This item is only available as the following downloads:
1 DISEASE By AMANDA GARCIA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013
2 2013 Amanda Garcia
3 To my loving family
4 ACKNOWLEDGMENTS I would like to thank Dr. Bruce Crosson for his support and mentorship, as well as my supervisory committee Dr. Reilly, Dr. Marsiske, Dr. Janicke, and Dr. Pereira, for their contributions. I would also like to extend thanks and acknowledgement to my lab ma tes for their guidance and aide on this project, especially Michelle Benjamin, Stephen Towler, and Kristin Moffett. Finally, I would like to thank my mother and my sister for their continued encouragement, love, and support.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 Language Production ................................ ................................ .............................. 10 ................................ ................................ ........... 12 (fMRI) ................. 13 .............................. 15 2 METHODS ................................ ................................ ................................ .............. 19 Participants ................................ ................................ ................................ ............. 19 Procedures ................................ ................................ ................................ ............. 21 Neuropsychological Testing ................................ ................................ ............. 21 fMRI Naming Task ................................ ................................ ............................ 23 fMRI Motor Task ................................ ................................ ............................... 24 Image Acquisition ................................ ................................ ............................. 25 Neuroimaging Data Analyses ................................ ................................ ........... 25 Region of Interest (ROI) Analyses ................................ ................................ ... 27 Voxel Based Morphometry (VBM) Analyses ................................ ..................... 27 3 RESULTS ................................ ................................ ................................ ............... 30 Behavioral Results ................................ ................................ ................................ .. 30 fMRI ROI Analysis Results ................................ ................................ ...................... 30 VBM Results ................................ ................................ ................................ ........... 31 VBM: Whole Brain Results ................................ ................................ ............... 31 VBM: ROI Analyses ................................ ................................ .......................... 31 4 DISCUSSION ................................ ................................ ................................ ......... 37 LIST OF REFERENCES ................................ ................................ ............................... 43 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 48
6 L IST OF TABLES Table page 2 1 Demographics of the healthy control (HC) and Alzheimer's disease/mild cognitive impairment (AD/MCI) groups ................................ ............................... 29 3 1 Neuropsychological raw/composite scores of the healthy control (HC) and Alzheimer's disease/mild cognitive impairment (AD/MCI) groups ....................... 33 3 2 Spearman correlations between functional activity and behavioral language performance ................................ ................................ ................................ ....... 33 3 3 Spearman correlations between grey matter density and behavioral language perofrmance ................................ ................................ ................................ ....... 34
7 LIST OF FIGURES Figure page 1 1 A depiction of the modular model of language. ................................ ................... 18 2 1 Schematic of picture naming task. ................................ ................................ ...... 29 3 1 Hemodynamic response function of activity in the right posterior perisylvian cortex (PPS) ; scatter plots of the relationship between average a rea under the curve in right PPS and language function ................................ ................... 35 3 2 Whole brain comparison of grey matter density between healthy control (HC) and Alzheimer s disease ( AD ) / Mild C o gnitive Impairment (MCI) groups. ........... 36
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DISEASE By Amanda Garcia M ay 2013 Chair: Bruce Crosson Major: Psychology lexical functioning normally mediated by the inferior temporal (ITL) and posterior perisylvian cortices (PPS), respectively. The current s tudy utilized functional magnetic resonance imaging (fMRI) during picture naming and voxel based morph o m e try (VBM) to determine the relationship between activity (fMRI) and structural changes (VBM) in these language areas and semantic and lexical performan ce in early AD and healthy controls (HC). The functional activity of 10 participants with mild AD or multi modality amnestic mild cognitive impairment (MCI) and 12 age and education matched controls was extracted from regions of interest (ROIs) (right and left ITL, PPS, and frontal lobe). Magnitude of functional activity was then correlated with a semantic composite score (Animal Fluency, Pyramids and Palm Trees) and a lexical composite score (Letter Fluency, Phonological Blocking). T tests of cortical dens ity between AD and HCs were also conducted. Average grey matter densities within areas of group difference within the ROIs were correlated with lexical and semantic scores. There was a significant negative correlation between the both lexical and semantic measures and functional activity in right PPS, r s (19) = .564, p = .048, r s (20) = .625, p = .012, respectively. There
9 was also a significant positive correlation between the cortical density in the right PPS and semantic composite score, r s (20) = .577, p = .03. These results do not support our a priori hypotheses; however, they may suggest that the increase in right PPS activity reflects an extension of healthy aging processes that is aggravated by AD pathology. Cortical density reductions in the same area may be accompanied by loss of inhibitory functions, such that the increased activity in this area interferes with semantic functions.
10 CHAPTER 1 INTRODUCTION (AD) is most commonly associated with degraded explicit memory, these patients also experience impaired communication. One common language deficit is most often iden tified clinically by diminished picture naming ability. Anomia presentation is not homogenous across patients, however, possibly due to the distinct cognitive processes involved in the task. This study specifically focused on the contributions of semantic understand the meaning of the picture they were being asked to name (semantic processing) and choose the correct word for that object (lexical retrieval). It further ex amined the neural correlates of these cognitive substrates to elucidate the variable nature of naming deficits. Language Production Language production, while highly complex, can be more readily understood when broken down into distinct but interactive cog nitive components (e.g. semantic, lexical, phonemic components). The modular model put forth by Ellis and Young (1988) emphasizes the importance of the semantic system in the language production process (Figure 1 1). In this model, multiple lexicons feed i nto the semantic system. Spoken and written words enter the model through the phonological and orthographic input lexicons, which recognize the words forms as familiar. Visual objects similarly enter the model through the structural description system, whi ch acts like an input lexicon to recognize the visual form as familiar. These familiar forms then enter the semantic system, where they are assigned meaning. Meaningful concepts are next prepared to be spoken or
11 written: the proper word form is retrieved f rom the phonologic or orthographic output lexicons, allowing the concept to be expressed out loud or on paper. In this way, a centralized semantic system is integral to both assigning meaning to concepts/objects that are encountered and generating the word s used to represent those concepts Though this model is by no means all encompassing, its relative simplicity is especially useful for the discussion of picture naming and the brain. The distinct cognitive processes it outlines (i.e. recognition, lexical retrieval, and semantic processing) are associated with functional neuroanatomical regions. For instance, lesion studies suggest that lexical and associated phonological processes are supported by posterior perisylvian cortices in the language dominant hem isphere (Nadeau, 2000; Ojemann, 1983) A similarly discrete area that supports semantic processes is more difficult to pinpoint given the distributed nature of the semantic system. However, when considering the semantic system in relation to the picture na ming task examined here, the left ventral visual stream emerges as important. This brain area, which consists of the fusiform, lingual, and parahippocampal gyri, has been implicated in the processing of semantic visual information (Chao, Haxby, & Martin, 1 999; Feinberg, Schindler, Ochoa, Kwan, & Farah, 1994; Galton et al., 2001; Gold et al., 2006) Additional areas of importance include the left inferior and middle temporal gyri, which are noted here for their role in managing the connection between semanti c concepts and their representations in the output lexicon (Foundas, Daniels, & Vasterling, 1998; Raymer, Moberg, Crosson, Nadeau, & Rothi, 1997) Of note, these semantic areas (i.e. left fusiform, lingual, parahippocampal, inferior temporal, and middle te mporal gyri) additionally correspond to one of the semantic convergence zones
12 proposed by Binder, Desai, Graves, and Conant (2009) Together, then, they are implicated in the integration of semantic features across modalities as well as semantic concept re trieval. documented (e.g. Hodges, Salmon, & Butters, 1991; Salmon, Butters, & Chan, 1999; Salmon, Heindel, & Lange, 1999) .These studies promoted the idea that naming deficits resulted from degraded semantic memory stores by remarking on loss of concept knowledge across task. Investigations of primed picture naming also suggested, however, that impaired lexical retrieval of intact semantic stores p l ay a role in expressed anomia. Faust et al. (2004) for example, found that mild AD patients but not healthy older adults were susceptible to interference by phonologically similar words. Such phonological interference stresses the lexical nature of their deficit. Other studies have found additional evidence of phonological errors (i.e. substitution, addition, omission of phonemes) across task and suggest that difficulty activating lexical representations may be responsible for some semantic errors expresse d downstream (Croot, Hodges, Xuereb, & Patterson, 2000; Galton, Patterson, Xuereb, & Hodges, 2000; Moreaud, David, Charnallet, & Pellat, 2001) Furthermore, case studies of atypical AD pathology reveal that these patients may present with aphasia as the pr ominent AD symptom or demonstrate phonological errors in word retrieval. In this way, it is likely that impaired semantic and lexical processing deficits contribute to poor naming performance in AD, though one may be more prominent than the other in any gi ven patient (Croot et al., 2000; Galton et al., 2000).
13 Though the behavioral contributions of semantic and lexical processing to naming performance in AD are well studied, fewer investigations have looked at the neurocorrelates of these processes. Broadly, a pattern has emerged with regards to the processing of verbal memory in these patients. Bookheimer et al. (2000) compared brain activation during a word pairing task in older adults at risk or n ot at risk for areas important for task performance, such as the left hippocampus, parietal lobe, and prefrontal cortex. Bookheimer et al. interpreted this activation as compensatory, which was supported by the fact that increased activation predicted cognitive decline two years after scanning. During a similar task, those with mild AD showed decreased activation in the left hippocampus and parietal lobe compared to healt hy older adults (Backman et al., 1999) When considered together, these findings suggest that pre symptomatic AD patients show increased activation in structures important for verbal memory. As deficits begin to emerge, however, deterioration of these stru ctures is expressed in a reduction of activation. Though the literature is somewhat inconsistent, this pattern is generally repeated in studies directly examining language and AD. Lexical and semantic processing have not been directly compared or distingu ished, however, which may account for the discrepancies. Grossman Koenig, Glosser e t al. (2003) for example, compared activation patterns during a semantic judgment task of nouns. They demonstrated decreased activation for AD patients compared to healthy older adults in left posterior perisylvian and superior temporal cortex, coupled with increased activation in adjacent left inferior temporal cortex. Similar activations patterns were shown during a semantic
14 judgment task of verbs. AD patients had weaker activation of left posterior perisylvian regions coupled with increased activation of smaller adj acent areas (Grossman, Koenig, DeVita, et al., 2003) In both cases the increased activation was cautiously interpreted as compensatory. Further analyses were not conducted, however, relating accuracy to activation increases, so the degree to which this ac tivity is truly beneficial is unclear. The frontal lobes additionally demonstrate increased activation in early AD patients. Some degree of frontal involvement in language tasks is expected, as these regions participate in selection and controlled semant ic retrieval (Binder et al., 2009; Wagner, Pare Blagoev, Clark, & Poldrack, 2001) However, several studies have shown the left dorsolateral prefrontal cortex to be more active in early AD patients than in healthy older adults during both verbal and semant ic memory tasks (Becker et al., 1996; Saykin et al., 1999) Because this brain region is relatively less affected by neurofibrillary tangles early in the disease process (Braak & Braak, 1991; Foster et al., 1984; Wilcock & Esiri, 1982) the activation has been viewed as compensating for atrophy of more posterior lexical semantic regions (i.e. the PPS and ITL). This viewpoint is supported by the positive association between activity of the left prefrontal cortex and semantic performance demonstrated by Sayki n et al. (1999). In studies relating such activity and atrophy, a positive association was detected between the two in the left inferior frontal gyrus (Johnson et al., 2000) The same relationship was not detected between atrophy and activity in the superi or termporal gyrus, raising the possibility that increased compensatory activity is an effect specific to this area. Wierenga and colleagues (2011) additionally found increased response in frontal regions in patients with AD, including the right inferior f rontal gyrus. This brain pattern has also been
15 documented in healthy aging and may have been an example of the hemispheric asymmetry reduction in older adults (HAROLD) phenomenon. Of note, the functional implications of HAROLD are controversial: while some view the increased positive activity as compensatory (Park & Reuter Lorenz, 2009) others have linked it to decreased semantic performance in older adults (Meinzer et al., 2012) Thus, the compensatory nature of increased activity in the right frontal lob patients is uncertain. Though it may have the same relationship to behavioral performance as activity in the left hemisphere, it may instead interfere with successful language functioning. Structural Neu roimaging Studies investigating the relationship between grey matter atrophy and language deficits are fairly consistent with the functional neuroimaging research. Galton and colleagues (2001) for example, were interested in common neurocorrelates of sema ntic with these diagnoses, they were able to detect a positive correlation between semantic functioning and grey matter in the inferior and middle temporal gyri. Thus, a trophy of areas previously implicated in semantic processing was associated with lower scores on neuropsychological measures of semantic functioning. Later studies utilized voxel based morphometry (VBM) to look at the relationship between atrophy and namin g deficits in AD specifically (Gee et al., 2003; Grossman et al., 2004) They showed that AD patients had a pattern of atrophy that included key language areas underlying semantic and lexical processes, such as the left anterior and posterior lateral tempo ral lobes, left inferior temporal lobe, and left posterior perisylvian cortex. Of these areas, left anterior and lateral temporal atrophy were associated with poorer naming performance.
16 This association supports the idea that these areas are associated wit h semantic deficits, as cognitive testing revealed the AD patients performed significantly worse than healthy controls on measures of animal fluency. Studies of atrophy patterns in atypical on of lexical deficits. Croot et al. (2000) reported on 13 patients who presented with deficits in articulatory and phonological processing. Structural neuroimaging of these patients showed associated with acquired speech and language disorders. That is, patients with lexical retrieval deficits appeared to have greater left posterior perisylvian pathology. In light of these findings, the current study will explore the nature of naming defici evidence of the relationship between the posterior perisylvian cortex and lexical deficits in these patients, we would expect a positive correlation between lexical functioni ng and functional activity during a language task in this region. We would expect a positive correlation between lexical functioning and grey matter density of this area, as well. In terms of semantic functioning, we would expect significantly positive cor relations between functional activity and grey matter density of its associated neurocorrelate, the inferior temporal lobe. Finally, AD patients seem to demonstrate increased frontal activity during language tasks, but this activity does not appear to be s pecifically related to lexical or semantic functioning. Thus, we hypothesized the increased activity would be negatively related to lexical and semantic functioning. Because increased activity of the frontal lobes corresponds to increased atrophy of those regions, we hypothesized
17 that grey matter density would be positively related to both measures of language functioning.
18 Figure 1 1. A depiction of the modular model of language.
19 CHAPTER 2 METHODS Participants Twelve older healthy adults at least 65 years of age (7 female) and ten participants (5 female) with early to moderate stage AD or multi modality amnestic Mild Cognitive Impairment (MCI) participated in the current study. Amnestic MCI patients were recruit disease (Dubois et al., 2010) and those with impairments across multiple domains have increased conversion rates to AD (Tabert et al., 2006) In the impaired group, six participants ha d a diagnosis of AD and four had a diagnosis of MCI with impairments in memory and at least one other domain. Of note, all participants with multi modalit y amnestic MCI met NINCDS criteria for probable degenerative dementia of the functional impairment is not required for this classification. As shown in Table 2 1, the AD/MCI group and healthy control group did not differ significantly in age or level of education. Healthy older adults were recruited from the Gainesville community at large. AD /MCI participants were recruited either through the University of Florida Memory and Cognitive Disorders neurology clinic (UF MCD) or the community. Given that the study involved participating in f unctional magnetic resonance imaging (fMRI) individuals with pacemakers, metal implant s, claustrophobia, or other conditions contraindicated for MRI were excluded. All participants were right handed as determined by the Edinburgh Handedness Inventory (Oldfield, 1971) and spoke English as their first language. General medical e xclusionary criteria for participants consisted of history of head trauma, neurological disorder (e.g., stroke), learning disability (e.g., dyslexia), psychiatric disorder (e.g., schizophrenia), drug or alcohol abuse, and any
20 chronic medical condition like ly to impair cognition (e.g., renal or hepatic failure). Medical records reviews were conducted on all AD /MCI participants prior to enrollment in the study. To rule out other causes for cognitive deficits, imaging results (e.g., clinical MRI, thyroid funct ion tests, B12 levels and syphilis test results (e.g., HATTS) were reviewed to rule out other causes of cognitive deficits. Patients recruited from the UF MCD had a consensus diagnosis of AD or multiple domain amnestic MCI. Additionally, for those particip ants recruited outside of the UF MCD, case consensus diagnosis of AD or multi modality amnestic MCI was established through the UF MCD team, with other forms of dementia or medical comorbidities likely to affect a clear cut diagnosis of AD (e.g., mixed dem entia with significant vascular impact) used as exclusionary criteria for the current study. Pre existing neuroradiological imaging results were reviewed by the UF MCD team. Participants with leukoaraisis compromising 50% or more of white matter (Junque sc ale; Junque et al., 1990) were excluded, and structural neuroimaging results collected during the current study were reviewed by the senior neuropsychologist and/or neurologist on the study to verify that participants did not currently meet exclusionary cr iteria with regards to major hemorrhagic or ischemic stroke, lacunar infarct, or leukoaraisis affecting greater than 50% of white matter. Pharmacological exclusionary criteria for healthy control participants included benzodiazapines, antiepileptic, antips ychotic, dopaminergic, and anticholinergic classes of medications, due to their potential effects on cognition. Participants with AD or MCI who were on antiepileptic medication were excluded. Participants with AD or MCI were not excluded if currently medic ally stable on pharmacological agents used to manage behavioral sequelae of AD or MCI (e.g., antidepressants, antipsychotics). Participants
21 with AD or MCI were also not excluded if currently taking medications to manage the cognitive sequelae of dementia ( e.g., Aricept, Namenda), given the current medical cognitive sequelae in AD. All but two participants in the AD/MCI group were taking a fixed dosage of such medication. All participants remained on their regular medication regimen at both study appointments. Informed consent was obtained from participants according to guidelines established by the Health Science Center Institutional Review Board at the University of Flor id a. Participants were paid US $200 for participation. Of note, an additional nine AD/MCI patients were consented but were not used for analyses due to inability to complete neuropsychological testing or the naming task. Procedures Neuropsyc hological T est ing All participants underwent neuropsychological assessment at a separate appointment within 1 to 2 months prior to the fMRI session. The neuropsychological assessment included the following measures: The Mini Mental State Exam (MMSE; Folstein, Folstei n, & McHugh, 1975) was administered to screen for possible dementia and Mild Cognitive Impairment (MCI) in the healthy control group and to document current level of general cognitive functioning in the AD group. Healthy older adults who scored below a 27 were excluded. In order to assess dementia severity, the Clinical Dementia Rating (Morris, 1993) semi str uctured interview was completed. The California Verbal Learning Test, Second Edition (CVLT II; Delis, Kramer, & Kaplan, 2000) was used to assess verba l learning and memory. Potential healthy controls scoring at 2 SD or below on the long delay free recall portion of the test were excluded Conversely, all AD participants were expected to score at least 2 SD below the mean on the long delay free recall an d thus were excluded if a verbal memory deficit was not present.
22 Constructional praxis and visuospatial immediate and delayed recall were evaluated using the Rey Osterrieth Complex Figure Test (Lezak, 1995) The test was scored using the Meyers and Meyers (1995) criteria. The Boston Naming Test (BNT; Kaplan, Goodglass, & Weintraub, 2001) is a 60 name objects. The Verbal Fluency subtest of the Delis Kaplan Executive Function System (D KEFS; Delis, Kaplan, & Kramer, 2001) was administered in order to assess phonological and semantic fluency and semantic fluency switching. During the phonological fluency portion of Verbal Fluency, participants were given 3 letters (i.e., F, A, S) and asked to generate items beginning with each letter. The semantic fluency portion of Verbal Fluency consists of generating items category switching subtest consists of alt ernating between providing items across 2 categories (e.g., fruits, furniture). Phonological Blocking task (PB) used by Faust et al (2004) that has demonstrated significant phonological blocking during picture naming in mild AD patients was administered to quantify lexical impairments in the current study. One hundred pictures were paired with neutral, unrelated, semantic, or phonological primes, all of which were real words. Phonological blocking consists of the difference between errors on picturing na ming for the phonological prime and errors for the unrelated prime. The picture version of the Pyramids and Palm Trees Test (P&PT; Howard & Patterson, 1992) was administered as a test of semantic functions. Participants were presented with a probe picture and 2 pictures from which to choose the one that is more appropriately related to the probe. All participants were additionally assessed f or visual agnosia and apraxia of speech. Participants who showed evidence of either condition were excluded. To further investigate the nature of lexical and semantic functioning in our AD/MCI participants, composite scores were created. The lexical comp osite score was based on performance on the letter fluency subportion of D KEFS and Phonological Blocking. The semantic composite score was based on performance on the animal fluency subportion of D KEFS and Pyramids and Palm Trees. In both cases, composi te scores were created by standardizing raw scores across both groups. The z scores of each
23 lexical measure and each semantic measure were then summed to create the composite scores. fMRI Naming T ask Within 48 hours prior to fMRI, patients and controls un derwent a mock scanning session during which no data was collected. Ten picture naming stimuli not used in the fMRI picture naming task were presented in the mock scanner with simulated scanning noise. The mock scanning session was designed to acclimate su bjects to the scanning environment in order to reduce the effects of novelty on the scanning session. At the MRI scanner, participants completed an overt picture naming task during five functional imaging runs (Figure 2 1) The picture naming task stimuli were the same as those used by Wierenga et al (2008) and consisted of grayscale photographs of 20 animals, 20 vehicles, and 20 tools equated for size and resol ution. Stimuli for each of the three categories were balanced for English frequency (Francis and Kucera, 1982) and familiarity (Coltheart, 1981) Each photograph was presented once for a total of 60 naming trials during the scanning session. Pictures were presented one at a time for eight seconds (four images) each, and participants named the picture aloud. An event related design was implemented to allow for overt responses, in order to a ss ess performance accuracy and response latency. Between overt picture naming trials, participants were instructed to rest and not think of any words while passively view ing abstract patterns. The abstract patterns presented during rest were created by pixelating photographs from the naming task using Adobe PhotoShop 7.0. Intertrial intervals were pseudorandomly varied between 14, 16 and 18 seconds (7, 8, and 9 images respectively) in order to minimize effects of physiological noise and to prevent subsequent overt responses contaminating the hemodynamic response (HDR) by
24 allowing the hemodynamic response to return to baseline before the patient spoke again. Experiment al runs began and ended with a rest interval, and there were 12 naming trials in each experimental run. Each naming fMRI run was 316 seconds in length and 158 functional images were acquired for each slice. Visual stimuli were projected onto a translucent Functional Imaging System (Philips) using E Prime Version 1 software. Overt verbal responses were monitored and recorded using a bidirectional dual microphone system (Commander XG, Resonance Technology, Inc .), noise canceling fiber optic microphone (Phon Or), laptop computer (Dell), and Audacity software (Adobe). Overt responses were scored for accuracy and reaction time off line. fMRI Motor T ask A motor task was used during separate experimental runs fro m the picture naming task to ensure that subjects were capable of generating blood oxygen level dependent ( BOLD ) HDRs. The motor task was not a control task for picture naming. Lack of detectable BOLD HDRs was an exclusionary factor due to some evidence suggesting that BOLD contrast HDR s can be diminished or absent in some older adults. Subjects failing to show a significant (R 2 left hand sensorimotor region during the motor task would have been assumed to have subnormal HDRs and been exclud ed from the study though none failed to show the response The motor task involved pressing the right index figure on a Button Response Unit in synchronization with a visually presented flashing green star. During the baseline task, participants viewed a static red star. Three runs were administered to all participants for a total of 30 activation events. Movement was performed for 2 seconds (1 image) following by variable intervals of rest consisting of 14, 16, or 18 seconds (7, 8,
25 or 9 images, respective ly). The total length of each imaging run was 206 seconds (103 images). Image A cquisition Images were acquired on a Philip 3 Tesla Achieva instrument with a SENSE multiple arrayed head coil. Head motion was minimized using foam padding. Functional images were obtained with a 1 short gradient echo EPI scan (TR=2000ms; TE=30ms; FOV=240mm; matrix size=80 x 80; 3mm x 3mm in plane resolution, flip angle=80 o ). Thirty eight 3mm thick axial slices covering the whole brain were acquired and the total number of images differed depending on which of the two fMRI paradigms was being implemented. A high resolution T1 weighted anatomic scan (TR=8.1ms; TE=3.7ms; FOV=240mm; FA=8 o ; matrix size=240 x 240; 180 X 1.0mm slices) was obtained prior to functional imaging to provide anatomic reference. FLAIR imaging was collected in order to verify that subjects did not meet le ukoaraisis exclusionary criteria for leukoaraisis affecting greater than 50% of white matter ( TR=8000ms; TE=337ms; FOV=256mm; matrix size=256 x 256; 18 0 1.0mm slices). These images were occasionally inspected when participants recruited from the community did not have adequate scans of white matter abnormalities prior to enrollment in the study. Neuroimaging Data A nalyses fMRI data were analyzed and ov erlaid onto structural images with the Analyses of Functional Neuroimaging (AFNI) program from the National Institutes of Health (Cox, 1996) The first seven images of each functional run were discarded to ensure that the spin lattice attain ed a steady state. To minimize the effects of head motion, time series images were spatially registered in three dimensional space to the first image of the first functional image run, since its acquisition immediately followed the acquisition of the
26 anatomic reference scan. Quality control procedures were applied to the data to detect residual motion or susceptibility artifact and distribution of voxelwise coefficient of variation as well as maximum displacement statistics were inspected. Images were addition ally visually inspected for gross artifacts. One subject was asked to be rescanned as a result of these quality control measures. All subjects in the current study demonstrated an adequate HDR in the left sensory motor cortex during the finger tapping task The five naming and three motor imaging runs were detrended for low frequency signal drifts and concatenated into a single time series for each task. To minimize the negative effects of overt speech on the functi onal neuroimaging results, the two images (four seconds) occurring at the time of stimulus presentation for each naming stimulus were removed from each of the naming runs prior to concatenation. Voxels in which the SD of acquired time series exceeded 8% of mean signal (SD/mean signal intensity > 08) were set to zero, to minimize large vessel effects downstream from activity changes or other artifacts. Functional images were co registered with structural images. To compensate for individual differences in structural and functional anatomy, function al images were spatially smoothed (Gaussian filter, 6 mm full width at half maximum) prior to analysis. HDRs for picture naming trials were derived with Area under the curve (AUC) of the deconvolved HDR was the dependent varia ble in the functional analyses. To calculate the AUC statistic, we summed signal intensities at each of the eight TRs comprising the H DR Anatomic and functional AUC images were non linearly warped to 2 x 2 x 2 mm MNI space and co registered using the FMRI B Software Library (FSL) package (Smith et al., 2004; Woolrich et al., 2009).
27 Region of Interest (ROI) A nalyses Three regions of interest (ROIs: inferior temporal lobe [ITL], posterior perisylvian cortex [PPS], and frontal lobes [FL]) were constructed for each hemisphere by combining regions from the Harvard Oxford atlas distributed with FSL (the anterior most portion of frontal polar cortex was eliminated from frontal ROIs but medial frontal and lateral frontal cortex were included ). The blurred and stand ard space transformed AUC images for each participant were masked with these ROIs. Clusters within the ROIs were retained if each voxel was significant at p < .001 and the cluster had a volume of at least 116 This threshold/volume combination was deter mined by Monte Carlo simulation in order to protect ROI wise probability of false positives of at least p < .05. The HDR s of each cluster within an ROI were then examined to ensure that conflicting directionality of the functions would not affect an averag e value for the entire ROI (e.g. a positive BOLD response would not be cancelled out by an equally large negative BOLD response in an adjacent cluster). The HDR s of all clusters within an ROI were next averaged to create one H DR per ROI. We then calculated between the AUC of this average H DR and lexical and semantic composite scores. Voxel Based Morphometry (VBM) A nalyses T1 weighted images DICOM files were converted to .nii gz format. Structural data was analysed with FSL VBM (D ouaud et al., 2007) an optimised VBM protocol (Good et al., 2001) carried out with FSL tools (Smith et al., 2004). First, structural images were skull stripped and grey matter segmented before being registered to the MNI 152 standard space using no n linear registration (Andersson, Jenkinson, & Smith 2007). A visual inspection of all images was completed to control the quality of brain extraction and registration of each image. These images were then averaged and flipped along the
28 x axis to create a left right symmetric, study specific grey matter template. Second, all native grey matter images were non linearly registered to this study specific template and "modulated" to correct for local expansion (or contraction) due to the non linear component o f the spatial transformation. The modulated grey matter images were then smoothed with an isotropic Ga ussian kernel with a sigma of 3 mm (FWHM = 7.05 mm) Finally, voxelwise GLM was applied using permutation based non parametric testing for comparison of g rey matter density between HC and AD/MCI groups (5000 permutations). FWE was used to correct for multiple comparisons across space, and threshold free cluster enhancement was used to assess cluster significance. To determine if areas of group difference wi thin our a priori language related ROIs were associated with task performance, we carried out additional ROI analyses. Average grey matter density for the areas of group difference that fell within our a prior ROIs was calculated for each participant. Thus we created average grey matter density scores for each ROI using the intersection between the significant group differences mask and the ROI masks. grey matter density scores and lexical and semantic composite scores.
29 Table 2 1. D emographics of the healthy control (HC) and Alzheimer's disease/mild cognitive impairment (AD/MCI) groups Diagnosis t value p value (two tailed) HC, n = 12 AD/MCI, n = 10 Mean SD Mean SD Age, year 70.92 4.54 72.60 3.27 1.01 0.33 Education, year 15.54 3.17 16.10 3.25 0.41 0.69 Figure 2 1 Schematic of picture naming task. Participants were presented with greyscale photographs of 20 tools, 20 animals, and 20 vehicles, which they were asked to name out loud. In between naming stimuli, participants were presented with pixelated versions of these same photographs. While these pixelated images we re on the screen, patients were instructed to just relax. 8000 ms 8000 ms 8000 ms 14000 ms 18000 ms
30 CHAPTER 3 R ESULTS Behavioral Results Neuropsychological assessment performance is presented in Table 3 1. Independent samples t tests were conducted to investigate significant differences in cognitive functioning between groups. As expected, the AD/MCI patients performed significantly worse t han healthy older adults on measures of global cognition, MMSE, t (11.037) = 3.55, p < .001; CDR, t (9) = 8.68, p < .001. They also performed significantly worse on measures of verbal learning and memory, CVLT total recall, t ( 20 ) = 8.49, p < .001; CVLT lon g delay free recall, t (14.38) = 11.81, p < .001, visuospatial memory (ROFT, t ( 20) = 5.35, p < .001), and general naming ability, BNT, t ( 11.8 ) = 2.85, p = .02. Contrary to our hypotheses, the AD/MCI patients and healthy older adults were not significantly different on letter fluency, t ( 20 ) = 1.31, p = .21, or phonological blocking, t ( 19 ) = 1.29, p = .21. It follows, then, that they were also not signific antly different in their lexical composite scores, t ( 19 ) = 1.50, p = .15. Finally, differences between the groups were not consistently found on measures of semantic functioning. While the AD/MCI patients performed significantly worse than healthy older a dults on category fluency, ( t ( 20 ) = 3.96, p = .001, there were no significant differences in performance on Pyramids and Palm Trees, t ( 20 ) = 1.41, p = .17. However, AD/MCI patients did have significantly lower semantic composite scores overall, t ( 20 ) = 2.9 6, p = .008. fMRI ROI Analysis Results Average AUC of suprathreshold clusters within a priori ROIs was calculated for each participant, and Spearman correlations were conducted between this measure and lexical and semantic composite scores (Table 3 2). Bo nferroni corrections were applied
31 to account for multiple comparisons Contrary to our hypotheses, neither the lexical nor semantic scores were significantly related to functional activity of their associated language cortices, left PPS r s (19) = .014, p = ns; left ITL r s (20) = .251, p = ns, respectively. Lexical and semantic functioning were similarly not significantly correlated with activity in the left or right frontal lobes, left FL r s (19) = .105, p = ns; right FL r s (19) = .002, p = ns, left FL r s (20) = .47, p = .168; right FL r s (20) = .32, p = .906, for lexical and semantic scores respectively. There was, however, a significant negative correlation between both the lexical and semantic measures and AUC of the right posterior perisylvian corte x, r s (19) = .56, p = .048, r s (20) = .63, p = .012, respectively. Specifically, less functional activity in the right PPS moderately predicted better lexical performance and semantic performance (Figure 3 1). VBM Results VBM: Whole Brain R esults To inv estigate whole brain differences between the AD/MCI patients and healthy older adults voxelwise GLM analyses were conducted (Figure 3 2 ). The analyses revealed the expected density reductions in medial temporal cortex for AD/MCI patients. A dditional bilat eral reductions were detected in regions of the frontal, temporal, parietal, o ccipital, and insular cortices. VBM: ROI A nalyses Further analyses were limited to areas of group difference located within a priori ROIs. Average grey matter density for these r egions was calculated for each participant, semantic composite scores (Table 3 3). Bonferroni corrections were applied to account for multiple comparisons Contrary to our hypo theses, neither the lexical nor semantic
32 scores were significantly related to grey matter density of their associated language cortices, left PPS r s (19) = .078, p = ns; left ITL r s (20) = .097, p = ns, respectively. Lexical and semantic functioning were sim ilarly not significantly correlated with grey matter density in the left or right frontal lobes, left FL r s (19) = .378, p = .55; right FL r s (19) = .17, p =ns, left FL r s (20) = .48, p = .15; right FL r s (20) = .45, p = .204, for lexical and semantic scores r espectively. There was, however, a significant positive correlation between the grey matter density in the right posterior perisylvian cortex and semantic composite score, r s (20) = .577, p = .03. Specifically, higher grey matter density in the right PPS moderately predicted better semantic performance.
33 Table 3 1. Neuropsychological raw/composite scores of the healthy control (HC) and Alzheimer's disease/mild cognitive impa irment (AD/MCI) groups Variables Diagnosis t value p value (two tailed) HC, n = 12 AD, n = 10 Mean SD Mean SD Global Cognition MMSE 29.00 1.21 25.40 3.27 3.30 0.01 a CDR 0.00 0.00 0.75 0.49 4.88 0.00 b Memory CVLT list 1 5 total recall 52.00 8.01 25.40 6.36 8.49 0.00 a CVLT list 1 5 long delay free recall 11.92 3.00 0.90 1.10 11.81 0.00 a RCFT long delay free recall 13.33 4.31 4.15 3.60 5.35 0.00 a General Language Boston Naming Test 55.91 2.91 49.40 6.72 2.85 0.02 a Lexical Functioning Letter fluency (FAS total) 39.75 9.75 34.20 10.06 1.31 0.21 Phonological blocking 0.34 1.82 1.38 1.84 1.29 0.21 c Lexical composite 0.49 1.35 0.53 1.76 1.45 0.15 c Semantic Functioning Category fluency (animals) 21.33 5.14 13.90 3.21 4.13 0.00 a Pyramids and Palm Trees 50.50 1.88 49.40 1.71 1.43 0.17 Semantic composite 0.86 1.76 1.03 1.09 3.09 0.01 a MMSE = Mini Mental State Exam; CDR = Clinical Dementia Rating; CVLT = California Verbal Learning Test; RCFT = Rey Complex Figure Test. a Statistically significant pairwise comparison AD/MCI < HC b Statistically siginificant pairwise comparison AD/MCI > HC c AD/MCI n = 9 Table 3 2. Spearman correlations between functional activity and behavioral language performance Measure PPS ITL FL Left Right Left Right Left Right Lexical composite .01 .56* .52 .24 .11 .002 Semantic composite .12 .63* .25 .14 .47 .32 PPS = Posterior Perisylvian Cortex; ITL = Inferior Temporal Lobe; FL = Frontal Lobe p < .05, FWE corrected
34 Table 3 3. Spearman correlations between grey matter density and behavioral language perofrmance Measure PPS ITL FL Left Right Left Right Left Right Lexical composite .08 .35 .09 .38 .38 .17 Semantic composite .18 .58* .1 0 .36 .48 .45 PPS = Posterior Perisylvian Cortex; ITL = Inferior Temporal Lobe; FL = Frontal Lobe p < .05, FWE corrected
35 Figure 3 1 Hemodynamic response function in the right PPS during the picture naming task is presented for each group. Scatter plots for the relationship between activity and semantic and lexical composite scores are shown for each group with the fitted least squares regression line. AD/MCI is represented in green; HC is re presented in blue. Average AUC Semantic Composite Lexical Composite r s = .63, p < .05 r s = .56, p < .05
36 Figure 3 2 Whole brain comparison of grey matter density between HC and AD/MCI groups. Thresholded and clustered results (protecting a whole brain p < .05) are presented. Warm colors represent areas where AD/MCI patients had signif icantly lower grey matter density than HCs. Results are overlaid on to a study specific average brain that has been non linearly transformed into MNI space and smoothed at 7.05mm FWHM.
37 CHAPTER 4 DISCUSSION This study aimed to investigate the neur al cor relates of naming impairment in hypothesized that not only semantic but also lexical deficits occur during language processing in these patients. Thus, functional brain activity during a naming task and average grey matter density in language related brain regions were correlated with offline measures of lexical and semantic functioning. Contrary to our hypotheses, lexical performance was not significantly related to functional a ctivity or grey matter density in the left posterior perisylvian cortex. Similarly, semantic performance was not significantly related to functional activity or grey matter density in the left inferior temporal lobe. Several significant associations were detected, however, with the right posterior perisylvian cortex. This area was included in analyses as the right hemisphere homologue to the chosen lexical region of interest. Our results indicate that functional activity in this region is negatively relate d to performance on lexical measures; thus, the greater the activation in the lexical right hemisphere homologue, the poorer the lexical functioning: the AD/MCI pat ients did not have significantly lower scores in this domain than the healthy older adults. In this sense, the negative association may not reflect deficits that arise from the disease process; instead, it may be related to changes in language processing t hat are associated with the aging process in general. Activity in the right posterior perisylvian cortex was also significantly negatively related to semantic functioning, while grey matter density in this area was significantly positively related to
38 seman tic functioning. Together these results suggest that both increased activity and increased atrophy of this region are associated with poor semantic performance. While unexpected, this association between the right posterior perisylvian cortex and several different measures of language functioning may be due to participant characteristics. As noted earlier, no lexical deficits were detected among our AD/MCI patients, which may be due in part to our inclusion criteria. We chose to include multi modality amne stic MCI patients, as these cases show higher conversion rates to impairments), as such p atients may be diagnosed with primary progressive aphasia. Croot et al. (2000), for example, reported that 3 of the 10 case studies included in t he i r description of atypical AD presentations were classified as progressive aphasia until postmortem analyses were performed. To interpret our results, then, we look outside the AD literature and into the realm of aging and language research in general. Our chosen correlational analyses lend themselves well to this type of interpretation in that our participants s pan the continuum of healthy to pathological aging. This assertion is especially apt in light of recent research regarding increased amyloid burden among a subset of cognitively normal older adults (Codispoti et al., 2012; Kikuchi et al., 2011; Sperling et al., 2009) These studies suggest that while up to 30% of health older adults may have amyloid deposition they do not show impaired functioning. Amyloid deposition may, however, impact functional brain activity, such that high amyloid burden in cognitively normal older adults
39 is associated with aberrant neural activity in networks that support memory functions (Sperling et al., 2009). In this sense, it is likely that som e cognitively normal older adults in our healthy control group demonstrated aberrant neural activity while successfully completing neuropsychological measures. Our analyses, then, allow us to ask if we are detecting a phenomenon of healthy aging that is ag gravated by the effects of neuropathology rather than an effect of the pathology itself. Recent studies of the hemispheric asymmetry reduction in older adults (HAROLD) phenomenon lend support to this idea. Meinzer and colleagues (2012) examined functiona l brain activity during semantic and phonemic fluency in healthy younger and older adults. They specifically looked at increased task positive activity as well as decreased task negative activity in older adults to investigate their relationship to languag e performance. Of note, they looked at both frontal and more posterior brain regions, including regions of interest of the current study. Overall, Meinzer and colleagues found that older adults performed better when their brain activity more closely resemb led that of younger adults. In other words, increased task positive activity was associated with poorer semantic performance and decreased task negative activity was associated with poorer semantic performance. Relevant to the findings of this study, they found a cluster of decreased negative BOLD in the right inferior parietal lobule, and this cluster fell within the right posterior perisylvian ROI we investigated. Adjacent to this cluster, Meinzer and colleagues found an area of increased positive BOLD in the right superior parietal lobule amongst older adults. These results resemble our findings in that increased positive BOLD in the right posterior perisylvian cortex
40 during a language task was associated with poorer semantic performance amongst healthy o lder adults. Zlatar and colleagues (in press) examined possible neural mechanisms underlying these differences in brain activity between healthy younger and older adults during a semantic fluency task. They looked at activation in young adults, sedentary older adults, and active older adults and foun d brain patterns very similar to those reported by Meinzer et al. (2012) for sedentary older adults. Older adults performed better when their activation more closely resembled that of younger adults. Like Meinzer et al. (2012), they found BOLD response pat terns in the right posterior perisylvian cortex that corresponded to those detected in the current study. While younger adults showed negative BOLD responses, both groups of older adults had task positive activity in this area. Furthermore, the increased p ositive activity was associated with decreased semantic functioning. Zlatar and colleagues related this increased BOLD response to decreased inhibition by examining the ipsilateral silent period (iSP, a measure of interhemispheric inhibition obtained durin g transcranial magnetic stimulation). They found that increased activity in the right PPS was negatively associated with the iSP, suggesting that the positive BOLD detected is the result of decreased suppression of the region. In this sense, the right post erior perisylvian cortex was active during a task in which it is usually suppressed, and this activation interfered with efficient semantic functioning in sedentary older adults. It is important to note that in both of the studies described the task that elicited increased activation of the right posterior perisylvian cortex was semantic fluency. We utilized a different task of language functioning, picture naming, to investigate the neural
41 correlates of anomia. However, our correlational analyses did not rely on naming accuracy; rather they used a measure of functioning that included semantic fluency. Thus, while our investigation does not align completely with those previously mentioned, the overlap in processes examined allows us to cautiously extend th eir findings. Indeed, these studies have interesting implications for the interpretation of our results. They suggest that the increased task positive activity we found in right PPS may be a part of the aging process. Similar activity in previous studies h as been linked to decreased interhemispheric inhibition, which begs the question: is this activity related to the loss of the fact that this activity is associated wit h poorer semantic and lexical functioning, which makes it unlikely that activation of the right posterior perisylvian cortex is compensatory. Rather, the increased activation of this region may be interfering with efficient language functioning, or it may be an indication of decreased efficiency in processing throughout the brain. Thus, we may have detected a process of healthy disease process itself. Research involving the de fault mode network (DMN) functioning and beta amyloid deposition have shown similar phenomena occur in the attention network. While younger adults typically show strong task negative activity within the DMN, healthy older adults show weaker task negative a ctivity and those with a high amyloid burden show task positive activity (Sperling et al., 2009). Moreover, while default mode connectivity is disrupted in normal aging, high amyloid burden is associated with further decreased connectivity that can be dete cted before clinical signs of AD are evident (Hedden et al., 2009)
42 In summary, the expected relationship between semantic and lexical deficits in AD patients and their associated neurocorrelates was not detected in the current study. Rather, we found tha t increased activation of the right posterior perisylvian cortex was associated with poorer semantic and lexical functioning. Increased atrophy of the same region was also associated with semantic deficits. In this sense, the right PPS may be sensitive to functional changes of the aging process, such as a loss of suppression, this phenomenon, future studies should address the relationship between the right PPS, cognition, and amyloid burden. A healthy younger adult group should additionally be included. Further investigation of this trend opens the door for new discussion of treatment of language deficits due to the healthy aging processes.
43 LIST OF REFERENCES Andersson, M., Jenkison, M., and Smith, S. (2007). Non linear registration, aka Spatial normalisation. FMRIB technical report. Backman, L., Andersson, J. L., Nyberg, L., Winblad, B., Nordberg, A., & Almkvist, O. (1999). Brain regions associated with episodic retrieval in normal aging and Alzheimer's disease. Neurology, 52 (9), 1861 1870. Becker, J. T., Mintun, M. A., Aleva, K., Wi seman, M. B., Nichols, T., & DeKosky, S. T. (1996). Compensatory reallocation of brain resources supporting verbal episodic memory in Alzheimer's disease. Neurology, 46 (3), 692 700. Binder, J. R., Desai, R. H., Graves, W. W., & Conant, L. L. (2009). Where Is the Semantic System? A Critical Review and Meta Analysis of 120 Functional Neuroimaging Studies. Cerebral Cortex, 19 (12), 2767 2796. Bookheimer, S. Y., Strojwas, M. H., Cohen, M. S., Saunders, A. M., Pericak Vance, M. A., Mazziotta, J. C., & Small, G. W. (2000). Patterns of brain activation in people at risk for Alzheimer's disease. N Engl J Med, 343 (7), 450 456. Braak, H., & Braak, E. (1991). Neuropathological stageing of Alzheimer related changes Acta Neuropathol, 82 (4), 239 259. Chao, L. L., Haxby, J. V., & Martin, A. (1999). Attribute based neural substrates in temporal cortex for perceiving and knowing about objects. Nature Neuroscience, 2 (10), 913 919. Codispoti, K. E., Beason Held, L. L., Kraut, M. A., O'Brien, R. J., Rudow, G., Pletnikova, O., Resnick, S. M. (2012). Longitudinal brain activity changes in asymptomatic Alzheimer disease. Brain Behav, 2 (3), 221 230. Coltheart, M. (1981). The MRC Psycholinguistic Database. Quarterly Journal of Experimental Psychology, 33A 497 505. Cox, R. W. (1996). AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res, 29 (3), 162 173. Croot, K., Hodges, J. R., Xuereb, J., & Patterson, K. (2000). Phonological and articulatory impairment in Alzheimer's disease: a case series. Brain Lang, 75 (2), 277 309. Delis, D. C., Kaplan, E., & Kramer, J.H. (2001). Delis Kaplan Executive Function System (D KEFS). San Antonio, TX: Pearson Assessment. Delis, D. C., Kramer, J.H., Kaplan, E., & Ober, B.A. (2001). Califo rnia Verbal Learning Test. Second Edition. San Antonio, TX: Pearson Assessment.
44 Douaud, G., Smith, S., Jenkinson, M., Behrens, T., Johansen Berg, H., Vickers, J., James, A. (2007). Anatomically related grey and white matter abnormalities in adolescen t onset schizophrenia. Brain, 130 (9), 2375 2386. Dubois, B., Feldman, H. H., Jacova, C., Cummings, J. L., Dekosky, S. T., Barberger Gateau, P., Scheltens, P. (2010). Revising the definition of Alzheimer's disease: a new lexicon. Lancet Neurol, 9 (11) 1118 1127. Ellis, A., & Young, A. (1988). Human Cognitive Neuropsychology Hillsdale, NJ: Lawrence Erbaum. Faust, M. E., Balota, D. A., & Multhaup, K. S. (2004). Phonological blocking during picture naming in dementia of the Alzheimer type. Neuropsychol ogy, 18 (3), 526 536. Feinberg, T. E., Schindler, R. J., Ochoa, E., Kwan, P. C., & Farah, M. J. (1994). Associative Visual Agnosia and Alexia without Prosopagnosia. Cortex, 30 (3), 395 411. 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. J Psychiatr Res, 12 (3), 189 198. Foster, N. L., Chase, T. N., Mansi, L., Brooks, R., Fedio, P., Patronas, N. J., & Di Chiro, G. ( 1984). Cortical abnormalities in Alzheimer's disease. Ann Neurol, 16 (6), 649 654. Foundas, A. L., Daniels, S. K., & Vasterling, J. J. (1998). Anomia: Case studies with lesion localization. Neurocase, 4 (1), 35 43. Francis, W., & Kucera, H. (1982). Frequen cy Analysis of English Usage: Lexicon and Grammar Boston: Houghton Mifflin Co. Galton, C. J., Patterson, K., Graham, K., Lambon Ralph, M. A., Williams, G., Antoun, N., Hodges, J. R. (2001). Differing patterns of temporal atrophy in Alzheimer's disea se and semantic dementia. Neurology, 57 (2), 216 225. 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, 123 484 498. Gee, J., Ding, L., Xie, Z., Lin, M., DeVita, C., & Grossman, M. (2003). disease and frontotemporal dementia exhibit distinct atrophy behavior correlates. Academic Radiology, 10 (12), 1392 1401.
45 Gold, B. T., Balota, D. A., Jones, S. J., Powell, D. K., Smith, C. D., & Andersen, A. H. (2006). Dissociation of automatic and strategic lexical semantics: Functional magnetic resonance imaging evidence for differing roles of multiple frontotemporal regions. Journal of Neuroscience, 26 (24), 6523 6532. Good, C. D., Johnsrude, I. S., Ashburner, J., Henson, R. N., Friston, K. J ., & Frackowiak, R. S. (2001). A voxel based morphometric study of ageing in 465 normal adult human brains. Neuroimage, 14 (1 Pt 1), 21 36. Grossman, M., Koenig, P., DeVita, C., Glosser, G., Moore, P., Gee, J., Alsop, D. (2003). Neural basis for verb processing in Alzheimer's disease: an fMRI study. Neuropsychology, 17 (4), 658 674. Grossman, M., Koenig, P., Glosser, G., DeVita, C., Moore, P., Rhee, J., Gee, J. (2003). Neural basis for semantic memory difficulty in Alzh eimer's disease: an fMRI st udy. Brain, 126 (2), 292 311. Grossman, M., McMillan, C., Moore, P., Ding, L. J., Glosser, G., Work, M., & Gee, J. (2004). What's in a name: voxel based morphometric analyses of MRI and naming difficulty in Alzheimer's disease, frontotemporal dementia and corticobasal degeneration. Brain, 127 628 649. Hedden, T., Van Dijk, K. R., Becker, J. A., Mehta, A., Sperling, R. A., Johnson, K. A., & Buckner, R. L. (2009). Disruption of functional connectivity in clinically normal older adults harboring amyloid burd en. J Neurosci, 29 (40), 12686 12694. Hodges, J. R., Salmon, D. P., & Butters, N. (1991). The Nature of the Naming Deficit in Alzheimers and Huntingtons Disease. Brain, 114 1547 1558. Howard, R., & Patterson, K. (1992). Pyramids and Palm Trees. Suffolk, England: Thames Valley Test Company. Johnson, S. C., Saykin, A. J., Baxter, L. C., Flashman, L. A., Santulli, R. B., McAllister, T. W., & Mamourian, A. C. (2000). The relationship between fMRI activation and cerebral atrophy: comparison of normal aging and alzheimer disease. Neuroimage, 11 (3), 179 187. Junque, C., Pujol, J., Vendrell, P., Bruna, O., Jodar, M., Ribas, J. C., Marti Vilalta, J. L. (1990). Leuko araiosis on magnetic resonance imaging and speed of mental processing. Arch Neurol, 47 (2), 15 1 156. Kaplan, E., Goodglass, H., & Weintraub, S. (2001). Boston Naming Test. Second Edition Austin, TX: PRO ED, Inc. Kikuchi, M., Hirosawa, T., Yokokura, M., Yagi, S., Mori, N., Yoshikawa, E., Ouchi, Y. (2011). Effects of brain amyloid deposition and reduced glucose metabolism on the default mode of brain function in normal aging. J Neurosci, 31 (31), 11193 11199.
46 Lezak, M. (1995). Neuropsychological Assessment New York: Oxford University Press. Meinzer, M., Seeds, L., Flaisch, T., Harnish, S., Co hen, M. L., McGregor, K., Crosson, B. (2012). Impact of changed positive and negative task related brain activity on word retrieval in aging. Neurobiol Aging, 33 (4), 656 669. Meyers, J. E., & Meyers, K.R. (1995). Rey Complex Figure and Recognition T rial: Professional manual Odessa, FL: Psychological Assessment Resources. Moreaud, O., David, D., Charnallet, A., & Pellat, J. (2001). Are semantic errors actually semantic?: Evidence from alzheimer's disease. Brain Lang, 77 (2), 176 186. Morris, J. C. (1 993). The Clinical Dementia Rating (CDR): current version and scoring rules Neurology, 43 (11), 2412 2414. Nadeau, S. E. (2000). Phonology. In S. E. Nadeau, Rothi, L. J. G., & Crosson, B. (Ed.), Aphasia & Language: Theory to Practice. (pp. 40 81). Newy Yo rk: Guilford Press. Ojemann, G. A. (1983). Brain Organization for Language from the Perspective of Electrical Stimulation Mapping. Behavioral and Brain Sciences, 6 (2), 189 206. Oldfield, R. C. (1971). The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia, 9 (1), 97 113. Park, D. C., & Reuter Lorenz, P. (2009). The adaptive brain: aging and neurocognitive scaffolding. Annu Rev Psychol, 60 173 196. Raymer, A. M., Moberg, P., Crosson, B., Nadeau, S., & Rothi, L. J. G. (1997). Lexical semantic deficits in two patients with dominant thalamic infarction. Neuropsychologia, 35 (2), 211 219. Salmon, D. P., Butters, N., & Chan, A. S. (1999). The deterioration of semantic memory in Alzheimer's disease. Canadian Journal of Experimental Psychology Revue Canadienne De Psychologie Experimentale, 53 (1), 108 117. Salmon, D. P., Heindel, W. C., & Lange, K. L. (1999). Differential decline in word generation from phonemic and semantic categories during the course of Alzheimer's disease: Implications for the integrity of semantic memory. Journal of the International Neuropsychological Society, 5 (7), 692 703. Saykin, A. J., Flashman, L. A., Frutiger, S. A., Johnson, S. C., Mamourian, A. C., Moritz, C. H., Weaver, J. B. (1999). Neuroanatomic substrates of semantic memory impairment in Alzheimer's disease: Patterns of functional MRI activation. Journal of the International Neurops ychological Society, 5 (5), 377 392.
47 Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E., Johansen Berg, H., Matthews, P. M. (2004). Advances in functional and structural MR image analysis and implementation as FSL. Neuroima ge, 23 Suppl 1 S208 219. Sperling, R. A., Laviolette, P. S., O'Keefe, K., O'Brien, J., Rentz, D. M., Pihlajamaki, M., Johnson, K. A. (2009). Amyloid deposition is associated with impaired default network function in older persons without dementia. Neuron, 63 (2), 178 188. Tabert, M. H., Manly, J. J., Liu, X., Pelton, G. H., Rosenblum, S., Jacobs, M., Devanand, D. P. (2006). Neuropsychological prediction of conversion to Alzheimer disease in patients with mild cognitive impairment. Arch Gen Psy chiatry, 63 (8), 916 924. Wagner, A. D., Pare Blagoev, E. J., Clark, J., & Poldrack, R. A. (2001). Recovering meaning: left prefrontal cortex guides controlled semantic retrieval. Neuron, 31 (2), 329 338. Wierenga, C. E., Benjamin, M., Gopinath, K., Perlste in, W. M., Leonard, C. M., Rothi, L. J., Crosson, B. (2008). Age related changes in word retrieval: role of bilateral frontal and subcortical networks. Neurobiol Aging, 29 (3), 436 451. Wierenga, C. E., Stricker, N. H., McCauley, A., Simmons, A., Jak A. J., Chang, Y. L., Bondi, M. W. (2011). Altered brain response for semantic knowledge in Alzheimer's disease. Neuropsychologia, 49 (3), 392 404. Wilcock, G. K., & Esiri, M. M. (1982). Plaques, tangles and dementia. A quantitative study J Neurol Sci, 56 (2 3), 343 356. Woolrich, M. W., Jbabdi, S., Patenaude, B., Chappell, M., Makni, S., Behrens, T., Smith, S. M. (2009). Bayesian analysis of neuroimaging data in FSL. Neuroimage, 45 (1 Suppl), S173 186. Zlatar, Z. Z., Towler, S., McGregor, K.M ., Dzierzewski, J.M., Bauer, A., Phan, S., Cohen, M., Marsiske, M., Manini, T.D., & Crosson, B. (in press). Functional language networks in sedentary and physically active older adults. Journal of International Neuropsychological Society
48 BIOGRAPHICAL SKETCH Amanda Garcia graduated in 2011 from the University of Florida with a Bachelor of Science in p sychology and a Bachelor of Arts in l inguistics and Spanish. In 2011, she began a Doctor of Philosophy program in the Department of Clinical a nd Health Psychology. Her research interests include semantic memory and dementia.