buckner, R. L. (2004). Memory and executive function in aging and AD: Multiple factors that cause decline and reserve factors that compensate. Neuron, 44(1), 195–208.
Budson, A. E., & Price, B. H. (2005). Memory dysfunction. The New England Journal of Medicine, 352(7), 692–699. doi:10.1056/NEJMra041071
Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychology and Aging, 17(1), 85–100. doi:10.1037//0882-7974.17.1.85
Calabresi, P., Picconi, B., Parnetti, L., & Di Filippo, M. (2006). A convergent model for cognitive dysfunctions in Parkinson's disease: The critical dopamine–acetylcholine synaptic balance. The Lancet Neurology, 5(11), 974–983.
Cameron, I. G. M., Watanabe, M., Pari, G., & Munoz, D. P. (2010). Executive impairment in Parkinson's disease: Response automaticity and task switching. Neuropsychologia, 48(7), 1948–1957. doi:10.1016/j.neuropsychologia.2010.03.015
Canolle, M., Messaoudi, M., Ayoub, B., Descours, I., Bocquet, P., Gely-Nargeot, M.-C., & Touchon, J. (2008). [Prototypic value of semantic intrusion errors in Alzheimer's disease]. Psychologie & Neuropsychiatrie Du Vieillissement, 6(1), 67–79. doi:10.1684/pnv.2008.0115
Carew, T. G., Lamar, M., Cloud, B. S., Grossman, M., & Libon, D. J. (1997). Impairment in category fluency in ischemic vascular dementia. Neuropsychology, 11(3), 400–412.
Catani, M., Jones, D. K., & ffytche, D. H. (2005). Perisylvian language networks of the human brain. Annals of Neurology, 57(1), 8–16. doi:10.1002/ana.20319
140
Catani, M., & Thiebaut de Schotten, M. (2008). A diffusion tensor imaging tractography atlas for virtual in vivo dissections. Cortex, 44(8), 1105–1132. doi:10.1016/j.cortex.2008.05.004
Cave, C. B., & Squire, L. R. (1992). Intact verbal and nonverbal short-term memory following damage to the human hippocampus. Hippocampus, 2(2), 151–163. doi:10.1002/hipo.450020207
Caviness, J. N., Driver-Dunckley, E., Connor, D. J., Sabbagh, M. N., Hentz, J. G., Noble, B.,…Adler, C. H. (2007). Defining mild cognitive impairment in Parkinson's disease. Movement Disorders: Official Journal of the Movement Disorder Society, 22(9), 1272–1277. doi:10.1002/mds.21453
Chen, J. J. (2010). Parkinson's disease: Health-related quality of life, economic cost, and implications of early treatment. The American Journal of Managed Care, 16(4), 87–93.
Chételat, G., Desgranges, B., La Sayette, de, V., Viader, F., Berkouk, K., Landeau, B.,…Eustache, F. (2003). Dissociating atrophy and hypometabolism impact on episodic memory in mild cognitive impairment. Brain, 126(9), 1955–1967. doi:10.1093/brain/awg196
Choo, I. H., Lee, D. Y., Oh, J. S., Lee, J. S., Lee, D. S., Song, I. C.,…Woo, J. I. (2010). Posterior cingulate cortex atrophy and regional cingulum disruption in mild cognitive impairment and Alzheimer's disease. Neurobiology of Aging, 31(5), 772–779. doi:10.1016/j.neurobiolaging.2008.06.015
Christidi, F., Bigler, E. D., Mccauley, S. R., Schnelle, K. P., Merkley, T. L., Li, X.,…Wilde, E. A. (2011). Diffusion tensor imaging of the perforant pathway zone and its relation to memory function in patients with severe traumatic brain injury. Journal of Neurotrauma. doi:10.1089/neu.2010.1644
Clément, F., Belleville, S., & Mellah, S. (2010). Functional neuroanatomy of the encoding and retrieval processes of verbal episodic memory in MCI. Cortex, 46(8), 1005–1015. doi:10.1016/j.cortex.2009.07.003
Cohn, M., Moscovitch, M., & Davidson, P. S. R. (2010). Double dissociation between familiarity and recollection in Parkinson's disease as a function of encoding tasks. Neuropsychologia, 48(14), 4142–4147. doi:10.1016/j.neuropsychologia.2010.10.013
Cooper, J. A., Sagar, H. J., Jordan, N., Harvey, N. S., & Sullivan, E. V. (1991). Cognitive impairment in early, untreated Parkinson's disease and its relationship to motor disability. Brain, 114, 2095.
Coutureau, E., & Di Scala, G. (2009). Entorhinal cortex and cognition. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 33(5), 753–761. doi:10.1016/j.pnpbp.2009.03.038
141
Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268–294.
Crosson, B. (1999). Subcortical mechanisms in language: Lexical-semantic mechanisms and the thalamus. Brain and Cognition, 40(2), 414–438. doi:10.1006/brcg.1999.1088
Cummings, J. (1986). Subcortical dementia. Neuropsychology, neuropsychiatry, and pathophysiology. The British Journal of Psychiatry, 149(6), 682. doi:10.1192/bjp.149.6.682
Cummings, J. L., & Benson, D. F. (1984). Subcortical dementia. Review of an emerging concept. Archives of Neurology, 41(8), 874–879.
Cummings, B. J., Head, E., Afagh, A. J., Milgram, N. W., & Cotman, C. W. (1996). Beta-amyloid accumulation correlates with cognitive dysfunction in the aged canine. Neurobiology of Learning and Memory, 66(1), 11–23. doi:10.1006/nlme.1996.0039
Damasio, A. R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed neural systems. Proceedings of the National Academy of Sciences of the United States of America, 90(11), 4957–4960.
Davidson, P. S. R., Anaki, D., Saint-Cyr, J. A., Chow, T. W., & Moscovitch, M. (2006). Exploring the recognition memory deficit in Parkinson's disease: Estimates of recollection versus familiarity. Brain, 129(7), 1768–1779. doi:10.1093/brain/awl115
Davis, K. L., Price, C. C., Kaplan, E., & Libon, D. J. (2002). Error analysis of the nine-word California Verbal Learning Test (CVLT-9) among older adults with and without dementia. The Clinical Neuropsychologist, 16(1), 81–89.
Delis, D. C. (2000). California verbal learning test (2nd Edition). San Antonio, TX: Psychological Corporation.
Delis, D. C., Massman, P. J., Butters, N., Salmon, D. P., Cermak, L. S., & Kramer, J. H. (1991). Profiles of demented and amnesic patients on the California Verbal Learning Test: Implications for the assessment of memory disorders. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 3(1), 19.
Della Sala, S., MacPherson, S. E., Phillips, L. H., Sacco, L., & Spinnler, H. (2004). The role of semantic knowledge on the cognitive estimation task. Journal of Neurology, 251(2), 156–164. doi:10.1007/s00415-004-0292-8
Diamond, B. J., Johnson, S. K., Kaufman, M., & Graves, L. (2008). Relationships between information processing, depression, fatigue and cognition in multiple sclerosis. Archives of Clinical Neuropsychology: The Official Journal of the National Academy of Neuropsychologists, 23(2), 189–199. doi:10.1016/j.acn.2007.10.002
142
Di Paola, M., Macaluso, E., Carlesimo, G. A., Tomaiuolo, F., Worsley, K. J., Fadda, L., & Caltagirone, C. (2007). Episodic memory impairment in patients with Alzheimer's disease is correlated with entorhinal cortex atrophy. A voxel-based morphometry study. Journal of Neurology, 254(6), 774–781. doi:10.1007/s00415-006-0435-1
Dobbins, I. G., Simons, J. S., & Schacter, D. L. (2004). fMRI evidence for separable and lateralized prefrontal memory monitoring processes. Journal of Cognitive Neuroscience, 16(6), 908–920.
Douaud, G., Jbabdi, S., Behrens, T. E. J., Menke, R. A., Gass, A., Monsch, A. U.,…Smith, S. (2011). DTI measures in crossing-fibre areas: Increased diffusion anisotropy reveals early white matter alteration in MCI and mild Alzheimer's disease. NeuroImage, 55(3), 880–890. doi:10.1016/j.neuroimage.2010.12.008
Dubois, B., Ruberg, M., Javoy-Agid, F., Ploska, A., & Agid, Y. (1983). A subcortico-cortical cholinergic system is affected in Parkinson's disease. Brain Research, 288(1-2), 213–218.
Duffau, H. (2008). The anatomo-functional connectivity of language revisited. Neuropsychologia, 46(4), 927–934. doi:10.1016/j.neuropsychologia.2007.10.025
Elbaz, A., Bower, J. H., Maraganore, D. M., McDonnell, S. K., Peterson, B. J., Ahlskog, J. E.,…Rocca, W. A. (2002). Risk tables for parkinsonism and Parkinson's disease. Journal of Clinical Epidemiology, 55(1), 25–31.
Fernandez, G., Brewer, J., Zhao, Z., Glover, G., & Gabrieli, J. (1999). Level of sustained entorhinal activity at study correlates with subsequent cued-recall performance: A functional magnetic resonance imaging study with high acquisition rate. Hippocampus, 9(1), 35–44.
Filoteo, J., Rilling, L., Cole, B., Williams, B., Davis, J., & Roberts, J. (1997). Variable memory profiles in Parkinson's disease. Journal of Clinical and Experimental Neuropsychology, 19(6), 878–888.
Filoteo, J. V., Salmon, D. P., Schiehser, D. M., Kane, A. E., Hamilton, J. M., Rilling, L. M.,...Galasko, D. R. (2009). Verbal learning and memory in patients with dementia with Lewy bodies or Parkinson's disease with dementia. Journal of Clinical and Experimental Neuropsychology, 31(7), 823–834. doi:10.1080/13803390802572401
Fischl, B., Stevens, A. A., Rajendran, N., Yeo, B. T. T., Greve, D. N., van Leemput, K.,...Augustinack, J. C. (2009). Predicting the location of entorhinal cortex from MRI. NeuroImage, 47(1), 8–17. doi:10.1016/j.neuroimage.2009.04.033
Fischl, B., Salat, D. H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C.,...Dale, A. M. (2002). Whole brain segmentation: Automated labeling of neuroanatomical structures in the human brain. Neuron, 33(3), 341–355.
143
Foltynie, T., Brayne, C. E. G., Robbins, T. W., & Barker, R. A. (2004). The cognitive ability of an incident cohort of Parkinson's patients in the UK. The CamPaIGN study. Brain: A Journal of Neurology, 127(3), 550–560. doi:10.1093/brain/awh067
Fonov, V. S., Evans, A. C., McKinstry, R. C., Almli, C. R., & Collins, D. L. (2009). Unbiased nonlinear average age-appropriate brain templates from birth to adulthood. NeuroImage, 47(S1), S102. doi:10.1016/S1053-8119(09)70884-5
Foster, P. S., Drago, V., Crucian, G. P., Skidmore, F., Rhodes, R. D., Shenal, B. V.,…Heilman, K. M. (2010). Verbal and visuospatial memory in lateral onset Parkinson disease: Time is of the essence. Cognitive and Behavioral Neurology: Official Journal of the Society for Behavioral and Cognitive Neurology, 23(1), 19–25. doi:10.1097/WNN.0b013e3181c20de7
Friederici, A. D. (2009). Pathways to language: Fiber tracts in the human brain. Trends in Cognitive Sciences, 13(4), 175–181. doi:10.1016/j.tics.2009.01.001
Galton, C. J., Patterson, K., Xuereb, J. H., & Hodges, J. R. (2000). Atypical and typical presentations of Alzheimer's disease: A clinical, neuropsychological, neuroimaging and pathological study of 13 cases. Brain: A Journal of Neurology, 123(3), 484–498.
Gattellaro, G., Minati, L., Grisoli, M., Mariani, C., Carella, F., Osio, M.,…Bruzzone, M. G. (2009). White matter involvement in idiopathic Parkinson disease: A diffusion tensor imaging study. American Journal of Neuroradiology, 30(6), 1222–1226. doi:10.3174/ajnr.A1556
Gauthier, S., Reisberg, B., Zaudig, M., Petersen, R. C., Ritchie, K., Broich, K.,…Winblad, B. (2006). Mild cognitive impairment. The Lancet, 367(9518), 1262–1270. doi:10.1016/S0140-6736(06)68542-5
Geschwind, N. (1971). Aphasia. New England Journal of Medicine, 284(12), 654–656.
Geschwind, N. (1972). Language and the brain. New York, NY: W. H. Freeman.
Gibb, W. R., & Lees, A. J. (1988). The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson's disease. Journal of Neurology, Neurosurgery, and Psychiatry, 51(6), 745–752.
Gold, J. J., & Squire, L. R. (2006). The anatomy of amnesia: Neurohistological analysis of three new cases. Learning & Memory, 13(6), 699–710. doi:10.1101/lm.357406
Goto, M., Abe, O., Miyati, T., Yoshikawa, T., Hayashi, N., Takao, H.,…Ohtomo, K. (2011). Entorhinal cortex volume measured with 3T MRI is positively correlated with the Wechsler Memory Scale-Revised logical/verbal memory score for healthy subjects. Neuroradiology, 53(8), 617–622. doi:10.1007/s00234-011-0863-1
144
Greve, D. N., & Fischl, B. (2009). Accurate and robust brain image alignment using boundary-based registration. NeuroImage, 48(1), 63.
Grossman, M. (1999). Sentence processing in Parkinson's disease. Brain and Cognition, 40(2), 387–413.
Hagmann, P., Kurant, M., Gigandet, X., Thiran, P., Wedeen, V., Meuli, R., & Thiran, J. (2007). Mapping human whole-brain structural networks with diffusion MRI. PLoS One, 2(7).
Halliday, G., Lees, A., & Stern, M. (2011). Milestones in Parkinson's disease—Clinical and pathologic features. Movement Disorders: Official Journal of the Movement Disorder Society, 26(6), 1015–1021.
Hardy, J. (2010). Genetic analysis of pathways to Parkinson disease. Neuron, 68(2), 201–206. doi:10.1016/j.neuron.2010.10.014
Hariz, G. M., & Forsgren, L. (2011). Activities of daily living and quality of life in persons with newly diagnosed Parkinson’s disease according to subtype of disease, and in comparison to healthy controls. Acta Neurologica Scandinavica, 123(1), 20–27.
Hawkes, C. H., Del Tredici, K., & Braak, H. (2007). Parkinson's disease: A dual‐hit hypothesis. Neuropathology and Applied Neurobiology, 33(6), 599–614.
Hawkes, C. H., Shephard, B. C., & Daniel, S. E. (1999). Is Parkinson's disease a primary olfactory disorder? Qjm, 92(8), 473–480.
Hebb, D. O. (1949). The Organization of Behavior. New York, NY: Wiley & Sons.
Helkala, E. L., Laulumaa, V., Soininen, H., & Riekkinen, P. J. (1988). Recall and recognition memory in patients with Alzheimer’s and Parkinson’s diseases. Annals of Neurology, 24(2), 214–217.
Hoehn, M. M., & Yahr, M. D. (1967). Parkinsonism: Onset, progression, and mortality. Neurology, 17(5), 427–442.
Holtgraves, T., & McNamara, P. (2010). Pragmatic comprehension deficit in Parkinson’s disease. Journal of Clinical and Experimental Neuropsychology, 32(4), 388–397.
Hoscheidt, S. M., Nadel, L., Payne, J., & Ryan, L. (2010). Hippocampal activation during retrieval of spatial context from episodic and semantic memory. Behavioural Brain Research, 212(2), 121–132. doi:10.1016/j.bbr.2010.04.010
Huber, S. J., Shuttleworth, E. C., & Freidenberg, D. L. (1989). Neuropsychological differences between the dementias of Alzheimer’s and Parkinson’s diseases. Archives of Neurology, 46(12), 1287–1291.
145
Hyman, B. T., Kromer, L. J., & Van Hoesen, G. W. (1988). A direct demonstration of the perforant pathway terminal zone in Alzheimer's disease using the monoclonal antibody Alz-50. Brain Research, 450(1-2), 392–397.
Hyman, B. T., Van Hoesen, G. W., Kromer, L. J., & Damasio, A. R. (1986). Perforant pathway changes and the memory impairment of Alzheimer's disease. Annals of Neurology, 20(4), 472–481. doi:10.1002/ana.410200406
Insausti, R., Amaral, D. G., & Cowan, W. M. (1987). The entorhinal cortex of the monkey: II. Cortical afferents. The Journal of Comparative Neurology, 264(3), 356–395. doi:10.1002/cne.902640306
Insausti, R., Juottonen, K., Soininen, H., Insausti, A., Partanen, K., Vainio, P.,...Pitkanen, A. (1998). MR volumetric analysis of the human entorhinal, perirhinal, and temporopolar cortices. American Journal of Neuroradiology, 19(4), 659-671.
Insausti, R. (1993). Comparative anatomy of the entorhinal cortex and hippocampus in mammals. Hippocampus, 3, 19–26.
Janecek, J. K., Rog., L. A., Prosje, M., Okun, M.S., Price, C. C., Bauer, R. M., & Bowers, D. (2011). MCI in Parkinson disease: Measures matter. Journal of the International Neuropsychological Society, 17(S1), i–334. doi:10.1017/S1355617711000415
Jellinger, K. A. (2009). Formation and development of Lewy pathology: A critical update. Journal of Neurology, 256(S3), 270–279. doi:10.1007/s00415-009-5243-y
Jellinger, K. A. (2010). Neuropathology in Parkinson's disease with mild cognitive impairment. Acta Neuropathologica, 829–830. doi:10.1007/s00401-010-0755-1
Jellinger, K. A. (1991). Pathology of Parkinson's disease. Changes other than the nigrostriatal pathway. Molecular and Chemical Neuropathology, 14(3), 153–197.
Jenkinson, M., Bannister, P., Brady, M., & Smith, S. (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. NeuroImage, 17(2), 825–841.
Jenkinson, M., & Smith, S. (2001). A global optimisation method for robust affine registration of brain images. Medical Image Analysis, 5(2), 143–156.
Jian, B., Vemuri, B. C., Ozarslan, E., Carney, P. R., & Mareci, T. H. (2007). A novel tensor distribution model for the diffusion-weighted MR signal. NeuroImage, 37(1), 164–176. doi:10.1016/j.neuroimage.2007.03.074
146
Jokinen, H., Kalska, H., Ylikoski, R., Hietanen, M., Mäntylä, R., Pohjasvaara, T.,… Erkinjuntti, T. (2004). Medial temporal lobe atrophy and memory deficits in elderly stroke patients. European Journal of Neurology: The Official Journal of the European Federation of Neurological Societies, 11(12), 825–832. doi:10.1111/j.1468-1331.2004.00870.x
Jokinen, P., Brück, A., Aalto, S., Forsback, S., Parkkola, R., & Rinne, J. O. (2009). Impaired cognitive performance in Parkinson's disease is related to caudate dopaminergic hypofunction and hippocampal atrophy. Parkinsonism & Related Disorders, 15(2), 88–93. doi:10.1016/j.parkreldis.2008.03.005
Jones, D. K. (2010). Challenges and limitations of quantifying brain connectivity in vivo with diffusion MRI. Imaging in Medicine, 2(3), 341–355.
Jones, D. K., & Cercignani, M. (2010). Twenty-five pitfalls in the analysis of diffusion MRI data. NMR in Biomedicine, 23(7), 803–820. doi:10.1002/nbm.1543
Kalaitzakis, M. E., Christian, L. M., Moran, L. B., Graeber, M. B., Pearce, R. K. B., & Gentleman, S. M. (2009). Dementia and visual hallucinations associated with limbic pathology in Parkinson's disease. Parkinsonism & Related Disorders, 15(3), 196–204. doi:10.1016/j.parkreldis.2008.05.007
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). Boston Naming Test. Philadelphia, PA: Lea & Febiger.
Karagulle Kendi, A., Lehericy, S., Luciana, M., Ugurbil, K., & Tuite, P. (2008). Altered diffusion in the frontal lobe in Parkinson disease. American Journal of Neuroradiology, 29, 501–505.
Kehagia, A. A., Barker, R. A., & Robbins, T. W. (2010). Neuropsychological and clinical heterogeneity of cognitive impairment and dementia in patients with Parkinson's disease. Lancet Neurology, 9(12), 1200–1213. doi:10.1016/S1474-4422(10)70212-X
Kim, H.-J., Park, S.-Y., Cho, Y.-J., Hong, K.-S., Cho, J.-Y., Seo, S.-Y.,…Jeon, B. S. (2009). Nonmotor symptoms in de novo Parkinson disease before and after dopaminergic treatment. Journal of the Neurological Sciences, 287(1-2), 200–204. doi:10.1016/j.jns.2009.07.026
Klein, A., Andersson, J., Ardekani, B. A., Ashburner, J., Avants, B., Chiang, M.-C.,...Parsey, R. V. (2009). Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration. NeuroImage, 46(3), 786–802. doi:10.1016/j.neuroimage.2008.12.037
Kobayashi, Y., & Amaral, D. G. (2003). Macaque monkey retrosplenial cortex: II. Cortical afferents. The Journal of Comparative Neurology, 466(1), 48–79. doi:10.1002/cne.10883
147
Kramer, J. H., Rosen, H. J., Du, A.-T., Schuff, N., Hollnagel, C., Weiner, M. W.,...Delis, D. C. (2005). Dissociations in hippocampal and frontal contributions to episodic memory performance. Neuropsychology, 19(6), 799–805. doi:10.1037/0894-4105.19.6.7999
Laakso, M. P., Partanen, K., Riekkinen, P., Lehtovirta, M., Helkala, E. L., Hallikainen, M.,…Soininen, H. (1996). Hippocampal volumes in Alzheimer’s disease, Parkinson’s disease with and without dementia, and in vascular dementia: An MRI study. Neurology, 46(3), 678–681.
Lamberty, G. J., Kennedy, C. M., & Flashman, L. A. (1995). Clinical utility of the CERAD word list memory test. Applied Neuropsychology, 2(3-4), 170–173. doi:10.1207/s15324826an0203&4_12
Le Bihan, D. (1995). [Diffusion, perfusion and functional magnetic resonance imaging]. Journal Des Maladies Vasculaires, 20(3), 209–214.
Lee, J. E., Park, H.-J., Song, S. K., Sohn, Y. H., Lee, J. D., & Lee, P. H. (2010). Neuroanatomic basis of amnestic MCI differs in patients with and without Parkinson disease. Neurology, 75(22), 2009–2016. doi:10.1212/WNL.0b013e3181ff96bf
Leemans, A., & Jones, D. K. (2009). The B-matrix must be rotated when correcting for subject motion in DTI data. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine, 61(6), 1336–1349. doi:10.1002/mrm.21890
Levin, E. D., & Simon, B. B. (1998). Nicotinic acetylcholine involvement in cognitive function in animals. Psychopharmacology, 138(3-4), 217–230. doi:10.1007/s002130050667
Levy, G., Jacobs, D. M., Tang, M. X., Côté, L. J., Louis, E. D., Alfaro, B.,...Marder, K. (2002). Memory and executive function impairment predict dementia in Parkinson's disease. Movement disorders, 17(6), 1221-1226.
Lewis, M. M., Smith, A. B., Styner, M., Gu, H., Poole, R., Zhu, H.,...Huang, X. (2009). Asymmetrical lateral ventricular enlargement in Parkinson's disease. European Journal of Neurology: The Official Journal of the European Federation of Neurological Societies, 16(4), 475–481. doi:10.1111/j.1468-1331.2008.02430.x
Lezak, M. D., Howieson, D. B., & Loring, D. W. (2004). Neuropsychological assessment. Oxford University Press, USA.
Loewenstein, D. A., Acevedo, A., Potter, E., Schinka, J. A., Raj, A., Greig, M. T.,...Duara, R. (2009). Severity of medial temporal atrophy and amnestic mild cognitive impairment: Selecting type and number of memory tests. The American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 17(12), 1050–1058. doi:10.1097/JGP.0b013e3181b7ef42
148
Markowitsch, H. J., Kalbe, E., Kessler, J., Stockhausen, von, H. M., Ghaemi, M., & Heiss, W. D. (1999). Short-term memory deficit after focal parietal damage. Journal of Clinical and Experimental Neuropsychology, 21(6), 784–797. doi:10.1076/jcen.21.6.784.853
Martin, S. B., Smith, C. D., Collins, H. R., Schmitt, F. A., & Gold, B. T. (2010). Evidence that volume of anterior medial temporal lobe is reduced in seniors destined for mild cognitive impairment. Neurobiology of Aging, 31(7), 1099–1106. doi:10.1016/j.neurobiolaging.2008.08.010
Marui, W., Iseki, E., Kato, M., Akatsu, H., & Kosaka, K. (2004). Pathological entity of dementia with Lewy bodies and its differentiation from Alzheimer’s disease. Acta Neuropathologica, 108(2), 121–128.
Massman, P. J., Delis, D. C., Butters, N., Dupont, R. M., & Gillin, J. C. (1992). The subcortical dysfunction hypothesis of memory deficits in depression: Neuropsychological validation in a subgroup of patients. Journal of Clinical and Experimental Neuropsychology, 14(5), 687–706.
Mattila, P. M., Rinne, J. O., Helenius, H., Dickson, D. W., & Röyttä, M. (2000). Alpha-synuclein-immunoreactive cortical Lewy bodies are associated with cognitive impairment in Parkinson's disease. Acta Neuropathologica, 100(3), 285–290.
Mattila, P. M., Röyttä, M., Lönnberg, P., Marjamäki, P., Helenius, H., & Rinne, J. O. (2001). Choline acetytransferase activity and striatal dopamine receptors in Parkinson's disease in relation to cognitive impairment. Acta Neuropathologica, 102(2), 160–166.
Matison, R., Mayeux, R., Rosen, J., & Fahn, S. (1982). “Tip‐of‐the‐tongue” phenomenon in Parkinson disease. Neurology, 32(5), 567–567.
Mazziotta, J., Toga, A., Evans, A., Fox, P., Lancaster, J., Zilles, K.,...Mazoyer, B. (2001). A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM). Philosophical Transactions of the Royal Society B: Biological Sciences, 356(1412), 1293–1322. doi:10.1098/rstb.2001.0915
Mesulam, M. (2004). The cholinergic innervation of the human cerebral cortex. Progress in Brain Research, 145, 67–78.
Metzler-Baddeley, C., Hunt, S., Jones, D. K., Leemans, A., Aggleton, J. P., & O'Sullivan, M. J. (2012). Temporal association tracts and the breakdown of episodic memory in mild cognitive impairment. Neurology, 79(23), 2233–2240. doi:10.1212/WNL.0b013e31827689e8
Meulenbroek, O., Rijpkema, M., Kessels, R. P. C., Rikkert, M. G. M. O., & Fernández, G. (2010). Autobiographical memory retrieval in patients with Alzheimer's disease. NeuroImage, 53(1), 331–340. doi:10.1016/j.neuroimage.2010.05.082
149
Minoshima, S., Giordani, B., Berent, S., Frey, K. A., Foster, N. L., & Kuhl, D. E. (1997). Metabolic reduction in the posterior cingulate cortex in very early Alzheimer's disease. Annals of Neurology, 42(1), 85–94. doi:10.1002/ana.410420114
Muzzio, I. A., Kentros, C., & Kandel, E. (2009). What is remembered? Role of attention on the encoding and retrieval of hippocampal representations. The Journal of Physiology, 587(12), 2837–2854. doi:10.1113/jphysiol.2009.172445
Naismith, S. L., Shine, J. M., & Lewis, S. J. G. (2010). The specific contributions of set-shifting to freezing of gait in Parkinson's disease. Movement Disorders: Official Journal of the Movement Disorder Society, 25(8), 1000–1004. doi:10.1002/mds.23005
Nakata, Y., Sato, N., Nemoto, K., Abe, O., Shikakura, S., Arima, K.,...Aoki, S. (2009). Diffusion abnormality in the posterior cingulum and hippocampal volume: Correlation with disease progression in Alzheimer's disease. Magnetic Resonance Imaging, 27(3), 347–354. doi:10.1016/j.mri.2008.07.013
Nishio, Y., Hirayama, K., Takeda, A., Hosokai, Y., Ishioka, T., Suzuki, K.,...Mori, E. (2010). Corticolimbic gray matter loss in Parkinson's disease without dementia. European Journal of Neurology: The Official Journal of the European Federation of Neurological Societies, 17(8), 1090–1097. doi:10.1111/j.1468-1331.2010.02980.x
Osborne, J. W. (2010). Improving your data transformations: Applying the Box-Cox transformation. Practical Assessment, Research & Evaluation, 15(12), 1–9.
Pagonabarraga, J., Kulisevsky, J., Llebaria, G., García Sánchez, C., Pascual Sedano,
B., & Gironell, A. (2008). Parkinson’s disease‐cognitive rating scale: A new cognitive scale specific for Parkinson’s disease. Movement Disorders: Official Journal of the Movement Disorder Society, 23(7), 998–1005. doi:10.1002/mds.22007
Papez, J. W. (1937). A proposed mechanism of emotion. 1995 reprint. The Journal of Neuropsychiatry and Clinical Neurosciences, 7(1), 103–112.
Park, A., & Stacy, M. (2009). Non-motor symptoms in Parkinson's disease. Journal of Neurology, 256(S3), 293–298. doi:10.1007/s00415-009-5240-1
Parkinson’s Disease Foundation. (2013). Statistics on Parkinson’s. Retrieved from http://www.pdf.org/en/parkinson_statistics.
Pepeu, G., & Giovannini, M. G. (2010). Cholinesterase inhibitors and memory. Chemico-Biological Interactions, 187(1-3), 403–408. doi:10.1016/j.cbi.2009.11.018
Perry, E. K., Curtis, M., Dick, D. J., Candy, J. M., Atack, J. R., Bloxham, C. A.,...Perry, R. H. (1985). Cholinergic correlates of cognitive impairment in Parkinson’s disease: Comparisons with Alzheimer’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 48(5), 413–421.
150
Pfefferbaum, A., Sullivan, E. V., Adalsteinsson, E., Garrick, T., & Harper, C. (2004). Postmortem MR imaging of formalin-fixed human brain. NeuroImage, 21(4), 1585–1595. doi:10.1016/j.neuroimage.2003.11.024
Portin, R., Laatu, S., Revonsuo, A., & Rinne, U. K. (2000). Impairment of semantic knowledge in Parkinson disease. Archives of Neurology, 57(9), 1338–1343.
Price, B. H., Gürvit, H., Weintraub, S., Geula, C., Leimkuhler, E., & Mesulam, M. (1993). Neuropsychological patterns and language deficits in 20 consecutive cases of autopsy-confirmed Alzheimer's disease. Archives of Neurology, 50(9), 931. doi:10.1001/archneur.1993.00540090038008
Price, C., Garrett, K. D., Jefferson, A., Cosentino, S., Tanner, J., Penney, D.,...Libon, D. (2009). Leukoaraiosis severity and list-learning in dementia. The Clinical Neuropsychologist, 23(6), 944–961. doi:10.1080/13854040802681664
Price, C. C., Favilla, C., Tanner, J. J., Towler, S., Jacobson, C. E., Hass, C. J.,...Okun, M. (2011). Lateral ventricle volume is poor predictor of post unilateral DBS motor change for Parkinson's disease. Parkinsonism & Related Disorders. doi:10.1016/j.parkreldis.2011.01.018
Price, C. C., Tanner, J., Schmalfuss, I., Gearen, P., Dickey, D., Heilman, M.,…Monk, T.G. (in revision). Presurgery neuroanatomical biomarkers for postoperative cognitive decline after total knee arthroplasty in older adults.
Price, C. C., Wood, M. F., Leonard, C. M., Towler, S., Ward, J., Montijo, H.,...Schmalfuss, I. (2010). Entorhinal cortex volume in older adults: Reliability and validity considerations for three published measurement protocols. Journal of the International Neuropsychological Society, 16, 846–855. doi:10.1017/S135561771000072X
Prins, N. D., van Dijk, E. J., den Heijer, T., Vermeer, S. E., Jolles, J., Koudstaal, P. J.,...Breteler, M. B. (2005). Cerebral small-vessel disease and decline in information processing speed, executive function and memory. Brain: A Journal of Neurology, 128(9), 2034–2041.
Rabin, L. A., Paré, N., Saykin, A. J., Brown, M. J., Wishart, H. A., Flashman, L. A., & Santulli, R. B. (2009). Differential memory test sensitivity for diagnosing amnestic mild cognitive impairment and predicting conversion to Alzheimer's disease. Neuropsychology, Development, and Cognition, 16(3), 357–376. doi:10.1080/13825580902825220
Randolph, C., Braun, A. R., Goldberg, T. E., & Chase, T. N. (1993). Semantic fluency in Alzheimer's, Parkinson‘s, and Huntington’s disease: Dissociation of storage and retrieval failures. Neuropsychology, 7(1), 82–88.
151
Reitz, C., Brickman, A. M., Brown, T. R., Manly, J., DeCarli, C., Small, S. A., & Mayeux, R. (2009). Linking hippocampal structure and function to memory performance in an aging population. Archives of Neurology, 66(11), 1385–1392. doi:10.1001/archneurol.2009.214
Rodrigue, K. M., & Raz, N. (2004). Shrinkage of the entorhinal cortex over five years predicts memory performance in healthy adults. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 24(4), 956–963. doi:10.1523/JNEUROSCI.4166-03.2004
Rodríguez-Ferreiro, J., Cuetos, F., Herrera, E., Menéndez, M., & Ribacoba, R. (2010). Cognitive impairment in Parkinson's disease without dementia. Movement Disorders: Official Journal of the Movement Disorder Society, 25(13), 2136–2141. doi:10.1002/mds.23239
Rodríguez-Ferreiro, J., Cuetos, F., Monsalve, A., Martínez, C., Pérez, A. J., & Venneri, A. (2011). Establishing the relationship between cortical atrophy and semantic deficits in Alzheimer’s disease and mild cognitive impairment patients through voxel-based morphometry. Journal of Neurolinguistics, 25, 139–149.
Roediger, H. L. (1990). Implicit memory: Retention without remembering. American Psychologist, 45, 1043-1056.
Rogers, D., Lees, A. J., Smith, E., Trimble, M., & Stern, G. M. (1987). Bradyphrenia in Parkinson's disease and psychomotor retardation in depressive illness. An experimental study. Brain: A Journal of Neurology, 110(3), 761–776.
Rohde, G. K., Barnett, A. S., Basser, P. J., Marenco, S., & Pierpaoli, C. (2003). Comprehensive approach for correction of motion and distortion in diffusion-weighted MRI. Magnetic Resonance in Medicine, 51(1), 103–114. doi:10.1002/mrm.10677
Rosas, H. D., Lee, S. Y., Bender, A. C., Zaleta, A. K., Vangel, M., Yu, P.,...Hersch, S. M. (2010). Altered white matter microstructure in the corpus callosum in Huntington's disease: Implications for cortical ‘disconnection’. NeuroImage, 49(4), 2995–3004. doi:10.1016/j.neuroimage.2009.10.015
Saint-Cyr, J. A. (2003). Frontal-striatal circuit functions: Context, sequence, and consequence. Journal of the International Neuropsychological Society, 9(1), 103–127.
Salat, D. H., Tuch, D. S., van der Kouwe, A. J. W., Greve, D. N., Pappu, V., Lee, S. Y.,...Rosas, H. D. (2010). White matter pathology isolates the hippocampal formation in Alzheimer's disease. Neurobiology of Aging, 31(2), 244–256. doi:10.1016/j.neurobiolaging.2008.03.013
152
Salmon, E., & Laureys, S. (2009). Brain imaging in prodromal and probable Alzheimer’s Disease. A focus on the cingulate gyrus. In B. A. Vogt (Ed.), Cingulate Neurobiology and Disease (pp. 749-762). New York, NY: Oxford University Press.
Sara, S. J. (2000). Retrieval and reconsolidation: Toward a neurobiology of remembering. Learning & Memory, 7(2), 73–84.
Sawamoto, N., Honda, M., Hanakawa, T., Aso, T., Inoue, M., Toyoda, H.,...Shibasaki, H. (2007). Cognitive slowing in Parkinson disease is accompanied by hypofunctioning of the striatum. Neurology, 68(13), 1062–1068. doi:10.1212/01.wnl.0000257821.28992.db
Schwarcz, R., & Witter, M. P. (2002). Memory impairment in temporal lobe epilepsy: The role of entorhinal lesions. Epilepsy Research, 50(1-2), 161–177.
Segonne, F., Dale, A. M., Busa, E., Glessner, M., Salat, D., Hahn, H. K., & Fischl, B. (2004). A hybrid approach to the skull stripping problem in MRI. NeuroImage, 22(3), 1060–1075.
Selden, N., Gitelman, D., Salamon-Murayama, N., Parrish, T., & Mesulam, M. (1998). Trajectories of cholinergic pathways within the cerebral hemispheres of the human brain. Brain: A Journal of Neurology, 121(12), 2249. doi:10.1093/brain/121.12.2249
Shallice, T., & Warrington, E. K. (1970). Independent functioning of verbal memory stores: A neuropsychological study. The Quarterly Journal of Experimental Psychology, 22(2), 261–273.
Smith, C. N., & Squire, L. R. (2009). Medial temporal lobe activity during retrieval of semantic memory is related to the age of the memory. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 29(4), 930–938. doi:10.1523/JNEUROSCI.4545-08.2009
Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E., Johansen-Berg, H.,…Filtney, D. E. (2004). Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage, 23, S208.
Squire, L. R., Stark, C. E. L., & Clark, R. E. (2004). The medial temporal lobe. Annual Review of Neuroscience, 27, 279–306. doi:10.1146/annurev.neuro.27.070203.144130
Sörös, P., Bose, A., Sokoloff, L. G., Graham, S. J., & Stuss, D. T. (2009). Age-related changes in the functional neuroanatomy of overt speech production. Neurobiology of Aging. doi:10.1016/j.neurobiolaging.2009.08.015
Stadlbauer, A., Salomonowitz, E., Strunk, G., Hammen, T., & Ganslandt, O. (2008). Quantitative diffusion tensor fiber tracking of age-related changes in the limbic system. European Radiology, 18(1), 130–137. doi:10.1007/s00330-007-0733-8
153
Stebbins, G. T., Carrillo, M. C., Dorfman, J., Dirksen, C., Desmond, J. E., Turner, D. A.,...Gabrieli, D. E. (2002). Aging effects on memory encoding in the frontal lobes. Psychology and Aging, 17(1), 44–55.
Stepkina, D. A., Zakharov, V. V., & Yakhno, N. N. (2010). Cognitive impairments in progression of Parkinson's disease. Neuroscience and Behavioral Physiology, 40(1), 61–67. doi:10.1007/s11055-009-9223-6
Stone, M., Gabrieli, J. D., Stebbins, G. T., & Sullivan, E. V. (1998). Working and strategic memory deficits in schizophrenia. Neuropsychology, 12(2), 278–288.
Stoub, T. R., Rogalski, E. J., Leurgans, S., Bennett, D. A., & deToledo-Morrell, L. (2010). Rate of entorhinal and hippocampal atrophy in incipient and mild AD: Relation to memory function. Neurobiology of Aging, 31(7), 1089–1098. doi:10.1016/j.neurobiolaging.2008.08.003
Stuss, D. T., & Alexander, M. P. (2007). Is there a dysexecutive syndrome? Philosophical Transactions of the Royal Society of London Series B, Biological Sciences, 362(1481), 901–915. doi:10.1098/rstb.2007.2096
Tanner, J. J., Price, C. C., Towler, S., Mitchell, S., Collazo, J., Moran, S.,...Libon, D. J. (2011). Dissociating gray and white matter contributions to verbal list learning and memory. Journal of the International Neuropsychological Society, 17(S1), i–334. doi:10.1017/S1355617711000415
Tinaz, S., Schendan, H. E., & Stern, C. E. (2008). Fronto-striatal deficit in Parkinson's disease during semantic event sequencing. Neurobiology of Aging, 29(3), 397–407. doi:10.1016/j.neurobiolaging.2006.10.025
Trinh, N. H., Hoblyn, J., Mohanty, S., & Yaffe, K. (2003). Efficacy of cholinesterase inhibitors in the treatment of neuropsychiatric symptoms and functional impairment in Alzheimer disease. JAMA: The Journal of the American Medical Association, 289(2), 210–216.
Tröster, A. I., Stalp, L. D., Paolo, A. M., Fields, J. A., & Koller, W. C. (1995). Neuropsychological impairment in Parkinson's disease with and without depression. Archives of Neurology, 52(12), 1164.
Tupler, L. A., Krishnan, K. R. R., Greenberg, D. L., Marcovina, S. M., Payne, M. E., MacFall, J. R.,...Doraiswamy, P. M. (2007). Predicting memory decline in normal elderly: Genetics, MRI, and cognitive reserve. Neurobiology of Aging, 28(11), 1644–1656. doi:10.1016/j.neurobiolaging.2006.07.001
Turken, U., & Dronkers, N. F. (2011). The neural architecture of the language comprehension network: Converging evidence from lesion and connectivity analyses. Frontiers in Systems Neuroscience, 5. doi:10.3389/fnsys.2011.00001
154
Uc, E. Y., Rizzo, M., Anderson, S. W., Qian, S., Rodnitzky, R. L., & Dawson, J. D. (2005). Visual dysfunction in Parkinson disease without dementia. Neurology, 65(12), 1907–1913.
Vakil, E., & Blachstein, H. (1993). Rey Auditory-Verbal Learning Test: Structure analysis. Journal of Clinical Psychology, 49(6), 883–890.
Valenstein, E., Bowers, D., Verfaellie, M., Heilman, K. M., Day, A., & Watson, R. T. (1987). Retrosplenial amnesia. Brain: A Journal of Neurology, 110(6), 1631–1646.
Van Hoesen, G. W., & Hyman, B. T. (1990). Hippocampal formation: Anatomy and the patterns of pathology in Alzheimer's disease. Progress in Brain Research, 83, 445–457.
Vann, S. D., Aggleton, J. P., & Maguire, E. A. (2009). What does the retrosplenial cortex do? Nature Reviews Neuroscience, 10(11), 792–802. doi:10.1038/nrn2733
Villain, N., Desgranges, B., Viader, F., La Sayette, de, V., Mézenge, F., Landeau, B.,...Chételat, G. (2008). Relationships between hippocampal atrophy, white matter disruption, and gray matter hypometabolism in Alzheimer's disease. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 28(24), 6174–6181. doi:10.1523/JNEUROSCI.1392-08.2008
Vogt, B. A., Pandya, D. N., & Rosene, D. L. (1987). Cingulate cortex of the rhesus monkey: I. Cytoarchitecture and thalamic afferents. The Journal of Comparative Neurology, 262(2), 256–270. doi:10.1002/cne.902620207
Vogt, B. A., Vogt, L. J., Purohit, D. P., & Hof, P. R. (2009) Cingulate subregional neuropathology in Dementia with Lewy Bodies and Parkinson’s Disease with Dementia. In B. A. Vogt (Ed.), Cingulate Neurobiology and Disease (pp. 707-725). New York, NY: Oxford University Press.
Vos, S. B., Jones, D. K., Viergever, M. A., & Leemans, A. (2011). Partial volume effect as a hidden covariate in DTI analyses. NeuroImage, 55(4), 1566–1576. doi:10.1016/j.neuroimage.2011.01.048
Wang, R., Benner, T., Sorensen, A. G., & Wedeen, V. J. (2007). Diffusion toolkit: A software package for diffusion imaging data processing and tractography. Presented at the Proceedings of the International Society of Magnetic Resonance in Medicine, 15, 3720.
Wechsler, D. (1997). Wechsler Memory Scale–Third Edition. San Antonio, TX: The Psychological Corporation.
Weintraub, D., Moberg, P. J., Culbertson, W. C., Duda, J. E., & Stern, M. B. (2004). Evidence for impaired encoding and retrieval memory profiles in Parkinson disease. Cognitive and Behavioral Neurology: Official Journal of the Society for Behavioral and Cognitive Neurology, 17(4), 195–200.
155
Welsh, K. A., Butters, N., Mohs, R. C., Beekly, D., Edland, S., Fillenbaum, G., & Heyman, A. (1994). The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part V. A normative study of the neuropsychological battery. Neurology, 44(4), 609–614.
Wernicke C. (1874). Der aphasische symptomencomplex. Breslau: Cohen and Weigert.
Whitehouse, P. J., Martino, A. M., Marcus, K. A., Zweig, R. M., Singer, H. S., Price, D. L., & Kellar, K. J. (1988). Reductions in acetylcholine and nicotine binding in several degenerative diseases. Archives of Neurology, 45(7), 722–724.
Wimo, A., & Prince, M. (2010). World Alzheimer's Report 2010. World Alzheimer’s Report. Retrieved December 31, 2012, from http://www.alz.co.uk/research/files/WorldAlzheimerReport2010.pdf
Wirdefeldt, K., Gatz, M., Reynolds, C. A., Prescott, C. A., & Pedersen, N. L. (2011). Heritability of Parkinson disease in Swedish twins: A longitudinal study. Neurobiology of Aging. doi:10.1016/j.neurobiolaging.2011.02.017
Wolk, D. A., Dickerson, B. C., & Alzheimer's Disease Neuroimaging Initiative. (2011). Fractionating verbal episodic memory in Alzheimer's disease. NeuroImage, 54(2), 1530–1539. doi:10.1016/j.neuroimage.2010.09.005
Yakovlev, P., & Locke, S. (1961). Limbic nuclei of thalamus and connections of limbic cortex: III. Corticocortical connections of the anterior cingulate gyrus, the cingulum, and the subcallosal bundle in monkey. Archives of Neurology.
Yan, C., Gong, G., Wang, J., Wang, D., Liu, D., Zhu, C., Chen, Z. J.,…He, Y. (2011). Sex- and brain size-related small-world structural cortical networks in young adults: A DTI tractography study. Cerebral Cortex, 21(2), 449–458. doi:10.1093/cercor/bhq111
Yasmin, H., Aoki, S., Abe, O., Nakata, Y., Hayashi, N., Masutani, Y.,...Ohtomo, K. (2009). Tract-specific analysis of white matter pathways in healthy subjects: A pilot study using diffusion tensor MRI. Neuroradiology, 51(12), 831–840. doi:10.1007/s00234-009-0580-1
Yassa, M. A., Muftuler, L. T., & Stark, C. E. L. (2010). Ultrahigh-resolution microstructural diffusion tensor imaging reveals perforant path degradation in aged humans in vivo. Proceedings of the National Academy of Sciences of the United States of America, 107(28), 12687–12691. doi:10.1073/pnas.1002113107
Yoshikawa, K., Nakata, Y., Yamada, K., & Nakagawa, M. (2004). Early pathological changes in the parkinsonian brain demonstrated by diffusion tensor MRI. Journal of Neurology, Neurosurgery, and Psychiatry, 75(3), 481–484.
156
Zahodne, L. B., Bowers, D., Price, C. C., Bauer, R. M., Nisenzon, A., Foote, K. D., & Okun, M. S. (2011). The case for testing memory with both stories and word lists prior to DBS surgery for Parkinson's disease. The Clinical Neuropsychologist, 25(3), 348–358. doi:10.1080/13854046.2011.562869
Zola-Morgan, S., & Squire, L. R. (1993). Neuroanatomy of memory. Annual Review of Neuroscience, 16, 547–563. doi:10.1146/annurev.ne.16.030193.002555
Zurif, E. B. (1980). Language mechanisms: A neuropsychological perspective. American Scientist, 68(3), 305–311.
157
BIOGRAPHICAL SKETCH
Jared J. Tanner grew up in Arizona in a family of seven children. He attended
Brigham Young University where he earned a Bachelor of Science in psychology, with a
minor in gerontology. He also earned a Master of Science degree from Brigham Young
University in psychology. His thesis was entitled Measuring Reliable Change in Acute
Respiratory Distress Syndrome.
Currently, Jared is a graduate student at the University of Florida in the Clinical
and Health Psychology Ph.D. program with an emphasis in neuropsychology. Jared is a
member of Catherine Price’s laboratory with a focus on neuroimaging and cognition and
memory in neurological disorders and after major surgery. His clinical internship was
completed at the Duke University Medical Center in the Department of Psychiatry and
Behavioral Sciences.