<%BANNER%>

Functional Connectivity and Dedifferentiation of Function in Older Adults During Category Member Generation

Permanent Link: http://ufdc.ufl.edu/UFE0024465/00001

Material Information

Title: Functional Connectivity and Dedifferentiation of Function in Older Adults During Category Member Generation
Physical Description: 1 online resource (40 p.)
Language: english
Creator: Cohen, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: aging, dedifferentiation, fmri, harold, neuroimaging, neuropsychology
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Older adults demonstrate a dedifferentiation of function whereby their cognitive abilities become more strongly related over time. Since the time of Spearman, this phenomenon has been shown with large samples and multivariate analyses. However, it is not clear what causes this phenomenon at the neural level. Functional neuroimaging has shown that older adults show greater activity and in more areas than young adults during the same cognitive task (termed the HAROLD phenomenon). However, this change in activity does not mean a dedifferentiation of function per se. The present investigation is a secondary analysis of a project designed to study the HAROLD phenomenon (Hemispheric Asymmetry Reduction in OLD adults) during category member generation. Here, we used functional connectivity (FC) to measure the level of synchrony and coherence between active regions to investigate whether and how HAROLD occurrences reflect of dedifferentiation of function. Twenty-two young (16-35 years old), eleven young-old (65-68), and eleven old-old (69-84) adults completed a block fMRI paradigm silently generating category exemplars (17 seconds/category; 16 categories). FMRI data were analyzed using AFNI. Results show that young-old adults demonstrate HAROLD-consistent activation during the task but that active regions are not strongly related functionally. Old-old adults, on the other hand, demonstrated widespread functional connections among (almost) all active regions. This suggests that in late life, HAROLD-typical activation is accompanied by functional dedifferentiation but that this stage is preceded by a stage where HAROLD activation occurs in the absence of functional integration of the active areas. We discuss these findings with three possible explanations. One describes the findings in the context of the brain recruiting more neural resources to account for age-related declines. The second possibility discusses the findings as a result of age-specific loss of dopamine receptors. The third possibility discusses the findings as the inability of the thalamus to selectively activate cortical networks needed for the task.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Cohen.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Crosson, Bruce A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024465:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024465/00001

Material Information

Title: Functional Connectivity and Dedifferentiation of Function in Older Adults During Category Member Generation
Physical Description: 1 online resource (40 p.)
Language: english
Creator: Cohen, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: aging, dedifferentiation, fmri, harold, neuroimaging, neuropsychology
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Older adults demonstrate a dedifferentiation of function whereby their cognitive abilities become more strongly related over time. Since the time of Spearman, this phenomenon has been shown with large samples and multivariate analyses. However, it is not clear what causes this phenomenon at the neural level. Functional neuroimaging has shown that older adults show greater activity and in more areas than young adults during the same cognitive task (termed the HAROLD phenomenon). However, this change in activity does not mean a dedifferentiation of function per se. The present investigation is a secondary analysis of a project designed to study the HAROLD phenomenon (Hemispheric Asymmetry Reduction in OLD adults) during category member generation. Here, we used functional connectivity (FC) to measure the level of synchrony and coherence between active regions to investigate whether and how HAROLD occurrences reflect of dedifferentiation of function. Twenty-two young (16-35 years old), eleven young-old (65-68), and eleven old-old (69-84) adults completed a block fMRI paradigm silently generating category exemplars (17 seconds/category; 16 categories). FMRI data were analyzed using AFNI. Results show that young-old adults demonstrate HAROLD-consistent activation during the task but that active regions are not strongly related functionally. Old-old adults, on the other hand, demonstrated widespread functional connections among (almost) all active regions. This suggests that in late life, HAROLD-typical activation is accompanied by functional dedifferentiation but that this stage is preceded by a stage where HAROLD activation occurs in the absence of functional integration of the active areas. We discuss these findings with three possible explanations. One describes the findings in the context of the brain recruiting more neural resources to account for age-related declines. The second possibility discusses the findings as a result of age-specific loss of dopamine receptors. The third possibility discusses the findings as the inability of the thalamus to selectively activate cortical networks needed for the task.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Cohen.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Crosson, Bruce A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024465:00001


This item has the following downloads:


Full Text

PAGE 1

FUNC TIONAL CONNECTIVITY AND DEDIFFER ENTIATION OF FUNCTION IN OLDER ADULTS DURING CATEGORY MEMBER GENERATION By MATTHEW L. COHEN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Matthew L. Cohen 2

PAGE 3

To m y family for their love and support 3

PAGE 4

ACKNOWL EDGMENTS I would first like to thank Dr. Bruce Cross on for his exceptional mentorship, guidance, training, and enthusiasm. I am also thankful for the time and comments of my committee members: Dr. William Perlstein, Dr. Deidre Pereira, and Dr. Sheila Eyberg. I would also like to thank my friends and colleagues for their support and assistance: Jonathan, Zvinka, Ilana, Keith, and Michelle. Finally, I extend my greatest thanks to my parents for their endless love and support and to my sister, Jessica who inspires me every day. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........6 LIST OF FIGURES.........................................................................................................................7 ABSTRACT.....................................................................................................................................8 1 INTRODUCTION................................................................................................................. .10 2 METHODS...................................................................................................................... .......16 Participants.............................................................................................................................16 Dementia Screen................................................................................................................ .....17 Assessment of Language........................................................................................................1 7 Functional Magnetic Resonance Imaging (f MRI) Category Member Generation Task........17 Image Acquisition.............................................................................................................. .....18 Data Analysis..........................................................................................................................18 Functional Connectivity Analysis...........................................................................................19 3 RESULTS...................................................................................................................... .........20 Behavioral Results............................................................................................................. .....20 Functional Magnetic Resonan ce Imaging (fMRI) Results.....................................................21 Within-Subject fMRI Results..........................................................................................21 Between-Subjects fMRI Results......................................................................................22 Functional Connectivity Results.............................................................................................22 Young Participants..........................................................................................................22 Young-Old Participants...................................................................................................23 Old-Old Participants........................................................................................................23 4 DISCUSSION................................................................................................................... ......28 Compensation.........................................................................................................................29 Age-Specific Deterioration of Basal Ganglia Circuitry..........................................................30 5 CONCLUSION................................................................................................................... ....35 LIST OF REFERENCES...............................................................................................................36 BIOGRAPHICAL SKETCH.........................................................................................................40 5

PAGE 6

LIST OF TABLES Table page 3-1 Within-subject activity against baseline............................................................................24 3-2 Between-subject activity................................................................................................... .25 6

PAGE 7

LIST OF FI GURES Figure page 3-1 Young adults against baseline............................................................................................26 3-2 Young-old adults against baseline.....................................................................................26 3-3 Old-old adults against baseline..........................................................................................27 4-1 Direct, indirect, and hyperdirect loops of the basal ganglia..............................................34 7

PAGE 8

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science FUNCTIONAL CONNECTIVITY AND DEDIFFER ENTIATION OF FUNCTION IN OLDER ADULTS DURING CATEGORY MEMBER GENERATION By Matthew L. Cohen May 2009 Chair: Bruce Crosson Major: Psychology Older adults demonstrate a dedifferentiation of function whereby their cognitive abilities become more strongly related ov er time. Since the time of Spearman this phenomenon has been shown with large samples and multivariate analyses. However, it is not clear what causes this phenomenon at the neural level. Functional neuroimaging has shown that older adults show greater activity and in more areas than young adu lts during the same cognitive task (termed the HAROLD phenomenon). However, this change in activity does not mean a dedifferentiation of function per se. The present investiga tion is a secondary analysis of a project designed to study the HAROLD phenomenon (Hemispheric Asymmetry Reduction in OLD adults) during category member generation. Here, we used functional connectivity (FC) to measure the level of synchrony and coherence between active region s to investigate whet her and how HAROLD occurrences reflect of de differentiation of function. Twenty-two young (16-35 years old), eleven young-old (65-68), and eleven old-old (6984) adults completed a block fMRI paradigm silently generating category exemplars (17 seconds/category; 16 categories). FMRI data were analyzed using AFNI. Results show that young-old adults demonstr ate HAROLD-consistent activation during the task but that active regions ar e not strongly related functionally. Old-old adults, on the other 8

PAGE 9

9 hand, demonstrated widespread functional connecti ons among (almost) all active regions. This suggests that in late life, HAROLD-typi cal activation is accompanied by functional dedifferentiation but that this stage is preceded by a stage where HAROLD activation occurs in the absence of functional integr ation of the active areas. We di scuss these findings with three possible explanations. One descri bes the findings in the context of the brain recruiting more neural resources to account for age-related d eclines. The second possibility discusses the findings as a result of age-specific loss of dopami ne receptors. The third possibility discusses the findings as the inability of the thalamus to sele ctively activate cortical networks needed for the task.

PAGE 10

CHAP TER 1 INTRODUCTION The idea that we can think of human cognition as both one core construct as well as many distinct aptitudes comes from Spearman (Spearman, 1904) who proposed a general factor of intelligence, g, as the degree of positive manifold (i.e. patterns of positive correlations) between tests of different cognitive abil ities (S. C Li & Lindenberger, 1 999) and posited that the factor structure of cognitive abilities va ries with age and ability level (Spearman, 1927). Since the time of Spearman it has been shown that abilities do indeed differentiate (i.e. become less related) in adolescence (Garrett, 1946) and then dedifferentiate in old age (Balinksy, 1941). Dedifferentiation remains one of th e most studied and most disput ed phenomena in the field of cognitive aging. There are two main approaches to studyi ng dedifferentiation. The first, termed the macroapproach (T. A. Salthouse, 2000), uses large samples, multivariate designs, and correlational analyses to invest igate patterns of relationships among variables. These methods have supported dedifferentiation by showing (a) age -related increases in th e correlations between levels of performance on cognitive abilities (P. B. Baltes & Lindenberger, 1997; de Frias, 2007; Ghisletta & Lindenberger, 2003; Hayslip & St erns, 1979), (b) age-related increases in correlations among cognitive changes, (c) higher interindividual variability in change in the older age groups (de Frias, 2007), and a weaker fact or structure among meas ures (P. B. Baltes, Cornelius, S.W., Spiro, A., Nesselroade, J.R., W illis, S.L., 1980). However, other studies have failed to show these patterns (Anstey et al., 2003; Sims, 2008; Tucker-Drob & Salthouse, 2008; Zelinski & Lewis, 2003). Possible e xplanations for the inconsistenc ies in the lite rature include differences in (a) sample age ranges, (b) the cogni tive measures included in the analyses, and (c) the types of statistical correc tions used (T. A. C. Salthouse, F.I.M., 2000; Sims, 2008). While 10

PAGE 11

m acroapproaches have the advant age of greater statistical powe r and multivariate techniques, they are less able to identify sp ecific mechanisms and processes and require greater inference in order to be translated into applications for neuropsychology and c ognitive neuroscience. Microapproaches, on the other hand, tend to focus on the age-re lated effects on a few, more focused variables and use smaller, more extr eme groups (e.g. young and old) and are better able to support specific mechanisms. This method has been less utilized in the study of dedifferentiation and was the approach taken by the present investigation. The general interpretation of de differentiation in old age is that the increase in correlations between abilities reflects declines in basic cognitive structures (Z elinski & Lewis, 2003), a pattern that is likely to emerge if the efficiency of an underlying structure or process is low (Deary, 1996; T. A. Salthouse, 2000). Though the exact nature of this factor or process is not known, several theories have been offered. Sa lthouse and colleagues (Salthouse et al., 1996) proposed that age-related declines are due to a general slowin g of processing speed, perhaps as a result of slower speeds of action potentials along single or multiple pathways, faulty neurotransmitters, or reduced synchronization of activation patterns. Lis neurocomputation model (S. C. Li et al., 2006; S. C Li & Lindenbe rger, 1999; S. C. Li et al., 2001) proposed that deficient neuromodulation of catecholamines (dopa mine, epinephrine, norepinephrine), known to confine neural activity (Oades, 1985; Sa waguchi & Goldman-Rakic, 1991), causes undifferentiated cortical representations and causes an in crease in random variability. These potential mediating f actors have been explored independently of the dedifferentiation phenomenon and have only been in ferred to be mediating variables. In fact, there is a noticeable lack of micro studies that investigate dedifferentiation mechanisms, probably because dedifferentiation is traditiona lly conceptualized as a multivariate phenomenon 11

PAGE 12

and dif ficult to conceptualize by studying a small number of cognitive variables. However, one possibility is that dedifferentiation is related to a decrease in functional specialization in old age and that a given brain region might be able to contribute to se veral different cognitive processes or the same cognitive process could be controlled by multiple neuroanatomical regions (Kinsbourne, 1980; T. A. Salthouse, 2000). In fact the latter occurrence has been by shown by functional neuroimaging (positron emission tomography, PET; functional magnetic resonance imaging, fMRI) for over a decade. Indeed, one of the most robust findings since the advent of functional neuroimaging is that, compared to younger adults who tend to compartmentalize functional activity during a given cognitive task, older adults tend to exhibit more diffuse activity and engage more structures bilaterally (both hemispheres). Th is consistent finding has been termed the HAROLD phenomenon (hemispheric asym metry reduction in older adults) (Cabeza, 2002) and has been shown across multiple cognitive domains. However, it is not clear whether the HAROLD phenomenon implies age-related dedifferentiation per se (Park et al., 2004). That is it is not known whether the additional activity reflects the recruitment of more neural resources that are f unctionally specialized for the task being performed (which does not reflect dedifferen tiation of function), or whether it reflects the incorporation of areas that are specialized in young adu lts but perform more general functions in older subjects (consistent with functional de differentiation) (Park et al., 2004; Rajah & D'Esposito, 2005). Yet another possibility is that the bilateral ac tivation is epiphenomenal and does not contribute to dedicated neural mechan isms. Finally, it is possible that the additional activity is unwanted and detracts from performance, as was suggested by Crosson et al. (B. Crosson et al., 2003), for example, during language production. 12

PAGE 13

Dedifferentiation at the neural level has been def ined by some as a change in the spatial pattern of brain activa tion with age as a result of decrea sed neural specificity (Cabeza, 2002; Park et al., 2004; Voss et al., 2008; Zarahn et al., 2007) but only a few studies have explored this occurrence and the results have not been discussed as relating to the (multivariate) dedifferentiation literature. Studi es that have operationally defi ned neural dedifferentiation in this way have quantified it by ca lculating the degree of spatia l overlap for a given region between cognitive tasks. However, spatial overlap alone may not fully capture the nature of the phenomenon and does not fully bridge the gap between HAROLD findings and the multivariate dedifferentiation literature. What is also needed is a measure of functional overlap to show that spatial differences seen during fMRI are related to functional integration of these regions, as well. One way of looking at functional relationships is with participants blood oxygenation level-dependent (BOLD) (Ogawa et al., 1990) responses. To our knowledge, only one study (Voss et al., 2008) has used a quantitative measure of dediffere ntiation that included BOLD response. Looking at the neural speci ficity in the visual cortex for processing color, words, faces, and places, these authors calculated a discrimini bility metric that was the average activation (percent signal change) to the preferred stimulus -localized region of inte rest (ROI), divided by the average standard deviation of the respons e amplitude for preferred and all non-preferred stimulus conditions within the preferred stimulus -localized ROI. We agree with these authors that a quantitative measure of dedifferentiati on that takes into account both magnitude and variance of activation would he lp examine not only age-relate d differences, but also could potentially be used as a tool to study individual differences within an aging cohort. A metric that includes BOLD response is better able to look at functional overlap (a s opposed to spatial overlap), is less sensitive to thresholding artif acts that can arise when comparing old and young 13

PAGE 14

subjects, an d helps account for the greater variab ility seen in older adults' response amplitude across time (Voss et al., 2008). The present analysis is based on data collected from another study designed to examine the HAROLD phenomenon during category member generation (Benjamin, Cohen et al., in preparation). Here, we draw from another methodology, functional connectivity, to provide another BOLD-derived index of functional dedi fferentiation. Functional connectivity is not a crisp methodological term but rather a class of analyses defined independently by different researchers and with different imaging modalities (fMRI, EEG, PET). For the present analysis, we adopt the definition provided by Rogers et al. (Rogers et al., 2007) that functional connectivity means the quantification of the operational interactions of multiple spatially distinct brain regions that are engaged simultaneously in a task. These authors have also noted that, while specific analytic techniques var y, a common feature of mu ltiple fMRI assessments of connectivity is the use of corre lations or covariances of activities derived from BOLD data. Correlated BOLD signal shows coherence and s ynchrony of activity between regions and has been interpreted as a marker of connectivity. Un like effective connectivity, a similar analysis, functional connectivity makes no attempt to describe or impose causality on separate regions or voxels. We believe this to be an appropriate te chnique to study dediffere ntiation since a loss of functional specificity does not ha ve an obvious directional component and it is not clear how to impose a priori hypotheses on the data. Previous BOLD fMRI studies of functional connectivity have looked at correlations of time-series data (Tivarus et al., 2008) and deconvolved impulse response functions. The present analysis inve stigates functional connectivity using the deconvolved hemodynamic response across blocked trials during category member generation. 14

PAGE 15

In summ ary, the present i nvestigation used a BOLD-derived measure of functional connectivity to investigate whether older ad ults diffuse, HAROLD-t ypical activity during category member generation is consistent with pr evious conceptualizatio ns of dedifferentiation of function in old age. Multivariate (macro) analyses have shown that in old age, cognitive abilities become more highly rela ted. We hypothesize that the pres ent investigation will reflect this phenomenon at the neural leve l during a single cognitive task ( micro level) as an increase in functional connectivity in old age. 15

PAGE 16

CHAP TER 2 METHODS Participants Twenty-two older adults (a ges 65-85, mean=71.3, SD=6.2) and twenty-two young adults (ages 18-35, mean=25.5, SD=4.5) participated and were matched for education level (young mean=16.4, young SD=2.3, older mean=16.3, older SD =3.2). For some analyses the group of older adults was subdivided by a median spl it for age to produce a Young-Old group (N=11, Mean=66.4, SD=0.92) and an Old-Old group (N =11, Mean=76.2, SD=5.3). In the Young group there were 10 men and 12 women. In the YoungOld group there were 4 men and 7 women. In the Old-Old group there were 9 men and 2 women. Young adults were recruited from the Univ ersity of Florida and Santa Fe College (Gainesville, FL) communities and older adults we re recruited from the community at large. All participants were native English speakers a nd right-hand dominant, as determined by the Edinburgh Handedness Inventory (Oldfi eld, 1971). Potential participan ts were excluded if they reported a history of neurological disease, dementia or mild cognitive impairment, cardiovascular disease, uncontro lled hypertension, Diagnostic and Statistical Manual of Mental Disorders American Psychiatric Association [DSM-IV-TR], 2000) Axis I diagnosis including learning disability, attention defic it hyperactivity disorder, and subs tance abuse, or if they were taking prescription psychoactive medications. Partic ipants were also extensively screened to make sure they were free of metal in their body (other than dental fil lings). Potential female participants were excluded if th ey were pregnant or trying to b ecome pregnant. Participants were instructed to abstain from caffein e on the day of the scan. All part icipants gave written informed consent in accordance with a protocol approved by the Health Center Institutional Review Board at the University of Florida and were paid US $25 as reimbursement for travel expenses. 16

PAGE 17

Dementia S creen Prior to being enrolled in the functional magnetic resonance imaging (fMRI) study, older participants were administered a standard health history qu estionnaire and completed brief cognitive tests to screen for dementia or Mild Cognitive Impairment-amnestic type, a precursor to pathological aging and a high-risk condition for the development of clinically probable Alzheimers disease (Petersen et al., 2001). Individuals were excluded had they reported subjective memory complaints, scored below 27 out of 30 on the Mini-Mental State Exam (MMSE) (Folstein et al., 1975)or obtained a total recall score (t rials 1) of less than 15 on the Hopkins Verbal Learning Test (HVLT) (Shapiro et al., 1999), a te st of auditory-verbal working memory. No subjects were excl uded for any of these reasons. Assessment of Language Outside of the scanner, all participants completed the category (animal) fluency subtest from the Delis-Kaplan Executive Functioning Sy stem (D-KEFS) (Delis, 2001) and responses were clustered into four 15-second intervals. Ol der participants were additionally administered the Boston Naming Test (Kaplan, 1983) to assess confrontation naming. Functional Magnetic Resonance Imaging (fMRI) Category Member Generation Task During fMRI, participants covertly (i.e. silen tly) generated members of the given category for 17 seconds (10 TR). Covert generation was de termined to be the best methodology due to the many difficulties associated with overt naming during fMRI tria ls (B. Crosson et al., 2007), including movement and respiratory artifact, an d because covert generation was known to elicit cortical as well as cortical activity (Crosson et al. 2003). Category members were presented visually and aurally at the same time with a stop cue at the end of the 17 seconds. Sixteen total categories (order counterbalanced) were presen ted across 2 fMRI echo-planar imaging (EPI) runs in a blocked design format with 10.2, 11.9, and 15.3 seconds intervals between active 17

PAGE 18

blocks, during which participants fixated on a cr oss, which served as a baseline of activity. Exam ples of categories includes: birds, bevera ges, lunch meats, occupations, ocean animals. Image Acquisition Images were acquired on a Siemens Allegra 3.0 Tesla head-only scanner with the standard quadrature radio frequency head coil. Functional images were obt ained with a 1-shot gradient echo planar imaging (EPI) scan: 240mm field of view (FOV), 64x64 matrix (3.75 x 3.75 in-plane resolution), repetition time (TR)=1700ms, echo time (TE)=25ms, flip angle (FA)=70. Thirtytwo 5mm axial slices covering th e whole brain were acquired. A high-resolution T1-weighed 3D rapid acquisition gradient echo (MP-RAGE )scan (TE=4.13ms, TR=2000ms; FOV=240mm; FA=8; matrix size=256x192mm l 128 1.3mm slices) was obtained to provide anatomic reference. Head motion was minimized using foam padding. Data Analysis fMRI data were analyzed and overlaid on struct ural images with the Analysis of Functional Neuroimaging (AFNI) software package from the National Institutes of Health (Cox, 1996). The first 6 images of the functional runs were discarded to ensure stability of the imaging matrix. Time series images were spatially registered in three-dimensional space to minimize the effects of head motion. The two imaging runs were c oncatenated into a single time series. Two participants were omitted from analysis, one for benign cortical calcifications and a second for use of a prescription calc ium ion blocker, leaving the 22 subjec ts per group (described above). For each voxel of each subject, the obtained fMRI intensity time-series was modeled with AFNIs 3dDeconvolution program as the convolution of the experimental stimulus vector and the estimated best-fit twelve-lag impulse re sponse, allowing the hemodynamic response (HDR) to return to baseline. Subject values were scaled to reflect Z scores. The mean value of the deconvolved impulse response function (mean-IRF) wa s the dependent variable for analyses and 18

PAGE 19

was calcu lated by adding the deconvolved image intensity at each deconvolved time point of the impulse response. The T1-weighted anatomic images and the mean-IRF functional activation maps were warped to the coordinates of the co-p lanar stereotactic atlas of Talairach and Tornoux (Talairach, 1988) and resampled at a 1mm3 resolution. Functional images were then spatially smoothed with a Gaussian kernel of 4mm full-width at half-maximum. Within-group comparisons of the area under the curve (AUC) during covert category member generation against zero were conducte d with voxel-wise t-tests for young and older adults. Between-group comparisons were made with voxel-wise t-tests, as well. A cluster threshholding technique was impl emented to determine which areas of activation on the t-maps were significant by threshholding at a single-voxel p-value less than 0.0001 for within-subject data and less than 0.005 for betw een-subject data. The cluster size was predetermined as a region of at least 250 contiguous voxels (i.e. 250mm3). Functional Connectivity Analysis (1) Masks of each active region identified from the clustered t-maps (generated from group data) were generated to produce one set of ma sks of active regions for the young group and a second set for the old group. (2) These masks were then applied to the individual subjects within each respective group and a mean-area under the curve value was extracted for each (groupidentified) region for each subject. (3) For each age group, subject values for each active region were correlated (Pearson r) with all other regions for all group members, re sulting in a matrix of intercorrelations among active regions for each group. (4)Correlations with an associated alpha less than 0.01 were isolated and patterns of functional connectivity were identified. 19

PAGE 20

CHAP TER 3 RESULTS For a more comprehensive report of the results the reader is referred to Benjamin, Cohen et al. (in preparation). Behavioral Results On the Delis-Kaplan Executive Function Sy stem (DKEFS) animal fluency test, Young participants produced a mean of 8.2 category members in the first 15 seconds (SD=1.6) and 23.5 words at 60 seconds (SD=5.6). Old (Young-Old + OldOld) participants ge nerated a mean of 9.0 words in the first 15 seconds and 20.8 words at 60 seconds (SD=4.0). Old and Young participants did not differ on the number of category members produced at 15 seconds (mean difference=-.82, t(42)=-1.78, p<.082) or at 60 seconds (mean difference=2.64, t(42)=1.8, p<.079). When the Old group was subdivided into Young-Old and Old-Old, the Young-Old group generated 8.36 (SD=1.2) category members at 15 seconds and 22.2 (SD=3.9) at 60 seconds. The Old-Old group generated 9.63 (SD=1.4) category members at 15 seconds and 19.4 (SD=3.7) at 60 seconds. Their performances were significant ly different at 15 seconds (mean difference=1.27, t(20)=-2.25, p<.036) with the Old-Old group pr oducing significantly more words than the Young-Old group, but there was no group difference at 60 seconds (mean difference=2.27, t(20)=1.7, p<.112). The Young-Old group did not differ from the Young group at 15 seconds (mean difference=-.18, t(31)=-.33, p<.741) or at 60 seconds (mean difference=1.27, t(31)=.67, p<.459). The Old-Old group, however, did produce significantly more words than the Young group at 15 seconds (mean difference=1.45, t(31)= -2.53, p<.016) and significantly fewer words at 60 seconds (mean difference=4.0, t(31)=2.145, p<.04). Within the Old group, Young-Old and Old-Old part icipants did not diffe r in their level of education or on their performances on the BNT or HVLT. They were significantly different with 20

PAGE 21

respect to their age (discussed earlier, section 2.1) and in the sex com position of the groups (t(20)=-2.3, p<.03). The Young-Old group had 4 males and 7 females, while the Old-Old group had 9 males and 2 females. Functional Magnetic Resona nce Imaging (fMRI) Results Within-Subject fMRI Results Young participants demonstrated 7 active regi ons at the level of thresholding described previously (t=4.7, p<.0001, 250 l, 1.8mm) (Table 3-1), including two clusters in the left middle frontal gyrus, one cluster in the le ft pre-central gyrus, one in the left caudate body, one in the left pre-supplementary motor area (pre-SMA), and one in the right cerebellum. The pre-SMA was defined for each age group as the portion of Brodma nns Area (BA) 6 anterior to a coronal plane passing through the anterior commissure of the Talairached T1 image. This line has been shown to be a good landmark for distinguishing between the two regions (Picar d & Strick, 1996). The SMA was defined as the region within BA 6 posterior to this line. Old (Young-Old + Old-Old) participants demonstrated 6 large clusters of strong activity at the selected level of thresholding (Table 3-1). Though th e older adults demonstrated fewer clusters at the same threshold, the clusters tended to be larger than the ones shown by young adults and traversed multiple anatomic regions. A more stringent threshold was not used in order to maintain equivalency with the Young adults and because at higher levels of activity (i.e. greater than t=4.7, p<.0001) right cortical and su bcortical activity becomes less prominent and these were areas of special interest. Some of the demonstrated clusters were manually divided based on anatomical boundaries (for example, SMA and pre-SMA, described previously) or based on boundaries that emerged at higher thresholds. These divisions resulted in 16 anatomically distinct regions. Additionally, though the Older participants were subdivided into Young-Old and Old-Old in analys es of BOLD response, they were kept together for the 21

PAGE 22

identification of clusters in order to r etain maximum statistical power and equivalency with the group of Young participants. Active regions include the left middl e frontal gyrus, left inferior frontal gyrus, left pre-central gyrus, left insula, pre-SMA bila terally, thalamus bilaterally, caudate body bilaterally, and right SMA. Between-Subjects fMRI Results When Old (all 22) participants' mean IRF was contrasted against Young participants (t=2.91,p<.005, 250ul, 1.8mm), 10 areas of difference were revealed, all of which demonstrated greater activity in Older particip ants (Table 3-2). These areas included the left middle frontal gyrus, left inferior front al gyrus, pre-SMA bilaterally, putame n bilaterally, right thalamus, right Broca's homologue (BA 47), and right SMA. When Old-Old (n=11) participants were c ontrasted with Young-Old (n=11) participants (t=2.91, p<.005, 250ul, 1.8mm), 9 areas of differences were revealed (Table 3-2), all of which demonstrated greater activity in the Old-Old participants. Results showed that Old-Old adults showed stronger activity than Young-Old adults in the left middle frontal gyrus, left inferior frontal gyrus, left superior fr ontal gyrus, lentiform nuclei b ilaterally, left thalamus, right cingulate cortex, as well as the right precuneus. An area of activati on was also shown in the right brain stem. Functional Connectivity Results Young Participants When the mean activation of all voxels within each cluster (within-subjects; 7 clusters total) was correlated with the average activation from all other within-subjects clusters, a matrix of 21 possible correlations was produced (Figur e 3-1). Of these 21 possible correlations, the group of Young participants demons trated significant (p <.01) correlations for 13 (62%) of them. The left insula was not correlated with any othe r cluster except the right cerebellum, the left 22

PAGE 23

m iddle frontal gyrus was not correl ated with the left pre-SMA or the left pre-central gyrus. The left pre-central gyrus also faile d to show significant correlati on with the right cerebellum. All other correlations were significant. Young-Old Participants Of the 120 possible correlations between the 16 active regions; within -subjects) of Older participants, only 8 (6.7%) rela tionships were found to be si gnificant in the Young-Old group (Figure 3-2). Five of these were areas correla ting with left and right pre-SMA clusters, one between the left post-central gyrus and the left superior temporal gyrus, one between the left and right caudate, and one between the left and right thalamus. Old-Old Participants Of the 120 possible intercorrelations between active areas for the Ol der participants, the Old-Old participants demonstrated 89 (74%) si gnificant relationships (Figure 3-3). Notably, almost all (24) of the nonsignificant correlations were with the left and right thalamus. 23

PAGE 24

Table 3-1. W ithin-subject activity against baseline Brain areas showing activity at or above t=4.78 ( p <.0001) and 250mm3 (contiguous voxels) Anatomical Location BA/Subregion Volume (mm3) Peak Intensity (x, y, z) YOUNG ADULTS L Premotor L Pre-SMA/SMA 2619 40, 0, 35 L Middle Frontal Gyrus L (lateral) BA 6 765 46, -1, 27 L BA 9 1140 37, 21, 24 L Motor L BA 4 462 39, 6, 50 L Caudate 790 14, -6, 13 L Insula L BA 13/14 380 27, -20, 7 R Cerebellum R Culmen 818 -35, 55, -26 OLDER (all 22) ADULTS L PCG, MFG, IFG* L BA 6,9, 45,46 8701 40, 0, 35 L/R pre-SMA/ SMA** L/R BA 6 5478 3, 2, 55 L Caudate, Lentiform, Thal* 1680 19, -5, 9 R Caudate, Lentiform, Thal* 905 -17, -9, 8 L Precuneus L BA 19 422 27, 70, 40 L MFG (superior) L BA 6 258 25, -5, 58 L=left; R=right; IFG=inferior frontal gyrus MFG=middle frontal gyru s PCG=precentral gyrus SMA=supplementary motor area Older adults demonstrated very strong activity. Clusters are reported here at the same threshold as young adults but the clusters marked with an (*) were manually divided based on anatomical divisions and boundaries that emerged at higher levels of thresholding (not shown). The manually divided clusters are the ones used in the dedifferentiation analysis. **SMA and pre-SMA were activated in both hemispheres. Because these areas demonstrated such strong activation, and because the SMA/pr e-SMA are medial structures and drain into the same sinus(FACT CHECK), the activation seen in both hemispheres was comb ined into one large cluster at this threshold and were manually isolated. 24

PAGE 25

Table 3-2. Between-su bject activity Brain areas showing activity at or above t=3.15 ( p <.005) and 250mm3 (contiguous voxels) Anatomical Location BA/Subregion Volume (mm3) Peak Intensity (x, y, z) OLDER(n=22) vs YOUNG(n=22) L MFG 631 L Middle Frontal Gyrus BA 10, 46 606 L Inferior Frontl Gyrus BA 47 308 L pre-SMA/SMA BA 6 786 L Superior Temporal Gyrus BA 22 290 L Lentiform Nucleus 1826 22, 4, 0 R Middle Frontal Gyrus 771 R Pre-Motor BA 6 (lat) 1043 -46, 0, 26 R Lentiform Nucleus 1992 -26, -12, 5 R Ventral Thalamus 2269 -6, 7, 12 OLD-OLD(11) vs YOUNG-OLD(11) L Inferior Frontal Gyrus L BA 4 284 34, 16, -6 L Middle Frontal Gyrus L BA 46 L Middle Frontal Gyrus L BA 47 L Superior Frontal Gyrus L BA 6 L Lentiform nucleus 584 32, 6, 2 L Subthalamic Nucleus 644 10, 9, 1 R Lentiform Nucleus 656 -18, 2, -2 R Brainstem 339 -6, 19, -26 R Cerebellum R Precuneus 660 -2, 36, 47 25

PAGE 26

Figure 3-1. Young adults against baseline. Figure 3-2 Young-old adu lts against baseline. 26

PAGE 27

Figure 3-3 Old-old adults against baseline. 27

PAGE 28

CHAP TER 4 DISCUSSION Consistent with previous investigations of HAROLD (hemispheric asymmetry reduction in older adults), Older adults' activ ity during this task was observed to be more diffuse than Young adults. That is, older adults showed stronger activity and in more areas than the young adults (including right hemisphere struct ures) and engaged the same stru ctures as young adults but with less precision. The Young adults did not show greater activity than older adults in any area. If the groups differed in the number of category member s generated, these results could have been explained by performance differe nces. However, the groups of Y oung and (all) Older adults did not differ on the test of spoken category member generation performed outside of the scanner at 15 seconds, so we believe that th eir covert generation in the scanner was most likely equivalent. It is therefore believed that the observed result s reflect true age-related differences and are not due to more superficial explanations. Also, Ol d-Old adults demonstrated 9 areas of greater activity than Young-Old adults, which suggest s that the HAROLD phenomenon is progressive with age. That is, the phenomenon is noticeable when comparing 66-year-olds to 25-year-olds, and also when comparing 76-year-olds to 66-year-olds. The consideration of functi onal connectivity data provides interesting insight into the HAROLD occurrence. Young adults display correlati ons (connections) that are consistent with the structures indicated by Crosson et al. (B. Cr osson et al., 2003) to be involved in the task, including connections between the left pre-SMA, caudate, and frontal cortical regions. The most interesting results emerge when the Young-Old a dults are contrasted with the Old-Old adults. Despite showing significant activity in 16 regions, the Young-Old participants only demonstrated 8 significant connections between them (out of a possible 120). Old-Old adults, however, showed 89 significant connections between these same regions. This suggests that spatial 28

PAGE 29

changes in activation in old age (e.g. HAROLD) do not necessarily mean functional changes. That is, in Old-Old adults (mean age=76), the di ffuse activation does appear to imply functional dedifferentiation. Young-Old adults however (mean age=66), s eem to show HAROLD-typical activity without functionally integrating active regions. Based on the specific patterns of functional conn ectivity in the different groups, we propose three possible explanations of the findings. The first explanation describes the findings in the context of the brain compensating for agerelated losses, like what has been previously described with HAROLD findings (Cabeza, 2002). The second involves age-specific deterioration of basal ganglia pathways. The third way reflects a lack of selective engagement of cortical structures from the thalamus (Nadeau & Crosson, 1997). Compensation The most commonly-endorsed explanation of HAROLD activity is that the diffuse activation reflects the brai n trying to recruit more neural reso urces to compensate for age-related deterioration. It is possible th at the results from the Young-Old group show an early stage in the recruitment of more resources. The neurons of th e newly-recruited areas ar e firing (and therefore show significant associated BOLD signal) but their contribution is not yet integrated into functional networks (as inferred from a lack of co rrelations). This is supp orted by the behavioral data that showed that Old-Old adults generated significantly mo re items than Young-Old adults in the first 15 seconds of the DKEFS semantic fluency task, suggesting that they have more successfully compensated for age-related declines than the Young-Old adults. It is also interesting to note that in the Young-Old adults, 6 of their 8 significant connections are between homologous or nearly-homologous re gions: left and right caudate, le ft and right thalamus, left and right pre-SMA. This suggests that the first connections that are ma de in the process of compensation are the ones between homologous regions. 29

PAGE 30

Age-Specific Deterio ration of Basal Ganglia Circuitry A second explanation of the results involves uni que basal ganglia deterioration associated with each age group. This discussion requires a brie f orientation to the functional neuroanatomy of category member generation, pr eviously explored by Crosson et al. (B. Crosson et al., 2003; B. Crosson, Benjamin, M., and Levy, I, 2007) and others. These authors modified previous ideas of basal ganglia functioning (Alexander et al., 1986; Mink, 1996; Penney & Young, 1986; Strick, 1995) and applied them to language tasks. Thei r model proposed that the basal ganglia are employed to help refine cortical activation so that the neural representations of individual lexical and semantic items (e.g. category members) are ra pidly enhanced and then suppressed in favor of the next produced item. Given a category (e.g. birds), the signal to noise ratio (SNR) is refined for a chosen response (e.g. birds) by enhanc ing the unique signal associated with that item (e.g. robin) while suppressing related and competing items (e.g. blue jay). The basal ganglia are then also recru ited to suppress robin in favor of the next item. Thus, the basal ganglia help improve the speed and accuracy of response, whil e also serving as a clu tch in the transition between items. Figure 4-1 describes the relati onship among structures in thre e subcortical circuits. The direct pathway is believed to enhance the signal of the desired output and involves projections from the pre-SMA to a dedicated neostriatal com ponent, and then to the medial globus pallidus (GPi). Striatal spiny output neur ons that participate in the direct loop cycl e through two states: a hyperpolarized down state and a less polarized up state that prepares the neurons for an action potential. Because excitato ry cortical inputs terminate rath er diffusely on striatal neurons, it takes a strong and coordinated co rtical input to excite those neurons into the up state. However, the transition to up state is facil itated by the activation of dopamine D1 receptors that are innervated with dopamine from the subs tantia nigra pars compacta (SNpc). Thus, active 30

PAGE 31

D1 recepto rs may facilitate selected actions or cognitions through prolonging the activity supporting them (B. Crosson, Be njamin, M., and Levy, I, 2007). Competing responses to the se lected output are attenuated by the indirect loop which involves projections from the striatum to the la teral globus pallidus (GPL) to the subthalamic nucleus (STN), and then to the medial globus pallidus. Input from the direct and indirect pathways summate at the GPm, which then dete rmines the degree of suppression that the GPm exerts on the thalamus. The greater the activity of the GPm, the more suppression is imposed on the thalamus. The indirect loop is believed to be mediated by dopamine D2 receptors, though the exact mechanics are far less understo od than for the direct pathway. The role of the basal ganglia during category member generation is further complicated by the presence of a third loop, the hyperdirect loop (Nambu et al., 2000) that projects excitatory information directly from the pre-SMA to the STN, which then causes the GPm to inhibit the thalamus. These three loops work together to form a temporal sequence of suppress-enhancesuppress. Nambu et al. showed that following ex citation of motor corti ces, the GPm experienced three waves of input. The first (8 ms after stimulation) was excita tory and was believed to be the hyperdirect pathway. By strongly activating the GPm, the thalamus is strongly inhibited and it is speculated that this may serve to reset the basal ganglia circuitry in prep aration for changing to a new activity (Nambu et al., 2002). The second wave (22 ms) was inhibitory on the GPm and was believed to be the direct pathway (enhancing desi red output). Finally, the third wave (30 ms) was excitatory and was believed to be the indirect pathway (suppre ss competing output). These three pathways and their temporal pattern will help in terpret the connectivity changes observed in the results of the present investigation. 31

PAGE 32

It is possible that the differe nt patterns of functional conne ctivity seen in Y oung-Old and Old-Old adults reflect age-specific losses of th e direct and indirect pathways. A loss of the efficacy of the direct (enhance) pathway might acc ount for the lack of f unctional recruitment of active regions seen in the Young-Old group, while a loss of the efficacy of the indirect pathway might cause the widespread functional engagement of regions seen in Old-Old adults. That is, they are no longer able to suppress unwanted noise It is already known th at aging results in a loss of basal ganglia dopamine receptors, both D1 (Wang et al., 1998) and D2 (Vernaleken et al., 2007). However, it is not know at what rate or du ring which critical period s these declines occur. Perhaps the results of the present investigation firs t reflect a loss of D1 receptors that mediate the direct pathway (age 66), followed by a loss of D2 receptors that mediate the indirect pathway (age 76). A third possibility is that the widespread f unctional connectivity seen in the Old-Old group reflects an inability to selectively engage task-related neural ne tworks. Nadeau & Crosson (1997) discussed the possibility that dysfunction of the pulvinar and lateral posterior nuclei of the thalamus may disrupt the ability of the thalamus to enlist specific neural nets to complete a task without engaging unnecessary reso urces. Perhaps the relative lack of connectivity with the thalamus in the Old-Old group is indicative of thalamic dysfunction and the widespread functional connectivity of all other regions results from the th alamus inability to confine neural activity to critical netw orks only. Though conceptually similar to the previous possibility, a decreased ability to selectively engage neural ne ts represents a recruiting deficiency, whereas the previous possibility represents an impaired ab ility to refine the output of already-engaged structures. 32

PAGE 33

There are several im portant lim itations of the investigation that should not be overlooked. First, it is not known to what degree these findi ngs will generalize to cognitive tasks other than category member generation or to groups of a lower education level. Second, as with any crosssectional analysis of dedifferentiation comes the assumption that between-person differences will be reflective of intraindividual changes across time. Though this assumption is an accepted one, further analysis with longitudinal analyses is necessary (Sims, 2008). 33

PAGE 34

Figure 4-1. Direct, indir ect, and hyperdirect loop s of the basal ganglia. 34

PAGE 35

35 CHAPTER 5 CONCLUSION Results show that patterns typical of the HAROLD phenomenon (hemispheric asymmetry reduction in older adults) can be informed by an alyses of the functiona l connectivity of active regions and that using the mean (blood oxygena tion level dependent; BOLD) impulse response function is one method of doing so. Specifically, f unctional connectivity anal ysis has shown that in Old-Old adults (mean age=76), HAROLD-typical activation does appear to reflect dedifferentiation of function, as supported by the functional integr ation of (almos t) all active regions. However, results from the Young-Old gr oup (mean age=66) suggest that there appears to be a preceding stage whereby many neuroanatomical regions are active but not functionally integrated. Perhaps this reflects an early stage of a compensatory mechanism that is recruiting more resources to account for ag e-related decline. Alternatively, the results might be explained as a loss of D1 receptors in the Young-Old group, followed by a loss of D2 receptors in the OldOld group. Finally, perhaps the results reflect an in ability of the thalamus to selectively engage networks needed for the task.

PAGE 36

LIST OF REFERE NCES Alexander, G.E., DeLong, M.R., & Strick, P.L. (1986). Parallel organi zation of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience, 9, 357-381. Anstey, K.J., Hofer, S.M., & Luszcz, M.A. (2003) Cross-sectional and longitudinal patterns of dedifferentiation in late-life cognitive and se nsory function: The effects of age, ability, attrition, and occasion of measurement. Journal of Experimental Psychology: General, 132(3), 470-487. Balinksy, B. (1941). An analysis of the mental factors of various groups from nine to sixty. Genetic Psychology Monographs, 23, 191-234. Baltes, P.B., Cornelius, S.W., Spiro, A., Nesselroa de, J.R., Willis, S.L. (1980). Integration versus differentiation of fluid/crystallized intelligence in old age. Developmental Psychology, 6, 625-635. Baltes, P.B., & Lindenberger, U. (1997). Emerge nce of a powerful connection between sensory and cognitive functions across the adult li fe span: A new window to the study of cognitive aging? Psychology and Aging, 12(1), 12-21. Cabeza, R. (2002). Hemispheric asymmetry redu ction in older adults: The HAROLD model. Psychology and Aging, 17(1), 85-100. Cox, R.W. (1996). AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical Research, 29(3), 162-173. Crosson, B., Benefield, H., Cato, M.A., Sadek, J.R., Moore, A.B., Wierenga, C.E., Gopinath, K., Soltysik, D., Bauer, R.M., Auerbach, E.J., Gokcay, D., Leonard, C.M., & Briggs, R.W. (2003). Left and right basal ganglia and frontal activity during language generation: Contributions to lexical, sema ntic, and phonological processes. Journal of the International Neurops ychological Society, 9(7), 1061-1077. Crosson, B., Benjamin, M., and Levy, I. (2007). Role of the Basal Ganglia in Language and Semantics: Supporting cast. In J.H.a.M. Kraut (Ed.), Neural Basis of Semantic Memory (pp. 219-243). Cambridge: Cambridge University Press. Crosson, B., McGregor, K., Gopinath, K.S., Conw ay, T.W., Benjamin, M., Chang, Y.L., Moore, A.B., Raymer, A.M., Briggs, R.W., Sher od, M.G., Wierenga, C.E., & White, K.D. (2007). Functional MRI of language in aphasi a: A review of the literature and the methodological challenges. Neuropsychology Review, 17(2), 157-177. de Frias, C.M.L., M; Lindenberger, U; Nil sson, L. (2007). Revisiti ng the dedifferentiation hypothesis with longitudinal multi-cohort data. Intelligence, 35, 381-392. Deary, I.E., V; Gibson, GJ; Aus tin, EJ; Brand, CR; Kellaghan, T. (1996). Intelligence and the Differentiation Hypothesis. Intelligence, 23, 105-132. 36

PAGE 37

Delis, D.C., Kaplan, E., & Kram er, J. H. (2001). Delis-Kaplan Executive Functioning System (DKEFS): Examiner's manual San Antonio: Psychol ogical Corporation. Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189-198. Garrett, H.E. (1946). A developmental theory of intelligence. American Psychologist, 1, 372378. Ghisletta, P., & Lindenberger, U. (2003). Age-based structural dynamics between perceptual speed and knowledge in the Berlin Ag ing Study: Direct evidence for ability dedifferentiation in old age. Psychology and Aging, 18(4), 696-713. Hayslip, B., Jr., & Sterns, H.L. (1979). Age differen ces in relationships be tween crystallized and fluid intelligences and problem solving. Journal of Gerontology, 34(3), 404-414. Kaplan, E.F., Goodglass, H ., & Weintraub, S. (1983). The Boston Naming Test Philadelphia: Lea & Febiger. Kinsbourne, M. (1980). Mapping a be havioral cerebral space. International Journal of Neuroscience, 11(1), 45-50. Li, S.C., Brehmer, Y., Shi ng, Y.L., Werkle-Bergner, M., & Lindenberger, U. (2006). Neuromodulation of associative and organizational plasticity across the life span: Empirical evidence and neurocomputational modeling. Neuroscience and Biobehavioral Reviews, 30(6), 775-790. Li, S.C., & Lindenberger, U. (1999). Cross-Level Unification: A Computational Exploration of the Link Between Deterioration of Neurotra nsmitter Systems and Dedifferentiaion of Cognitive Abilities in Old Age. In L.-G Nilsson & H. Markowitsch (Eds.), Cognitive Neuroscience of Memory (pp. 103-146). Kirkland: Hogrefe & Huber. Li, S.C., Lindenberger, U., & Sikstrom, S. (2001) Aging cognition: From neuromodulation to representation. Trends in Cognitive Science, 5(11), 479-486. Mink, J.W. (1996). The basal ganglia: Focused selection and inhibition of competing motor programs. Progress in Neurobiology, 50(4), 381-425. Nadeau, S.E., & Crosson, B. (1997). Subcortical aphasia. Brain and Language, 58(3), 355-402; discussion 418-323. Nambu, A., Tokuno, H., Hamada, I., Kita, H., Im anishi, M., Akazawa, T., Ikeuchi, Y., & Hasegawa, N. (2000). Excitatory cortical inputs to pallidal neurons via the subthalamic nucleus in the monkey. Journal of Neurophysiology, 84(1), 289-300. Nambu, A., Tokuno, H., & Takada, M. (2002). Functional significance of the corticosubthalamo-pallidal 'hyperdirect' pathway. Neuroscience Research, 43(2), 111-117. 37

PAGE 38

Oades, R.D. (1985). The role of noradrenaline in tuning and dopam ine in switching between signals in the CNS. Neuroscience and Biobehavioral Reviews, 9(2), 261-282. Ogawa, S., Lee, T.M., Kay, A.R., & Tank, D.W. (1990). Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proceedings of the National Academy of Science U S A, 87(24), 9868-9872. Oldfield, R.C. (1971). The assessment and anal ysis of handedness: The Edinburgh inventory. Neuropsychologia, 9(1), 97-113. Park, D.C., Polk, T.A., Park, R., Minear, M., Savage, A., & Smith, M.R. (2004). Aging reduces neural specialization in ventral visual cortex. Proceedings of the National Academy of Sciences U S A, 101(35), 13091-13095. Penney, J.B., Jr., & Young, A.B. (1986). Striatal inhomogeneities and basal ganglia function. Movement Disorders, 1(1), 3-15. Picard, N., & Strick, P.L. (1996). Mo tor areas of the medial wall: A review of their location and functional activation. Cerebral Cortex, 6(3), 342-353. Rajah, M.N., & D'Esposito, M. (2005). Region-specifi c changes in prefrontal function with age: A review of PET and fMRI studies on working and episodic memory. Brain, 128(Pt 9), 1964-1983. Rogers, B.P., Morgan, V.L., Newton, A.T., & Gore, J.C. (2007). Assessing functional connectivity in the human brain by fMRI. Magnetic Resonance Imaging, 25(10), 13471357. Salthouse, T.A. (2000). Methodological Assumpti ons in Cognitive Aging Research. In T.A.C. Salthouse, F. I. M. (Ed.), The Handbook of Aging and Cognition (pp. 467-498). Mahwah, NJ: Erlbaum. Salthouse, T.A., Hancock, H.E., Meinz, E.J., & Ha mbrick, D.Z. (1996). Interrelations of age, visual acuity, and cognitive functioning. Journals of Gerontology Se ries B: Psychological Sciences and Social Sciences, 51(6), P317-330. Salthouse, T.A.C., F.I.M. (2000). Closing Comments. In S. Craik (Ed.), The Handbook of Aging and Cognition (pp. 689-703). Mahwah, NJ: Erlbaum. Sawaguchi, T., & Goldman-Rakic, P.S. (1991). D1 dopamine receptors in prefrontal cortex: Involvement in working memory. Science, 251(4996), 947-950. Shapiro, A.M., Benedict, R.H., Schretlen, D., & Brandt, J. (1999). Construct and concurrent validity of the Hopkins Ve rbal Learning Test-revised. Clinical Neuropsychology, 13(3), 348-358. 38

PAGE 39

Sim s, R.C., Allaire, J.C., Gamaldo, A.A., Edward s, C.L., Whitfield, K.E. (2008). A Examination of Dedifferentiation in Cognition Am ong African-American Older Adults. Journal of Cross Cultual Gerontology. Spearman, C. (1904). "General Intelligence," Objectively Determined and Measured. American Journal of Psychology, 15, 201-293. Spearman, C. (1927). The abilities of man: Th eir nature and measurement New York: Macmilliam. Strick, P.L., Dum, R.P, & Pica rd, N. (1995). Macro-organization of the circuits connecting the basal ganglia with the cortical motor areas. In J.L.D. J.C. Houk, & D.G. Beiser (Ed.), Models of information pro cessing in the basal ganglia (pp. 117130). Cambridge: MIT Press. Talairach, J., Tournoux, P. (1988). Planar Stereotaxic Atlas of the Human Brain New York: Thiem Medical Publishers. Tivarus, M.E., Hillier, A., Schmalbrock, P., & Beversdorf, D.Q. (2008). Functional connectivity in an fMRI study of semantic and phonologi cal processes and the effect of L-Dopa. Brain and Language, 104(1), 42-50. Tucker-Drob, E.M., & Salthouse, T.A. (2008). Adult age trends in the rela tions among cognitive abilities. Psychology and Aging, 23(2), 453-460. Vernaleken, I., Weibrich, C., Siessmeier, T., Buchholz, H.G., Rosch, F., Heinz, A., Cumming, P., Stoeter, P., Bartenstein, P., & Grunder, G. (2007). Asymmetry in dopamine D(2/3) receptors of caudate nucleus is lost with age. Neuroimage, 34(3), 870-878. Voss, M.W., Erickson, K.I., Chaddock, L., Prakash, R.S., Colcombe, S.J., Morris, K.S., Doerksen, S., Hu, L., McAuley, E., & Kramer, A.F. (2008). Dedifferentiation in the visual cortex: An fMRI investigation of individual differences in older adults. Brain Research Wang, Y., Chan, G.L., Holden, J.E., Dobko, T., Ma k, E., Schulzer, M., Huser, J.M., Snow, B.J., Ruth, T.J., Calne, D.B., & Stoessl, A.J. (1998). Age-dependent decline of dopamine D1 receptors in human brain: A PET study. Synapse, 30(1), 56-61. Zarahn, E., Rakitin, B., Abela, D., Flynn, J., & St ern, Y. (2007). Age-related changes in brain activation during a delaye d item recognition task. Neurobiology of Aging, 28(5), 784-798. Zelinski, E.M., & Lewis, K.L. (2003). Adult ag e differences in multiple cognitive functions: Differentiation, dedifferentiation, or process-specific change? Psychology and Aging, 18(4), 727-745. 39

PAGE 40

40 BIOGRAPHICAL SKETCH Matthew L. Cohen graduated magna cum laude from the University of Delaware (UD) in 2007 with an honors Bachelor of Arts degree with distinction in psychology and with minors in biology and Spanish. His resear ch at UD focused on the in terrelationship between young childrens theory-of-mind and langua ge acquisition. He began his graduate studies in clinical psychology, specializing in clinical neuropsychology at the University of Florida in the Fall of 2007 and was awarded an Alumni Fellowship. He has been mentored by Dr. Bruce Crosson and his research has used neuroimaging to study s ubcortical contributions to language production both in healthy older adults a nd clinical popula tions. He obtained his M.S. degree in May, 2009 and is pursuing a Ph.D.