Citation
Age-Related Changes in Word Retrieval: Frontal-Executive vs. Temporal-Semantic Substrates

Material Information

Title:
Age-Related Changes in Word Retrieval: Frontal-Executive vs. Temporal-Semantic Substrates
Creator:
WIERENGA, CHRISTINA ELIZABETH ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Business executives ( jstor )
Hemispheres ( jstor )
Hemodynamic responses ( jstor )
Magnetic resonance imaging ( jstor )
Memory ( jstor )
Memory retrieval ( jstor )
Older adults ( jstor )
Prefrontal cortex ( jstor )
Semantics ( jstor )
Words ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Christina Elizabeth Wierenga. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
2/28/2006

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

AGE-RELATED CHANGES IN WORD RETRIEVAL: FRONTAL-EXECUTIVE VS. TEMPORAL-SEMANTIC SUBSTRATES By CHRISTINA ELIZABETH WIERENGA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2004 by Christina Elizabeth Wierenga

PAGE 3

This dissertation is dedicated to my parents, Edward and Wilma Wierenga.

PAGE 4

ACKNOWLEDGMENTS I would like to acknowledge my family and friends, as well as my dissertation committee and labmates, for their steady guidance and support. I would especially like to acknowledge Dr. Bruce Crosson for providing remarkable mentorship throughout my doctoral training. I extend sincere gratitude to Michelle Benjamin, Timothy Conway, Allison Cato, Kaundinya Gopinath and Keith White for their valuable contributions. This work was supported by the Evelyn F. & William L. McKnight Brain Institute and the Department of Veteran Affairs Office of Academic Affiliations and the Rehabilitation Research and Development Service Pre-Doctoral Associated Health Rehabilitation Research Fellowship. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.......................................................................................................................xi CHAPTER 1 INTRODUCTION........................................................................................................1 2 COGNITIVE AGING...................................................................................................9 Neuropsychological Changes in Aging........................................................................9 Age-related Changes in Executive Functions of Language................................10 Age-related Changes in Semantic Memory.........................................................12 Neuroanatomical and Vascular Changes in Aging.....................................................15 Functional Neuroimaging of Cognitive Aging...........................................................20 3 SEMANTIC PROCESSING......................................................................................28 Semantic Representation............................................................................................28 The Semantic System..........................................................................................28 Models of Semantic Representation....................................................................29 Evidence for the role of category.................................................................32 Evidence for the role of attributes................................................................36 Evidence for the role of modality.................................................................41 The Matrix Theory..............................................................................................44 Semantic Processing in the Ventral Visual Stream....................................................46 Global and Local Features...................................................................................48 Retrieval of Semantic Knowledge..............................................................................50 4 HYPOTHESES...........................................................................................................55 Hypothesis 1...............................................................................................................55 v

PAGE 6

Hypothesis 2...............................................................................................................56 Hypothesis 3...............................................................................................................56 Hypothesis 3a......................................................................................................58 Hypothesis 3b......................................................................................................58 Hypothesis 3c......................................................................................................58 Hypothesis 3d......................................................................................................58 5 PILOT STUDY...........................................................................................................59 Participants.................................................................................................................59 Experimental Design and Task...................................................................................59 Stimuli.................................................................................................................60 Design and Task..................................................................................................60 Results.........................................................................................................................61 6 METHODS.................................................................................................................64 Participants.................................................................................................................64 Experimental Design and Tasks.................................................................................65 Dementia Screening.............................................................................................65 Assessment of Language.....................................................................................66 FMRI Naming Task.............................................................................................68 FMRI Motor Task................................................................................................71 Image Acquisition.......................................................................................................72 Image Analysis...........................................................................................................73 Group Analyses of the Naming Task..................................................................75 Behavioral Analyses...................................................................................................75 7 RESULTS...................................................................................................................77 Behavioral Results......................................................................................................77 Naming Performance during FMRI.....................................................................77 Neuropsychological Testing Performance..........................................................78 FMRI Results..............................................................................................................80 Comparison of Age on Picture Naming..............................................................82 Comparison of the Interaction between Age and Semantic Category.................88 Direct Comparisons of Semantic Category.........................................................89 8 DISCUSSION...........................................................................................................101 Frontal-Executive Changes in Word Retrieval with Age.........................................101 Temporal Substrates of Semantic Representation....................................................107 Conclusion................................................................................................................112 vi

PAGE 7

APPENDIX A ADDITIONAL HRF ANALYSES FOR MAIN EFFECT OF AGE IN FRONTAL CORTICES...............................................................................................................114 B EXPLORATORY ANALYSIS FOR MAIN EFFECT OF CATEGORY...............117 LIST OF REFERENCES.................................................................................................119 BIOGRAPHICAL SKETCH...........................................................................................141 vii

PAGE 8

LIST OF TABLES Table page 6-1. Demographic Variables for Younger and Older Adults.............................................65 7-1. Performance on the FMRI Naming Task for Younger and Older Adults..................78 7-2. Raw Scores on the Neuropsychological Tests for Younger and Older Adults..........80 7-3. A Priori Volumes of Tissue in Medial and Lateral Frontal Cortices Showing Significant Activity during Naming for Older Compared to Younger Adults.........84 7-4. A Posteriori Volumes of Tissue Showing Significant Activity during Naming for Older Compared to Younger Adults........................................................................87 7-5. A Priori Volumes of Tissue in the Inferior Temporal Cortex Showing Differential Responses to Animals, Tools, and Vehicles for Pairwise Comparisons..................91 7-6. A Posteriori Volumes of Tissue Showing Differential Responses to Animals, Tools, and Vehicles..................................................................................................97 7-7. A Posteriori Volumes of Tissue in the Lateral Temporal Lobe (>200 l) Showing Differential Responses to Animals, Tools, and Vehicles for Pairwise Comparisons.............................................................................................................99 B-1. A Posteriori Volumes of Tissue Showing Differential Responses to Animals, Tools, and Vehicles for Main Effect of Category..................................................117 viii

PAGE 9

LIST OF FIGURES Figure page 3-1. Depiction of Ellis and Young’s (1988) model of lexical and semantic functions......29 3-2. Depiction of the matrix theory....................................................................................45 3-3. The ventral visual stream of the macaque..................................................................47 5-1. Schemata of an ascending sequence of 9 filtering levels of a stimulus in each category for the pilot study.......................................................................................61 5-2. Average threshold level of spatial filtering for identification of animals, tools, and vehicles.....................................................................................................................62 5-3. Percentage of correct responses for animals, tools, and vehicles at each level of spatial filtering..........................................................................................................62 6-1. Sample portion of a pseudorandomized event-related run alternating picture naming with passive viewing of pixilated images...................................................70 7-1. Regions of the a) medial (rostral cingulate zone) and b) lateral (Broca’s homologue in the right hemisphere) frontal cortex activated by older adults relative to younger adults during picture naming, along with corresponding hemodynamic response functions............................................................................86 7-2. Activity in left frontal cortex, including Broca’s area, for older adults and younger adults during picture naming compared to baseline..................................88 7-3. Lateral and medial fusiform gyrus activated by the comparison of animals and tools, along with corresponding hemodynamic response functions.........................92 7-4. Activity in the fusiform gyrus bilaterally for the comparison of vehicles to tools, along with corresponding hemodynamic response functions..................................93 7-5. Activity in the fusiform gyrus bilaterally for the comparison of vehicles to animals, along with corresponding hemodynamic response functions....................94 7-6. Activity in the a) right inferior frontal sulcus, b) left inferior parietal lobe and leftt middle temporal gyrus associated with the main effect of category across participants....................................................................................................97 ix

PAGE 10

7-7. Left and right middle temporal gyrus activated for the comparison of animals to tools......................................................................................................................99 A-1. Region of the bilateral supplementary motor area activated by older adults relative to younger adults during picture naming, along with corresponding hemodynamic response functions..........................................................................114 A-2. Region of the right inferior frontal gyrus activated by older adults relative to younger adults during picture naming, along with corresponding hemodynamic response functions..........................................................................115 A-3. Region of the right anterior cingulate gyrus activated by older adults relative to younger adults during picture naming, along with corresponding hemodynamic response functions..........................................................................115 x

PAGE 11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AGE-RELATED CHANGES IN WORD RETRIEVAL: FRONTAL-EXECUTIVE VS TEMPORAL-SEMANTIC SUBSTRATES By Christina Elizabeth Wierenga August 2005 Chair: Bruce Crosson Major Department: Clinical and Health Psychology People over the age of 65 frequently complain of word-finding difficulties. The current study tests two competing hypotheses to determine whether word retrieval problems in older adults result from changes in neural substrates in the frontal lobe subserving strategic processes to access lexical and semantic information or whether neural substrates of semantic information in the inferior temporal lobe change in function with normal aging. We investigate the role of category (living, nonliving) and visual attribute (global form, local details) in semantic representation in the fusiform gyrus. Forty adults (20 younger, 20 older) named pictures of animals, tools, and vehicles during FMRI. Results show that compared to younger adults, older adults activate a larger frontal network with decreased lateralization during word retrieval. In particular, older adults show greater activity in Broca’s area homologue (BA 45) in the right hemisphere, an anterior region of the right inferior frontal gyrus, and the rostral cingulate zone and supplementary motor area bilaterally. The lack of an age-associated difference in the xi

PAGE 12

inferior temporal cortex indicates that older and younger adults do not differ in terms of how the presumed substrates for semantic functions operate. Results also support the role of visual attribute in the organization of semantic information. Collapsed across subjects, findings indicate that categories are processed in the lateral and medial regions of the fusiform gyrus according to whether they are living (animals) or nonliving (tools, vehicles), respectively. In contrast, visual attributes of global form (animals) are processed more by the right fusiform gyrus and local details (tools) are processed more by the left fusiform gyrus. When both attributes are relevant to processing (vehicles), cortex from both left and right fusiform gyri is active. The left-right distinction for visual features is not restricted to the fusiform gyrus but includes the lateral temporal cortex as well. Taken together, these findings suggest that a deterioration of executive functions underlies age-related changes in word retrieval. Additionally, results support the role of features (visual attribute) and category in semantic representation. xii

PAGE 13

CHAPTER 1 INTRODUCTION People over the age of 65 frequently complain of word-finding difficulties, in which they are unable to remember the words they wish to use to describe objects, actions, and concepts, and may render them unable to communicate efficiently (Kempler & Zelinski, 1994). Such complaints have been objectively verified on tests of naming visually presented objects, such as the Boston Naming Test (BNT: Kaplan et al., 1983). Between the ages of 60 and 80, raw scores on the BNT may drop by a standard deviation or more (Ivnik et al., 1996; Mitrushina & Satz, 1995; Van Gorp et al., 1986). Older adults also tend to be slower and less accurate on tasks involving naming to definition (Bowles & Poon, 1985) and verbal fluency (McCrae et al., 1987). Despite the prevalence of word-finding problems in older adults and the resulting impairment in effective communication, little systematic research has been conducted to examine the cause of word retrieval difficulties. Identification of the underlying mechanisms of word retrieval problems in aging is critical to understanding the nature of cognitive aging and possible contributors to cognitive decline. In addition, elucidation of the cause of word-finding problems is vital for the development of effective interventions to promote efficient communication in healthy older adults as well as individuals with more severe language impairments resulting from stroke or Alzheimer’s disease. Word retrieval requires access to memories for word forms (i.e., lexical forms) and meaning (i.e., semantics). There are two alternative explanations for word-finding difficulties in normal aging. First, word-finding problems may be caused by an inability 1

PAGE 14

2 to focus attention and select an appropriate word. This may result from deterioration of the neural substrates that underlie the ability to strategically access and retrieve existing lexical or semantic information. Such mechanisms involved in word retrieval fall under the category of executive functions. Generally, the term “executive functions” has been used to refer to a set of cognitive processes that include planning, initiating, and implementing strategies for behavior, monitoring performance, using feedback to adjust future behavior, and inhibiting task irrelevant information and prepotent responses (Alexander et al., 1989; Bryan & Luszcz, 2000; Lezak, 1995; Spreen & Strauss, 1998; Stuss & Alexander, 2000; Stuss & Levine, 2002). Executive functions involved in language include operations such as searching semantic stores, monitoring and evaluating potential words to represent a concept, and exerting cognitive control to select the appropriate word from among competing alternatives (West, 1996). These executive functions have been localized to the frontal lobe structures. The occurrence of language disorders following lesions to the left frontal lobe has implicated the left frontal operculum and regions of the dorsolateral convexity in word-finding, grammatical usage, and comprehension (Alexander et al., 1989). More specifically, the neural substrates of executive functions in language involve medial and lateral frontal structures. Medial frontal structures, especially those at the border of pre-SMA and the rostral cingulate zone, are involved in initiation of language, cognitive control and monitoring conflict between competing responses (Barch et al., 2000; Carter et al., 2000; Crosson et al., 1999). Damage to this region may result in akinetic mutism (Damasio & Anderson, 1993). The lateral frontal cortex of the language dominant hemisphere, including Broca’s area, cortex along the inferior frontal sulcus, and possibly

PAGE 15

3 pars orbitalis is involved in selection, retrieval and execution of the lexical-semantic response appropriate to internal and external constraints (Barch et al., 2000; Crosson et al., 1999; Gabrieli et al., 1998; MacDonald et al., 2000; Thompson-Schill et al., 1997; Wagner et al., 2001). Age-related decreased performance on word-finding tasks may be due to a deterioration of executive functions resulting in an inability to access words from semantic memory stores. Chapter 2 discusses age-related changes in cognition and underlying neurophysiology that may contribute to declines in word retrieval. Thus, a difference in level or extent of activity for older vs. younger persons in either medial or lateral frontal structures would indicate a change in the neural substrates for executive functions. A medial location would suggest difficulty in intentional substrates for language or in monitoring conflict between competing responses. A lateral frontal location would suggest difficulty in response selection or execution, diminished ability to formulate search strategies, or increased semantic processing (Crosson et al., 2003). Given findings in functional imaging research with aging, it is likely that changes in activity with aging would involve increased activity and reduced lateralization thought to reflect compensation for reduced efficiency of functioning (Cabeza, 2001; Grady, 2000; Grady et al., 1994; Madden et al., 1997). A second explanation for word-finding difficulties is that the semantic system involved in representing concepts may deteriorate to the point where the partial information that is available is insufficient to activate a word on occasion. The semantic system attaches meaning to the lexical representation of words (Raymer & Rothi, 2000). This information includes knowledge of semantic categories and attributes, such as visual features or functions of objects. The structure of the semantic system has been inferred

PAGE 16

4 from patterns of category-specific language impairments resulting from brain injury. Category-specific and semantic deficits are characterized by losses that are specific for a subset of words in an individual's language. The most commonly reported category-specific naming impairment is the dissociation between living and nonliving objects, marked by an inability to name members of either a superordinate living or nonliving category (Damasio et al., 1996; De Renzi & Lucchelli, 1994; Hillis & Caramazza, 1991; Warrington & Shallice, 1984). Other category-specific naming impairments include deficits in naming animals (Barry & McHattie, 1995; Laws et al., 1995), tools (Hillis et al., 1990; Sacchett & Humphreys, 1992), fruits and vegetables (Farah & Wallace, 1992), body parts (Goodglass & Wingfield, 1993), and medical items and instruments (Crosson et al., 1997). Case studies and functional neuroimaging studies have implicated regions of the temporal lobe in semantic representation. Many neuroimaging studies suggest that semantic processing occurs in a distributed network of topographically organized processors, i.e., category domain or type of attribute (Chao et al., 1999; Ishai et al., 1999; Vandenberghe et al., 1996). Vandenberghe et al. (1996) identified a distributed multimodal semantic network (shared by pictures and words) that extends from the left superior occipital gyrus through the middle and inferior temporal cortex to the inferior frontal gyrus. An integration of case-study and neuroimaging findings suggests that semantic information is stored in a complex matrix involving the modality in which information is processed (i.e., visual vs. verbal), attributes relevant to identifying objects (e.g., shape, color, emotional connotation), and the semantic category of objects (e.g., living vs. nonliving) (Crosson et al., 2000; Hart & Gordon, 1992; Warrington & Shallice, 1984). Each of these semantic

PAGE 17

5 characteristics is thought to have specific neural substrates that can be identified from unique patterns of activity with functional imaging. Although there are many competing theories of semantic organization to account for discrepant findings in both functional imaging and lesion studies, there are relatively consistent findings for the involvement of the fusiform gyrus, which runs along the inferior surface of the temporal lobe, in semantic processing of visual attributes and category domain (e.g., Chao et al., 1999; Ishai et al., 1999; Thompson-Schill et al., 1999). Previous neuroimaging studies (Chao et al., 1999; Ishai et al., 1999; Thompson-Schill et al., 1999) have shown that the lateral and medial areas of the fusiform gyrus are differentially activated during processing of animate and inanimate objects, respectively. Living objects activate the lateral fusiform gyrus whereas nonliving objects activate the medial fusiform gyrus. Differences in location of activity in the fusiform gyrus may be due to either categorical organization of visual semantic knowledge or to physical attributes. Recently, Kraut et al. (2002) provided support for the role of attributes independent of category in semantic organization and identified a common region of activity for fruits and tools in the premotor cortex (BA 6/44). Since fruits are manipulable, they share a similar characteristic with tools even though they belong to a different category domain (living vs. nonliving). Thus, Kraut et al. (2002) concluded that this area of the premotor cortex may be responsible for the representation of attribute knowledge of object use regardless of object category. The distinction between the role of attributes and category in semantic organization is of particular interest to the current study and may provide a means to detect subtle changes in semantic representation, if they exist, in older adults that may contribute to

PAGE 18

6 word-finding problems. A preliminary study in our laboratory with younger adults provided evidence for a double dissociation between processing of visual-semantic features in the posterior portion of the fusiform gyrus and processing of categories in the anterior portion of the fusiform gyrus (Wierenga et al., 2001). It was based upon the concept that vehicles are members of the superordinate category of nonliving things but share a visual feature with living things, specifically, reliance on global form for identification. Global visual attributes refer to the basic shape of an object (e.g., an object's outline); local visual attributes refer to the fine details of an object's visual features that comprise the larger configuration. In contrast, tools, like many other nonliving objects, are assumed to be distinguished from one another visually by their local details rather than their global visual features. For instance, a shovel is differentiated from a rake by the nature of the extension attached to its handle. Global features are processed in the right hemisphere and local visual features are processed in the left hemisphere in the ventral visual stream (Delis et al., 1992; Doyen & Milner, 1991; Fink et al., 1997). In our study, both animals and vehicles activated an area in the right posterior fusiform gyrus thought to represent processing global form while implements activated an area in the left posterior fusiform gyrus that subserves analysis of local features. However, in the anterior fusiform gyri, animals activated a lateral area because they are living things while implements and vehicles activated a medial area because they are nonliving things. Although these findings replicated the findings of Chao et al. (1999) and Ishai et al. (1999), further research is needed to reliably support the differentiation of visual attributes and category in semantic representation.

PAGE 19

7 If semantic substrates in the inferior temporal lobe become less efficient with aging, then during confrontation naming, we might expect to see either a breakdown in lateralization of processing for global and local features in the left and right fusiform gyrus and/or diffuse activity within the medial and lateral fusiform gyrus without respect to whether items are living or nonliving things. Differences in semantic organization between younger and older adults in terms of how visual attributes and categories are processed would support the claim that decreased word-retrieval in older adults results from deterioration of semantic information. To build a rationale for the hypotheses regarding unique brain activity related to semantic information, a review of semantic processing will be provided in Chapter 3. Chapter 4 describes the hypotheses for the current study and provides a rationale for each hypothesis. Chapter 5 describes a pilot study that was conducted to empirically verify that information about global form is sufficient to identify both animals and vehicles whereas local details are needed to identify tools. Spatial frequency content is a primary perceptual factor for recognizing visual objects; filtering spatial frequency of photographs is a technique that manipulates the amount of detail in an object while preserving global form. Pictures were presented in a way that removes most local detail by filtering high spatial frequency content, and then detail was gradually added until normal subjects could identify the picture. As expected, animals could be identified with the least amount of local detail, tools required the most amount of local detail, and vehicles were intermediate in the amount of local detail needed for identification (Wierenga et al., 2003). Results of this pilot study informed our selection of stimuli for the FMRI study to experimentally manipulate the categories so that animals and vehicles

PAGE 20

8 were similar in amount of global visual information needed for identification. Chapter 6 describes the methodology used in the current study, Chapter 7 presents the results, and Chapter 8 provides a discussion of the findings. The purpose of the current study is two-fold. First, the study initiates an inquiry into the differences in neural substrates of word retrieval in healthy young and older adults using functional magnetic resonance imaging (FMRI) to determine the contribution of frontal-executive vs. temporal-semantic substrates of word retrieval. Specifically, the study tests two competing hypotheses to ascertain whether neural substrates subserving strategic processes to access lexical and semantic information change with normal aging or whether neural substrates of semantic information change in function with normal aging. The second aim of the study is to inform and constrain models of semantic representation in the inferior temporal lobe, specifically to assist in more precisely understanding the role of category and visual attribute in semantic organization. Category and attribute will be examined by using items from which a major attribute matches that of items from another category. Specifically, pictures of animals, tools, and vehicles will be presented for naming. As mentioned, vehicles are identified by the attribute of global form like animals, but vehicles are a member of the category of nonliving things, like tools. In contrast, tools, like many other nonliving objects, are assumed to be distinguished from one another visually by their local details rather than their global visual features. Results from the Kraut et al. (2002) study suggest that this type of dissociation between attributes and category is possible, although it has not been successfully demonstrated in the inferior temporal lobe.

PAGE 21

CHAPTER 2 COGNITIVE AGING Neuropsychological Changes in Aging The effects of healthy aging on cognition are not well understood. One challenge to the study of cognitive changes in aging is the ability to discriminate the effects of normal aging from pathological processes that disrupt cognition, especially since normal aging is associated with changes in the neural correlates of cognition. Therefore, it has become increasingly important to relate cognitive changes in aging to their underlying structural and functional neural substrates. Normal aging disproportionately affects certain cognitive functions while other cognitive functions remain intact late into life. For example, executive control, strategic planning and speed of processing have been shown to decline more rapidly with age than other functions (Park, 1996; Salthouse, 1996). Working memory (the manipulation and maintenance of information on-line) and episodic memory (the ability to encode new memories of facts) also tend to decline across the lifespan from age 20 to 80, whereas semantic memory performance is less vulnerable to decline with aging (Balota et al., 2000; Mayr & Kliegl, 2000; Park et al., 2002; Schaie, 1996). Of note, a disorder of the semantic memory system is considered a hallmark sign of Alzheimer’s disease, although the nature of this semantic disorder is under debate (see Milberg, 1999). Short-term memory and vocabulary remain relatively stable until around age 70, after which they show sharper declines (Hedden and Gabrieli, 2004; Park et al., 2002; Schaie, 1996). Autobiographical memory, emotional processing and implicit memory appear to remain 9

PAGE 22

10 relatively unchanged throughout life (Hedden & Gabrieli, 2004; La Voie & Light, 1994). Of particular interest to the current study is the impact of life-long changes in executive functions underlying the ability to access and manipulate information and the apparent relative preservation of semantic memory on word retrieval. Age-related Changes in Executive Functions of Language As noted, aging appears to differentially affect executive functioning abilities (Park 1998; Rubin, 1999). Since executive functions underlie the ability to retrieve information from memory, declines in naming and verbal fluency performance with age may be attributed to difficulty accessing lexical information from memory stores rather than to a deterioration of long-term lexical-semantic memories themselves (Kempler & Zelinski, 1994). However, changes in executive functions of language with aging have not been directly investigated. Rather, performance on tasks of confrontation naming and verbal fluency has been interpreted to reflect changes in underlying executive functions. For example, poor performance on the Boston Naming Test (BNT) may signify problems with selection and retrieval of lexical-semantic information. Older adults consistently perform worse than younger adults on the BNT. To illustrate, a score attaining the 50th percentile for an 80 year-old would be ranked at or below the 16th percentile when compared to a 60 year-old person (Ivnik et al., 1996). Although this difference in scores is statistically significant, it is relatively modest. The scores differ by only approximately 5 of a possible 60 correct responses, even though this represents a difference of two standard deviations. However, individuals with 12 years of schooling or more tend to perform better on confrontation naming, regardless of their age (Neils et al., 1995; Welch et al., 1996).

PAGE 23

11 Increased variability in performance on the BNT has also been reported with advancing age (Van Gorp et al., 1986). Errors most often involve no response, circumlocution or semantic errors (Tombaugh & Hubley, 1997). An increase in semantic errors on the BNT in a group of individuals in their seventies, including circumlocutions, semantically related associates, and nominalizations (word that describes the function of an object) suggests a discrepancy between the ability of older individuals to access a lexical representation and their knowledge of the semantic or conceptual representation involved in word retrieval (Albert et al., 1988). In contrast, other studies of confrontation naming support the preservation of executive functions involved in word retrieval in aging. For example, there was no interaction between age and lexical properties (including familiarity, number of letters, frequency of occurrence, and number of syllables of word representations) of the items on the BNT, suggesting that the process of lexical access was similar in younger and older adults (Moberg et al., 2000). Previous research has revealed a moderate but reliable age-related decline in performance on tests of semantic fluency, which involve active search and retrieval of members of a specified semantic category as quickly as possible (Bckman & Nilsson, 1996; Kempler et al., 1998; Mayr & Kliegl, 2000; McCrae et al., 1987; Troyer et al., 1997). For example, a score for semantic fluency attaining the 50th percentile for an 80 year-old would be ranked at or below the 16th percentile when compared to a person under 69 (Lucas et al., 1998). Relatively constant age differences on elements of verbal fluency tests such as retrieval position and response latency suggest that decreased performance results from impaired executive control processes, such as initiating a semantic search and selecting an appropriate response (Mayr & Kliegl, 2000). The use of

PAGE 24

12 switching and clustering strategies has been assessed to differentiate between semantic and executive functions involved in semantic fluency (Troyer et al., 1997). Switching involves finding a new category and relies on effective search processes supported by frontal structures, whereas clustering involves producing consecutive words within the same semantic or phonemic category and is thought to depend on an individual’s store of available words that are maintained by temporal structures (Stuss & Alexander, 2000). Older adults tend to switch less frequently, which contributes to decreased performance and suggests less effective search strategies due to aging. Arguably, although standard neuropsychological tests of language provide valuable qualitative and quantitative information about behavioral changes in aging, they are limited in their ability to elucidate the neural correlates of executive functions subserving language production and to distinguish between frontal and temporal neural systems underlying word retrieval deficits in aging. Furthermore, since neuropsychological tests of executive functions were developed to be sensitive to frontal lobe damage rather than to assess theoretical concepts of executive functions, their ability to differentiate between components of executive functions or between executive and “nonexecutive” functions is questionable (Bryan & Luszcz, 2000). The debate between the changing role of executive vs. semantic functions in word retrieval in aging is difficult to resolve through evaluation of behavioral performance alone without respect to the neuroanatomical correlates of these functions. Age-related Changes in Semantic Memory Although verbal and language-based knowledge is generally well-maintained into old age, subtle changes in the organization of semantic memory with aging may contribute to word finding difficulties (Eustache et al., 1998). However, the semantic

PAGE 25

13 system has not been adequately studied in older adults. Although the organization of semantic memory and potential age-related changes are difficult to assess with traditional neuropsychological or cognitive tests, attempts have been made to infer semantic functioning from behavioral performance. Qualitative analysis of the verbal responses in semantic fluency tasks reveals that older and younger adults demonstrate differences in selection and frequency of the most common responses to semantic categories, which may suggest different compositions of semantic categories in older and younger adults (Brosseau & Cohen, 1996). In contrast, Brown and Mitchell (1991) found no age difference in retrieving dominant responses for a high-constraint task (picture naming) and a low-constraint task (category exemplar generation), but older adults produced less consistent responses than younger adults with broadly defined categories. Although these studies provide some support for age-related differences in semantic processing, it is difficult to determine whether these differences result from changes due to aging or whether they merely represent cohort effects. An alternative approach to examine age-related changes in semantic organization is the use of priming tasks that indirectly probe semantic function and may be more sensitive than traditional neuropsychological tests to the role of semantic representation in age-related changes in word retrieval. Priming results in the facilitation of a response (e.g., lexical decision or picture naming) when the target is preceded by a semantically related stimulus or prime. The prime is assumed to activate its own representation in memory and that activation “spreads” and activates representations of related items. Therefore, an age difference in semantic priming raises the possibility of changes in the structural representation of lexical-semantic information. However, converging evidence

PAGE 26

14 suggests equivalent semantic priming effects in younger and older adults. The lack of age-related changes in priming has been interpreted as evidence against the loss of semantic knowledge (Bowles, 1994; Kempler & Zelinski, 1994). In addition, when priming effects are reported for older adults, the larger priming effect is often due to a greater age difference in reaction times to unrelated primes than in reaction times to related primes (Laver & Burke, 1993; Myerson et al., 1997). In summary, although word-finding problems in aging are well-documented, whether they are due to changes in the neural substrates subserving the ability to access semantic stores or a deterioration in the neural representation of semantic memory itself is difficult to resolve with behavioral performance on language tasks alone, since these tasks do not elucidate the changes in neural substrates underlying word retrieval. Despite inconsistent findings, behavioral performance on various language tasks provides clues into the nature of word retrieval problems. Although decreased accuracy on these tasks has predominantly been attributed to declines in executive functioning, there is some evidence to suggest a deterioration of semantic processing may impact performance. Yet, these behavioral studies are limited by the ability of these assessment measures, such as visual picture naming and semantic fluency, to yield pure reflections of semantic processing or executive functioning, since both processes are inherently implicated to some degree in the performance of these tasks. Clearly, the dissociation of executive and semantic processes in age-related changes in word retrieval has not been adequately addressed. These inconsistent findings highlight the need for a more rigorous examination of the underlying mechanisms responsible for word retrieval difficulties in aging to address the potential dissociation between the process of access and

PAGE 27

15 representation in word retrieval. Functional neuroimaging techniques may contribute to the resolution of this debate by relating behavioral performance to changes in functioning of the neural substrates underlying executive functions and semantic processes involved in word retrieval in normal aging. Neuroanatomical and Vascular Changes in Aging Aging is associated with notable alterations in brain structure and function. Compared to younger adults, the brains of older adults tend to have lower volumes of gray matter (Haug & Eggers, 1991; Mesulam, 2000; Raz, 2000; Resnick et al., 2003). The decreased brain volume results from lower synaptic densities in older adults, rather than from a loss of neurons (Terry, 2000). However, changes in brain volume are not uniform across regions. The prefrontal cortex and medial temporal structures are particularly affected by either normal or pathological aging, whereas the occipital cortex remains relatively unaffected by age (Raz et al., 2004; West, 1996). Older brains also demonstrate expansion of the cerebral ventricles, gyral atrophy, enlargement of cerebral sulci, and a global increase in cerebrospinal fluid (Good et al., 2001; Guttmann et al., 1998; Mesulam, 2000; Raz, 2000; Resnick et al., 2003). It remains unclear whether gray matter (cell bodies and synapses in the cortex and subcortical nuclei) and white matter (myelinated axons) are differentially sensitive to the effects of aging (Raz, 2000). There are relatively consistent findings that global gray matter volume decreases linearly with age. Yet, regional differences in gray matter volume changes are under dispute. Although accelerated loss of volume in the prefrontal cortex (dorsolateral and orbitofrontal), inferior temporal cortex, insula, superior parietal gyri, central sulci and cingulate sulci is generally agreed upon, there are conflicting reports of decline in the hippocampus and entorhinal cortex (Good et al., 2001; Raz et al.,

PAGE 28

16 1997; Resnick et al., 2003). However, at least one study reports an absence of a substantial decrease in gray matter (Guttman et al., 1998). Similarly, studies examining white matter changes in aging yield widely inconsistent results. Despite several reports that global white matter volumes do not decline with age, there is evidence of a loss of white matter density and an increase in white matter lesions as a result of age (Guttman et al., 1998). White matter changes are most pronounced in the prefrontal cortex and the anterior corpus collosum, although age-related decline in white matter integrity and increase in white matter hyperintensities is widespread and may represent demyelinating changes in older adults (Bartzokis et al., 2003; Head et al., 2004; Raz, 2000; Resnick et al., 2003). Several white matter pathways, including frontal white matter tracts, optic radiations, and posterior limbs of the internal capsule also exhibit a relative accelerated loss of volume (Good et al., 2001). White matter abnormalities have been associated with poor performance on tasks of executive functions, processing speed and immediate and delayed memory (Gunning-Dixon & Raz, 2000). Taken together, age-related changes in the gray and white matter of the cortex may mediate behavioral patterns of cognitive aging in healthy adults (Hedden & Gabrieli, 2004), despite some of the inconsistent findings, which are likely due to different morphometric techniques used across studies and the reciprocity between gray matter and white matter that makes it difficult to distinguish increases in one from decreases in the other. The prefrontal cortex is particularly vulnerable to the effects of aging, as demonstrated by a more pronounced reduction in volume due to shrinkage of cortical neurons and decreased synaptic density and dendritic arborization (Raz, 2000). Structures of the prefrontal cortex experience an estimated average volume decline of

PAGE 29

17 about 5% per decade after the age of 20, although it is not known whether atrophy is due to changes in white matter or gray matter (Raz et al., 2004; Raz et al., 1998; Resnick, 2003; Tisserand et al., 2002). However, the frontal lobes are structurally and functionally very heterogeneous, and subregions are differentially affected by age. Specifically, the lateral regions of the prefrontal cortex undergo the largest declines in healthy older adults whereas the anterior cingulate, frontal pole and precentral gyrus remain relatively unchanged (Raz et al., 2004; Salat et al., 2001; Tisserand et al., 2002). In addition to increased frontal atrophy with age, regions in which the frontal lobes have dense reciprocal projections such as the thalamus and striatum demonstrate significant atrophy (Raz, 2000; Rubin, 1999; Van der Werf et al., 2001). In fact, changes in the frontostriatal system are thought to underlie normal aging and account for the differential declines in the volume and function of the prefrontal cortex (Gabrieli 1996; Hedden & Gabrieli, 2004; Rubin, 1999). Changes in the striatum tend to be smaller and decline by about 3% per decade (Gunning-Dixon et al., 1998). The caudate has been found to decrease by 15% from age 25 to 75 (Rubin, 1999). The volume of the thalamus also decreases with age, and is correlated with slower cognitive processing speed (Van der Werf et al., 2001). In line with these volumetric changes, neurotransmitters in the prefrontal cortex and striatum also experience age-related changes. Dopamine concentration, transporter availability, and dopamine D2 receptor density decline with age, which are associated with lower glucose metabolism in the frontal cortex and hypometabolism in the anterior cingulate cortex, temporal cortex, and caudate nucleus (Goldman-Rakic & Brown, 1981; Volkow et al., 1996; Volkow et al., 2000).

PAGE 30

18 In contrast to the relatively large age-related changes in the prefrontal cortex, striatum, and frontal white matter tracts, temporal structures undergo minimal age-related changes in healthy aging (Raz et al., 2004). For instance, the degree of volume reduction in the temporal lobes for adults older than seventy has been estimated to be only about 1% (West, 1996). However, the volume of the hippocampus and the parahippocampal gyrus declines by 2-3% per decade, although dendritic growth continues until the 9th decade in the hippocampal CA subregions and the parahippocampal gyrus (Rapp et al., 2002; Rasmussen et al., 1996; Raz et al., 2004; West, 1993). These declines in volume are primarily unrelated to cognitive function over the lifespan, although hippocampal volume tends to predict explicit memory performance after the age of 60 (Raz et al., 1998; Rosen et al., 2003). Given the relative preservation of the temporal cortex in healthy aging, this region may be a more sensitive predictor of pathological cerebral changes in late life and changes in the volume of the temporal lobes may be predictive of the onset of dementia (Raz, 2000). Specifically, it has been suggested that a reduction of the volume of the fusiform gyrus may be a better predictor of cognitive decline than changes in hippocampal volume (Raz, 2000). Changes in cerebrovascular dynamics have also been observed in aging. This has important implications for the study of aging using FMRI, since FMRI relies on the hemodynamic properties of the vasculature surrounding regions of neuronal activity (to be discussed later in this chapter). Medical conditions such as hypertension, diabetes, and high cholesterol are relatively common in older adults and are the main risk factors for cerebrovascular disease. Even in healthy aging, arteriosclerotic changes, such as fibrohyaline thickening of the vessel wall (Furuta et al., 1991), necrosis of smooth muscle

PAGE 31

19 cells (Masawa et al., 1994) and thickening of the basement membrane (Nagasawa et al., 1979) gradually increase with age and compromise the integrity of the cerebral vasculature (D’Esposito et al., 2003). These changes may decrease the elasticity and compliancy of affected vessels, including capillaries, arterioles and cerebral arteries and eventually occlude cerebral vessels (D’Esposito et al., 2003). Sixty-five percent of adults above the age of 60 have venous alterations, which in severe cases can completely occlude veins (Moody et al., 1997). The tortuosity, or winding and coiling, of some vessels also increases with age, most notably in the arteriole-venous-capillary bed (Fang et al., 1976). Since blood flow in the capillary bed significantly contributes to the blood oxygenation level-dependent (BOLD) signal (Menon et al., 1995), these age-associated differences in vasculature may produce age-related differences in the BOLD FMRI signal. Partially occluded vessels may result in chronic ischemia, a condition of insufficient oxygen delivery tissue (D’Esposito et al., 2003). Sustained dilation of vessels distal to partially occluded vessels may compensate for cerebral ischemia and provide adequate tissue perfusion, but may impede the additional blood flow increase that is responsible for the BOLD signal examined in FMRI (Daut et al., 1990; D’Esposito et al., 2003). A significant decrease in the resting cerebral blood flow (rCBF) in the cortical and subcortical parenchyma as well as in large cerebral arteries is also associated with aging (Bentourkia et al., 2000; Kawamura et al., 1993; Krejza et al., 1999). Similarly, a decrease in the resting regional cerebral metabolic rate of oxygen utilization (rCMRO2) has also been documented in aging (Madden & Hoffman, 1997), although the impact of this on age-related changes in activity-induced CMRO2 have not been examined

PAGE 32

20 (D’Esposito et al., 2003). For example, selective reduction in rCBF has been found in the temporal and prefrontal cortices (West, 1996). In older adults, a pattern of hypofrontality, characterized by decreased blood flow in the anterior as compared with posterior regions has been observed, in contrast to a pattern of hyperfrontality commonly seen in younger adults (West, 1996). The differences in resting CBF between younger and older adults may have serious implications for FMRI studies that make comparisons between an active task and a resting baseline, since the BOLD signal is a relative measure of signal change between two conditions. Although such declines have been found in rCBF during resting states, age-related differences in the pattern or magnitude of activity during cognitive activity has not consistently been demonstrated (Gur et al., 1987). Despite clear age-associated anatomical and vascular changes, it has yet to be determined whether or to what extent these structural and physiological changes result in reduced functional integrity of the normal aging brain demonstrated by changes in behavior and cognition. Functional Neuroimaging of Cognitive Aging Functional magnetic resonance imaging (FMRI) indirectly measures neural functioning through the blood oxygenation level-dependent (BOLD) signal. The BOLD signal depends on neurovascular coupling, the process by which neural activity influences the hemodynamic properties of the surrounding vasculature (Buckner et al., 2000; D’Esposito et al., 2003). The BOLD response derives from a local increase in blood oxygen content due to changes in blood flow and volume resulting from increased neural activity. Specifically, local blood flow increases following a neural event. The increase in blood flow and volume overshoots the tissue’s increased demand for oxygen. This results in a decrease in the concentration of deoxygenated hemoglobin in the

PAGE 33

21 microvasculature surrounding the activated region and leads to an increase in the BOLD signal. The BOLD signal reflects the ratio of non-paramagnetic oxygenated hemoglobin to paramagnetic deoxygenated hemoglobin, and is altered by neural activity that influences several factors, including cerebral blood flow (CBF), cerebral blood volume, and cerebral blood oxygen consumption (D’Esposito et al., 2003). Therefore, the BOLD mechanism is dependent on the integrity of the local vasculature (Buckner et al., 2000). Basic changes in the brain’s physiology with aging, such as changes in cerebral cytoarchitecture, neurochemistry, vasculature, and cerebral hemodynamics represent potential confounds to the investigation of aging and cognition using functional neuroimaging, since any alteration in the cerebrovascular dynamics could affect neurovascular coupling. Since the BOLD response may have different properties in young and older adults, the ability to differentiate between changes in neurovascular coupling and changes in the underlying neural activity is a challenge to interpreting age-related changes in FMRI, especially because direct comparisons of groups rely on the assumption of comparable neurovascular coupling. The ability to correlate changes in BOLD signal with task performance increases the likelihood that group differences represent true correlates of changes in neural activity rather than physiological alterations. Given the likelihood of vascular compromise that may affect neurovascular coupling in older adults, characteristics of the BOLD signal may differ between younger and older adults. Studies that have compared hemodynamic response properties in younger and older adults have yielded conflicting results. For instance, there are reports of similar hemodynamic response amplitudes and shapes between young and old adults in

PAGE 34

22 the primary motor cortex (Buckner et al., 2000; D’Esposito et al., 1999). Despite this similarity, D’Esposito et al. (1999) also reported increased variability and a reduced signal-to-noise ratio in healthy older adults, and found that younger adults activated four times as many voxels as their older counterparts. However, a substantial portion of the elderly subjects did not exhibit detectable BOLD responses during the finger tapping task, which suggests a weakened or nonexistent coupling between neural activity and FMRI signal in these subjects. This study is noteworthy because D’Esposito et al. (1999) chose a simple motor task in which associated neural activity is not expected to differ between younger and older adults, thus strengthening their ability to discern age-related changes in neurovascular coupling. In addition, Taoka et al. (1998) report an age-related lag in time for the BOLD signal to reach half its maximum level in the precentral gyrus for a blocked design hand-grasping task, which may be due to decreased local vascular reactivity. In contrast, reduced signal amplitude (percent signal change) has been found in the visual cortex of older adults during FMRI of photic stimulation (Buckner et al., 2000; Ross et al., 1997), although Ross et al. (1997) did not find a change in spatial extent of activity as measured by the number of pixels exceeding threshold. Alternatively, Huettel et al. (2001) found a decrease in spatial extent and increased noise levels in the visual cortex in older adults evoked by viewing checkerboard stimuli despite similar amplitudes of the BOLD signal. A decrease in signal amplitude and number of activated voxels was also found in the sensorimotor cortex associated with age during a finger-tapping task (Hesselman et al., 2001). Since histological studies have not revealed significant neuronal loss in the primary motor cortex in normal aging (Haug, 1997; Haug & Eggers,

PAGE 35

23 1991), functional differences in this region more likely reflect a disturbance of neurovascular coupling. Furthermore, within-subject age-associated regional differences in the BOLD signal have been found between the visual and motor cortex elicited by viewing a flickering checkerboard and pressing a key; BOLD signal amplitude was decreased in the visual cortex, whereas there was no change in BOLD signal amplitude in the motor cortex (Buckner et al., 2000). Buckner et al. (2000) attempted to resolve these signal differences between visual and motor areas by suggesting that there may be regional variation in the coupling of neuronal activity to hemodynamic response. Whether regional variations also occur in the neural substrates subserving tasks with greater cognitive demands, such as word retrieval, is yet to be determined. Despite the potential confound of age-associated changes in neurocoupling with age-related changes in neural activity, the comparison of patterns of activity during cognitive tasks in older and younger adults can provide valuable information relating cognitive changes to their neural substrates, especially when behavioral performance is known. Functional neuroimaging studies have shown that older and younger people may use different brain areas on cognitive tasks, even when performing at the same level of proficiency (Grady, 2000). Several functional neuroimaging studies report differences in the extent of activity for older adults when compared to younger adults. Of particular interest to the current study are age-associated activity changes in executive functions and semantic memory. Older adults tend to exhibit different patterns of prefrontal cortex activity than younger adults during tasks involving executive functions. However, investigations of age-related activity changes during executive processing tasks report contradictory results. While some studies report that older adults demonstrate less

PAGE 36

24 activity in the prefrontal cortex during executive functioning (Logan et al., 2002; Rypma & D’Esposito, 2000; Stebbins et al., 2002), other studies indicate that differences exist between the regions activated within the prefrontal cortex in younger and older adults (Cabeza et al., 2002; Grady et al., 1998). In fact, older adults sometimes show increased activity in regions not activated by younger adults, often in areas contralateral to those activated by younger adults (Cabeza et al., 2002; Grady et al., 1998; Logan et al., 2002; Rypma & D’Esposito, 2000; Stebbins et al., 2002). Increased recruitment of the prefrontal cortex may indicate an increased need for executive functions in older adults, (such as monitoring of responses, alternating between tasks, or resolving conflict between competing goals) or a decreased suppression of competing responses, even during tasks not specifically designed to elicit executive control. For example, evidence suggests that older adults recruit medial and dorsolateral frontal areas subserving executive functions involved in task-switching even when such demands are not fundamental to the task (DiGiralamo et al., 2001). In contrast, few studies have examined age-related changes in semantic memory using functional neuroimaging techniques. Madden et al. (2002) report minimal age-related change in activity for retrieval of semantic information during a lexical decision task. Instead, older and younger adults demonstrated a similar pattern of activity in the occipitotemporal and inferior prefrontal regions of the left hemisphere (Madden et al., 2002). Of note, to our knowledge there have not been any studies to date that have investigated age-related changes in naming using FMRI. Because the relation between neural activity and cognitive performance is complex, interpreting age-related changes in patterns of activity can be challenging. Increased activity in older adults may reflect a variety of processing mechanisms, including

PAGE 37

25 decreased efficiency, compensatory functioning, use of alternative strategies, or reduced suppression or recruitment of irrelevant neural resources. For instance, higher activity levels may be necessary for older adults to maintain the same level of performance as younger adults. Rypma et al. (2001) found increases in frontal lobe activity were associated with decreases in reaction time in younger adults but with increases in reaction time in older adults during a working memory task, which they interpret as evidence that age-related declines might affect the neural correlates of processing efficiency. Older adults may use different functional networks to compensate for reductions in efficiency in task-related brain regions. Alternatively, there are reports that additional activity in the prefrontal cortex for older adults is often seen only in high-performing older adults (Cabeza, 2002; Cabeza et al., 2002; Reuter-Lorenz et al., 2000; Reuter-Lorenz et al., 2002). For example, in a strategic encoding task, high-performing older adults exhibited bilateral PFC activity, whereas younger adults and low-performing seniors showed only left-sided PFC activity (Rosen et al., 2002). In addition, age-related activity changes may be modulated by strategy choice, and older adults may adopt different strategies to compensate for declines in cognitive ability or neural deficits (Stebbins et al., 2002). However, additional activity in older adults does not always correlate with equivalent or better performance, but may sometimes represent non-selective recruitment of irrelevant or competing brain regions (Logan et al., 2002) accompanying poor performance. Such non-selective recruitment could occur if a break-down in inhibitory connections accompanies aging (Head et al., 2004). Alternatively, a decrease in activity may reflect more efficient processing or decreased processing (Cabeza, 2001).

PAGE 38

26 A relatively consistent finding in functional neuroimaging studies of cognitive aging is that prefrontal cortex activity tends to be less lateralized in older adults than younger adults in a variety of cognitive domains, including episodic memory, semantic memory retrieval, working memory, perception and inhibitory control (Logan et al., 2002; Madden et al., 1997; Reuter-Lorenz et al 2000; Stebbins et al., 2002). Cabeza (2001) conceptualizes this decreased laterality in terms of a model called Hemispheric Asymmetry Reduction in Old Adults (HAROLD). Two hypotheses have been proposed to account for decreased laterality. According to a compensation hypothesis in old age, bihemispheric involvement emerges to help counteract age-related neurocognitive decline, whereas according to a dedifferentiation hypothesis, bihemispheric activity reflects difficulty in recruiting specialized neural mechanisms. Preliminary studies of highand lowperforming older adults on memory tasks suggest that increased bilateral activity reflects compensatory mechanisms, since low-performing adults did not activate bilateral prefrontal cortex regions seen in high-performing older adults (Cabeza et al., 2002; Rosen et al., 2002). In summary, results from functional neuroimaging studies of cognitive aging suggest that comparisons of the neural mechanisms underlying age-related changes in cognitive activities are possible. Although there are several challenges in this area of research, potential methodological techniques have been suggested to aid in the interpretability of obtained results by reducing the confounds of age and performance. Event-related FMRI paradigms have the advantage of monitoring performance and distinguishing between successful and unsuccessful trials and allow for these trials to be analyzed separately so that activity associated with successful trials can be attributed to

PAGE 39

27 mechanisms responsible for successful performance (Cabeza, 2001). In addition, incorporation of a control task, such as a motor task, in which age-related neural activity is not thought to differ can be used to assess the BOLD response and allows for the elimination of participants who demonstrate an atypical or decreased hemodynamic response likely attributed to compromised vasculature associated with aging. Studies that demonstrate changes in activity in the prefrontal cortex and the ventral visual pathway suggest that the comparisons in the current study are possible. However, there is much to learn about changes in hemodynamic response properties with age and how to interpret age-related changes in functional activity during cognitive tasks.

PAGE 40

CHAPTER 3 SEMANTIC PROCESSING Semantic Representation The Semantic System Word retrieval involves semantic memory, which Tulving (1972) defined as a long-term memory store for context-free knowledge about language, facts and common concepts. The semantic system represents meaning and stores knowledge of concepts in the brain. It is the mechanism that attaches meaning to the sensory and productive experiences of spoken and written words and viewed objects (Raymer & Rothi, 2000). Ellis and Young (1988) proposed a model in which the semantic system receives sensory input from two modalities; it receives auditory input in the form of phonological input (heard words) or sounds, and visual input in the form of orthographic input (written words) or structural object representations (Figure 3-1). The auditory and visual processing systems are responsible for recognizing words, objects, and sounds as familiar but do not provide information about meaning. Rather, they allow access to meaning in the semantic system. The semantic system functions to attach meaning to the auditory and visual input and enables meaningful spoken or written output. Other theorists have elaborated on the model of the semantic system proposed by Ellis and Young (1988) to include other sensory inputs, such as viewed gestures and actions, olfactory information and tactile information, as well as productive outputs such as action or pantomime (Raymer & Rothi, 2000). The utility of Ellis and Young's model of the semantic system is its ability to explain many language disorders in terms of damage to specific input or 28

PAGE 41

29 output systems and differentiate these disorders from semantic language disorders. However, this model is incomplete because it does not provide an account of the underlying organization of information in the semantic system. Figure 3-1. Depiction of Ellis and Young’s (1988) model of lexical and semantic functions. Models of Semantic Representation There are two predominant models of semantic representation: modular and distributed. Modular models assume that the brain consists of anatomically distinct modules or cognitive processors which function independently to produce mental states (Fodor, 1983). Modular models of semantic organization posit that knowledge is compartmentalized in the brain in independent but interacting modules, and assume a one-to-one correspondence between semantic knowledge and underlying brain regions responsible for this information. Ellis and Young’s (1988) model of the semantic system represents a modular model. Support for modular models comes from semantic deficits

PAGE 42

30 that appear to respect the boundaries of category or attribute knowledge or modality of processing. An advantage of modular models is that they tend to map well onto neuroanatomical regions. However, since they assume that processes are discrete and processing occurs in a sequential, unidirectional fashion, they do not represent an efficient system; instead modular models presume replication of information. For example, separate lexicons (phonological, orthographic) are suggested for the input and output processes of the semantic system as depicted by Ellis and Young (1988). In contrast, distributed models of semantic representation do not assume divisions of the semantic system but conceptualize it as unified and widespread throughout the brain. Distributed models are guided by computational theories including spread of activation or parallel distributed processing (PDP) models, which maintain that semantic information is contained within the connections between different nodes (which represent concepts) and place less emphasis on neuroanatomical localization (McClelland & Rumelhart, 1985). The relative distance or weighting of connections between nodes represents the relative relationship between concepts. Information processing is efficient and spreads iteratively in a bi-directional manner with feedforward and feedbackward loops. Several computational models suggest that the semantic system is comprised of a network of attributes, similar to a neural network, with the properties of parallel processing and distributed representation, and semantic categories are an emergent property of the interaction of attributes (Caramazza et al., 1990; Caramazza & Shelton, 1998; Farah & McClelland, 1991; Hillis & Rapp, 1995; Rapp et al., 1993). Attributes are modeled as units that are interconnected with every attribute in the network. The strength of connections between attributes differs depending on the likelihood of occurring

PAGE 43

31 together (Farah and McClelland, 1991). Semantic representation results from a spreading pattern of activation within this network and semantic information is located in the connections between attributes. Therefore, category-specific deficits are seen as artifactual or due to lost attributes following an alteration of the dynamics of the semantic network, rather than loss of a semantic category. Although the computational theories of distributed models may reflect neuronal processing, they are limited in their ability to describe the underlying neural architecture of semantic processing. Neither modular and distributed models of semantic organization provide an adequate explanation for the diversity of semantic deficits observed in patients who have sustained brain damage nor the patterns of brain activity reported in the functional neuroimaging literature. Evidence suggests that category, attribute and modality of processing are involved to some extent in semantic representation, although the degree to which each dimension contributes to semantic organization is disputed. Modular and distributed models differ in the emphasis placed on the organizing role of category (modular models) or attributes (distributed models) and ascribe little importance to the role of modality (e.g., visual or verbal) in semantic organization. Many neuroimaging studies suggest that semantic processing occurs in a distributed network of topographically organized processors, i.e., category domain or type of attribute (Chao et al., 1999; Ishai et al., 1999; Vandenberghe et al., 1996). For instance, Binder et al. (1997) identified a large network involving the frontal lobe including the inferior and middle frontal gyri, the temporal lobe including superior, middle, and inferior temporal gyri, and parahippocampal gyri, the parietal lobe including the angular gyrus, and a retrosplenial area activated during an animal monitoring task. Vandenberghe et al. (1996)

PAGE 44

32 identified a distributed multimodal semantic network (shared by pictures and words) that extends from the left superior occipital gyrus through the middle and inferior temporal cortex to the inferior frontal gyrus. A discussion of the evidence from lesion and functional imaging studies supporting the role of category, attribute, and modality of processing follows, and a reconciliation of these theories is presented in the matrix theory (Crosson et al., 2000). An understanding of the organization of semantic knowledge is necessary to detect differences in semantic organization that may exist between younger and older adults and potentially contribute to decreased word-retrieval in older adults. Evidence for the role of category Numerous case studies of category-specific impairments suggest that knowledge is organized according to category domain (Caramazza & Shelton, 1998; Hillis & Caramazza, 1991; Warrington & Shallice, 1984). Theories of a categorically organized semantic system posit the existence of multiple functionally and anatomically distinct subsystems in the brain (analogous to modules) each specialized for the knowledge of a particular category domain. Accordingly, category-specific deficits are explained in terms of localized damage to the cognitive module responsible for a particular category domain. Theoretically, damage to one module does not affect the function of another module, which accounts for fairly circumscribed semantic deficits (i.e., for naming animals: Barry & McHattie, 1995; Laws et al., 1995; tools: Hillis et al., 1990; Sacchett & Humphreys, 1992; fruits and vegetables: Farah & Wallace, 1992; body parts: Goodglass & Wingfield, 1993; medical items and instruments: Crosson et al., 1997). However, boundaries between categories are difficult to specify and the number of category modules in the brain is potentially unlimited.

PAGE 45

33 The case-study literature is marked by discrepancies in the interpretation of category-specific language deficits, and studies with similar methodologies have reached diametrically opposed conclusions of semantic organization. For instance, Caramazza and Shelton (1998) reported a case of a patient with a category-specific deficit for animals relative to other living things and artifacts. The deficit cut across visual and verbal modalities and was not specific to visual (has a mouth) or functional (breathes) attributes. Caramazza and Shelton (1998) concluded that conceptual knowledge is organized categorically in the brain according to the animate-inanimate distinction. In a similar study, Hart and Gordon (1992) reported a category-specific deficit for animals with a selective impairment for visual properties restricted to the verbal domain. They interpret this finding as not simply supporting the role of categorical information in semantic representation, but as evidence that the semantic system can be further fractionated according to modality of input/output and object attribute. They concluded that the semantic system consists of at least two modality-specific representational systems: a language-based system that subserves verbal knowledge and a visual-based system that subserves visual knowledge. They suggest that there are multiple divisions within the semantic system and that categorical information functions to facilitate knowledge of properties or attributes. Of note, the etiology of the lesions in these two studies could have contributed substantially to the different behavioral presentations. Caramazza and Shelton’s patient suffered a left cerebral vascular accident resulting in encephlomalacia in the left posterior frontal and parietal lobes, whereas Hart and Gordon’s patient demonstrated bilateral temporal lobe pathology due to a paraneoplastic syndrome.

PAGE 46

34 Although category-specific deficits constrain theories of semantic organization and support the notion that storage of semantic information is at least in part organized according to category domain, case studies are limited in their ability to interrogate the semantic system. Since lesions result from various cerebral insults, including stroke, herpes simplex encephalitis, and temporal lobe epilepsy, they do not respect functional anatomic boundaries; this makes interpretation of the neuroanatomical substrates of semantic information difficult. Also, it is difficult to determine whether the loss of category knowledge is due to loss of lesioned tissue or due to loss of connections with other cortical areas that may be responsible for maintaining semantic information. Thus, debates over the organization of semantic representation are not resolvable with case studies since brain systems are variably damaged and result in diverse patterns of deficits. For this reason, case studies have not succeeded in delineating the basic structure of semantic knowledge that leads to category-specific deficits. A number of neuroimaging studies that have investigated the functional neuroanatomy of category specificity have reported differential activity according to category domain (Chao et al., 1999; Ishai et al., 1999; Martin et al., 1996; Moore & Price, 1999). Investigation of the neural substrates of category specificity for living and nonliving things has resulted in a relatively consistent pattern of category-related activity in the posterior cortex (Chao et al., 1999; Damasio et al., 1996; Ishai et al., 1999; Martin et al., 1996; Moore & Price, 1999; Thompson-Schill et al., 1999). For instance, living things have been found to activate the left posterior superior temporal sulcus (for animals and faces, Chao et al., 1999), occipitotemporal sulcus (for faces, Ishai et al., 1999) and bilateral anterior temporal and right posterior middle temporal cortices (for animals and

PAGE 47

35 fruits, Moore & Price, 1999). In comparison, nonliving objects selectively activate the left middle temporal gyrus (Chao et al., 1999; Martin et al., 1996; Mummery et al., 1998), left posterior temporal cortex (Moore & Price, 1999) and a region in the inferior temporal gyrus (for chairs, Ishai et al., 1999). In addition, tool naming is associated with left premotor activity (Martin et al., 1996). However, there are also inconsistencies in reports of categorical activity. For example, activity on the medial surface of the temporal lobe has been attributed to the representation of animals versus tools (Damasio et al., 1996), tools versus animals (Martin et al., 1996) and for nonliving versus living objects (Mummery et al., 1998). These discrepant findings are likely due to methodological differences, such as method of neuroimaging used, sample size, analyses procedures, and baseline task employed (Cato et al., 2001). In addition, this variability may be due to differences in power between studies; a category may reach threshold for one category one time and the other category at a different time. Despite some discrepant findings for categorical organization, converging evidence indicates a reliable medial/lateral distinction in the fusiform gyrus for the representation of nonliving and living things, respectively, (Chao et al., 1999; Ishai et al., 1999). That is, animals activate the lateral fusiform gyrus to a greater degree than tools, whereas tools activate the medial fusiform gyrus more than animals during multiple tasks, including viewing, delayed matched to sample, and naming visually presented line drawings. Although category-specific deficits and neuroimaging results provide evidence that category domain is an important dimension of the organization of semantic knowledge, interpretation of these findings is difficult due to methodological differences. In addition, there is evidence that category domain is not the only dimension of semantic organization

PAGE 48

36 and that a model of the semantic system based solely on categorical distinctions is an over-simplification of the structure of semantic representation. Evidence for the role of attributes An alternative interpretation of semantic organization proposes that attributes are the fundamental organizing dimension of semantic information (Carbonnel et al., 1997; Devlin et al., 1998; Farah & McClelland, 1991; Gonnerman et al., 1997; Moore & Price, 1999; Mummery et al., 1998; Warrington & McCarthy, 1987; Warrington & Shallice, 1984). Attributes are features that distinguish members of categories from each other, and include visual form, function, human action, color, movement and emotional connotation. Categories are thought to emerge from underlying semantic features that are distributed throughout the semantic system. Thus, category-specific deficits can be explained as resulting from damage to these distributed semantic features. Warrington and Shallice (1984) suggested that the essential difference between living and nonliving things may be the relative degree to which their defining attributes are respectively sensory or functional. The sensory/functional theory claims that animals or living things are characterized more by their visual features, such as form and color, and implements or nonliving things are characterized by their function. Proponents of the sensory/functional theory argue that semantic knowledge is organized into two subsystems: one that stores information about the visual characteristics of objects, and one that stores information about the uses of objects and other nonsensory properties. Farah and Wallace (1992) argue that living objects are more easily compromised because their form and function are disparate, versus implements whose form defines their function. Thus, the loss of perceptual information about a living object renders the system unable to access the functional information via an alternate route.

PAGE 49

37 Evidence for the role of attributes in the semantic system comes from case studies of patients that demonstrate impaired knowledge of attributes, such as functional knowledge (Laws et al., 1995; Sheridan & Humphreys, 1993) and visual perceptual features (Coltheart et al., 1998; De Renzi & Lucchelli, 1994; Gainotti & Silveri, 1996; Silveri & Gainotti, 1988) independent of modality of presentation. Some reports of category-specific deficits have attributed the specificity of the semantic deficit to both categorical and attribute organization. For example, Coltheart et al. (1998) reported a patient who demonstrated a selective deficit for visual perceptual knowledge that cut across living and nonliving categories. Because the patient’s non-perceptual knowledge (e.g., knowledge of dangerousness) of objects was preserved, they interpret this finding to suggest two distinct subsystems in semantic memory, one containing information about the visual properties of objects and another containing information about non-perceptual properties. In addition, a study that looked at naming in patients who had undergone a left anterior temporal lobectomy reported impaired knowledge for items characterized by human action (implements and human action verbs) (Lu et al., 2002). Furthermore, information regarding object use or action appears to be stored in the premotor cortex as well, as evidenced by impairments in action knowledge following a lesion of the left premotor area (Damasio & Tranel, 1993). Recent functional neuroimaging studies have attempted to localize areas in the brain responsible for processing semantic attributes such as visual features, form, function, color, movement and emotional connotation. It is generally thought that knowledge of sensory or motor attributes is stored in cortical regions proximal to the primary cortical sensory areas (i.e., primary visual area, primary motor area) involved in

PAGE 50

38 the perception or mediation of these attributes (Crosson et. al., 1999). Knowledge of motion associated with using objects has been localized to the left lateral and middle temporal cortex, and the ventral premotor area has been suggested as the site for stored knowledge of how objects are used (Beauchamp et al., 2003; Chao et al., 1999; Chao & Martin, 2000; Martin et al., 1995; Martin et al., 1996). Selective activity for tools in the left ventral premotor cortex may be related to the retrieval of information about hand movements associated with the use of manipulable nonliving objects (Chao et al., 1999). Recently, a distinction among action words according to the body part used to execute the movement (e.g., lick-tongue, throw-arm, kick-leg) has been found in the premotor and motor cortex (Hauk et al., 2004; Pulvermller, 2001; Pulvermller et al., 2001). Activity for action words that reference a body part is consistent with the topographical organization of the motor cortex. In other words, the “face” area, known to be responsible for control of facial muscles, is activated during action words that reference the face, such as licking or eating. Similarly, the “arm” area is activated during action words related to the use of the arm (i.e., throwing) and the “leg” area is activated during action words related to the use of the leg (i.e., kicking). The discrete localization of areas in close proximity to each other in the premotor and motor cortex involved in semantic processing of action words indicates that a similar degree of specificity in distinguishing attributes is possible in other regions of the brain, such as the ventral visual stream. Similarly, knowledge of color is represented in the left inferior temporal, left frontal, and left posterior parietal cortices, in close proximity to the cortical regions that mediate perception of color (Chao & Martin, 1999; Martin et al., 1995;). Recent functional imaging research conducted in our lab suggests that the semantic attribute, emotional

PAGE 51

39 connotation, is independent of object category and is localized to the frontal pole and retrosplenial cortex (Cato et al., 2004). The majority of functional imaging studies that attempt to localize semantic attributes are faced with a conundrum; the attributes under investigation are confounded with object category. To date, few studies have been designed to disentangle the contributions of attribute independent of category. For instance, the fusiform gyrus has consistently been demonstrated to play a role in visual semantic information, and thus has been a region of interest of many functional imaging studies. Specifically, categorical differentiation in the fusiform gyrus has been consistently reported, with a medial/lateral distinction between implements and animals, respectively (Chao et al., 1999). Yet, this categorical distinction has been interpreted as evidence that the semantic system is organized according to attribute since attributes such as visual form and motion are also used to distinguish between tools and animals (Chao et al., 1999). Unfortunately the design of the study did not assess the interaction of features with object category since the naming stimuli were divided on the basis of semantic category, not on the basis of semantic attribute, so conclusions that category-specific effects represent semantic organization by attribute are premature and not consistent with the data. Ishai et al. (1999) found this categorical distinction in the fusiform gyrus for chairs, faces and houses. However, they show support for a more distributed system in which there is a spread of activation for each category, such that one object type activates to some degree an area maximally activated for another category. Based on these results, they suggest that this region of cortex represents information about object form, although this was not

PAGE 52

40 directly tested, and explain these category effects as resulting from cortex specialized for different features that distinguish categories. There have been few attempts to investigate the interaction of category and semantic attribute. Physical features, such as form, shape, and internal detail constitute a large portion of visual attributes that characterize an object. Moore and Price (1999) identified distinct regions responsible for processing visual configuration and category domain by comparing complex and simple living and nonliving items. They found both an attribute and category effect. Specifically, the right occipitotemporal, fusiform and the medial extrastriate cortices were activated by multicomponent relative to simple shaped objects and bilateral anterior temporal and right posterior middle temporal cortices were activated for living objects (animals and fruits) (Moore & Price, 1999). These results suggest a possible distinction in the ventral temporal cortex for semantic processing, whereby the anterior ventral temporal cortex is organized by category and the posterior ventral temporal cortex is organized by attribute. However, this type of breakdown of the semantic system according to anterior and posterior specialization has not been systematically investigated. Kraut et al. (2002) recently reported a dissociation between attribute and category, in which both tools and fruits activated a similar region of the premotor cortex (BA 6/44). Since tools and fruits differ in terms of category domain (nonliving vs. living, respectively) the authors interpret activity in BA 6 to represent knowledge of attributes corresponding to the use of the objects, since both tools and fruits are manipulable. However, since this task did not include a motor baseline task to control for the button response in the categorization task, the extent to which this activity reflects preparation to

PAGE 53

41 respond using the button press is unknown. In addition, my master’s thesis (Wierenga, 2001) identified a dissociation between category and visual attribute in the fusiform gyrus in younger adults during a naming task similar to the one used by Chao et al. (1999). Results suggest an anterior/posterior distinction in the fusiform gyrus for processing category and visual attributes, respectively. The right posterior fusiform gyrus was activated for both animals and vehicles, thought to be identified by global form, whereas the left posterior fusiform gyrus was activated by tools, thought to be identified by their local details. In contrast, a categorical distinction between living and nonliving items was seen in the anterior fusiform gyrus, similar to that reported by Chao et al. (1999). Consistent with previous findings, animals activated a region of the lateral fusiform gyrus whereas vehicles and tools activated a medial region of the fusiform gyrus bilaterally. The majority of functional imaging studies have investigated attributes associated with categories, rather than assessing attributes independent of category. In order to conclude that the semantic system is organized according to attribute, the effects of category and attribute must be disentangled. This can be accomplished by juxtaposing category against attribute. The current study experimentally pits the category specific hypothesis against the attribute hypothesis by manipulating category and attribute to extricate their effects in the anterior and posterior fusiform gyrus. Evidence for the role of modality Semantic knowledge may divide along the modality by which the information is processed or analyzed. The semantic system has traditionally been thought to be organized according to two primary subsystems based on modality of input. The visual semantic system subserves meaning pertaining to objects and events that are viewed, and the verbal semantic system concerns meaning that is distinctly related to language

PAGE 54

42 (Paivio, 1971, 1986; Warrington & Shallice, 1984). Arguments for a modality-specific semantic system are supported by evidence that semantic information can be independently accessed from visual or auditory input. However, this model also assumes that the modality-specific subsystems are connected in some form to allow for the integration of visual and verbal semantic information. The interaction of category-specific and modality-specific deficits demonstrated in patients with category-specific impairments constrained to the verbal or visual modality provides support for the distinction between visual and verbal semantic systems (Hart & Gordon, 1992; Warrington & McCarthy, 1994; Warrington & Shallice, 1984) and is concordant with the neuroanatomical substrates underlying visual and auditory processing in the brain (Chao et al., 1999; Ishai et al., 1999). In particular, case studies have reported patients with an impairment for animals confined to the verbal modality (McCarthy & Warrington, 1988), a category-specific deficit for the visual characteristics of animals when assessed through the verbal system (Hart & Gordon, 1992), and an impairment for artifacts and body parts presented visually (Sacchett & Humphreys, 1992). Further evidence for a separation between the verbal and visual semantic system was presented in a patient with a category-specific visual agnosia (characterized by the inability to assign meaning to visually presented stimuli) for nonliving objects (Warrington & McCarthy, 1994). Localization of semantic information according to modality has typically been assumed to correspond to cortices primarily responsible for processing auditory or visual information, i.e., the auditory and visual cortices. However, some evidence suggests that localization of auditory/verbal and visual information must be proximal in the brain to account for category-specific deficits that involve both modalities (Cato et al., 2001;

PAGE 55

43 Crosson et al., 2000). Arguments against the localization of modality suggest that visual and verbal modality share a similar neural network, as demonstrated by a lack of differences in neural substrates of semantic processing involving verbal and visual information (Vandenberghe et al., 1996). However, often modality of input and modality of attribute are confounded, which contributes to the challenge of localization of modality. For example, although Vandenberghe et al. (1996) attempted to differentiate between visual and verbal modalities, his stimuli (pictures and words) were presented visually. Similarly, Thompson-Schill et al. (1999) attempted to examine visual attributes using auditory input. Further research needs to be conducted to disentangle the contribution of attribute and modality in an attempt to localize semantic modality. However, modality-specific disorders including the agnosias, support the notion that modality of processing semantic information is also an important dimension of semantic organization. In summary, although lesion and functional imaging studies may draw disparate conclusions regarding the semantic system, it is clear that category, attribute and modality are important dimensions in the organization of semantic information. Nevertheless, the independent contribution and the effects of the interactions between these organizing factors have not yet been fully determined. A resolution between modular and distributed models of semantic memory may provide a more sufficient account of the relative importance of category, attribute, and modality in the organization of the semantic system. For example, information may be stored in modules but within these modules semantic information pertaining to category, attributes, or modality distributed and weighted. Thus, activity within modules and communication between

PAGE 56

44 modules may occur in a manner consistent with parallel distributed or computational processing. The Matrix Theory Recently, the matrix theory was proposed to reconcile the discrepant findings from category-, attributeand modality-specific naming deficits (Crosson et al., 2000) and integrate models of semantic organization. According to this theory, the semantic system resembles a matrix that includes the modality in which information is processed (i.e., visual vs. verbal), attributes relevant to identifying objects (e.g., shape, color, emotional connotation), and the semantic category of objects (e.g., living vs. nonliving), see Figure 3-2. Each of these semantic characteristics is thought to have specific neural substrates that can be identified from unique patterns of activity with functional imaging. In accordance with the previous literature (Hart & Gordon, 1992; McCarthy & Warrington, 1988; Warrington & McCarthy, 1994; Warrington & Shallice, 1984), the matrix theory recognizes that the semantic system can be divided according to modality of processing (visual vs. verbal) and that these two dissociable systems are most easily accessed by their corresponding input modality. In order to account for Hart and Gordon’s (1992) patient who could not access visual information when using the verbal modality, the matrix theory asserts that categorical information cuts across these modalities. Furthermore, Crosson et al. (2000) acknowledged the importance of sensory and functional attributes in conceptual processing, and propose that other attributes, such as emotional connotation, visual form, and motoric action may be important in semantic organization. Such attributes become associated with representations of conceptual information in both the visual and verbal modalities. The categorical division results from common defining attributes. In order to process multiple defining attributes,

PAGE 57

45 convergence of information streams in the brain must occur. The matrix theory posits that attributes are incorporated into the object representation in the temporal lobe via converging input from cortical and limbic structures in a manner that is not random. The placement of categorical information is driven by the convergence of the incoming information streams (consisting of attributes) that are most salient for distinguishing a class of items. The categorical division along the temporal lobe emerges as common defining attributes cluster based on inputs to heteromodal cortex, reminiscent of computational models of semantic organization. Figure 3-2 illustrates that category knowledge is represented in the verbal and visual modalities and that the relative contribution of attributes differs by modality and category. Figure 3-2. Depiction of the matrix theory.

PAGE 58

46 Semantic Processing in the Ventral Visual Stream The ventral temporal cortex has been implicated in processing visual attributes during object identification (Chao et al., 1999; Ishai et al., 1999) and is an area of interest in the current study. Specifically, the ventral visual stream has been identified as responsible for determining "what" an object is (Ungerleider & Mishkin, 1982). The role of the inferior temporal cortex in feature processing and object identification is supported by evidence that damage to regions in the ventral visual stream results in associative agnosia (an inability to recognize objects, despite an apparent perception of the object), prosopagnosia (an inability to recognize familiar faces), and achromatopsia (color blindness). In order to make specific predictions about the organization of the visual semantic system, particularly regarding areas activated by vehicles, tools, and animals, it is necessary to understand how the brain processes visual information. Although visual processing is relatively well understood based on work in animal models, semantic processing in the visual ventral stream is not as well defined because study in nonverbal species is difficult. The occipitotemporal pathway, or ventral visual stream, extends from the striate cortex along the course of the inferior longitudinal fasciculus to the anterior part of the inferior temporal cortex (area TE) and is involved in object identification (Mishkin et al., 1983; Ungerleider & Haxby, 1994), see Figure 3-3. The occipotemporal pathway also has connections to the limbic system and the frontal lobes, enabling associations between vision and emotion or motor movements (Mishkin et al., 1983). The ventral visual stream only receives visual input and thus remains modality-specific throughout its course, implicating this area in picture or visual object identification. Visual information is processed in a hierarchical fashion within the ventral visual stream (Mishkin et al.,

PAGE 59

47 1983; Ungerleider & Haxby, 1994; Ungerleider & Mishkin, 1982). Studies of the primate visual system in the macaque reveal a multitude of visual processors and pathways interconnecting cortical visual areas (e.g., Van Essen et al., 1992). Through successive stages of processing, low-level visual information from the retina is processed for identification and transformed into useful representations and assigned meaning (Mishkin et al., 1983). As information proceeds through the occipitotemporal pathway, neuronal responses become increasingly complex. Early in the ventral visual stream cells function to identify contours or distinguish the object from the background, and cells progressively respond to more of the object's physical properties, such as color, size, and texture. Cells in the inferior temporal cortex selectively respond to global or overall object features, such as form and faces and integrate the information into a complete representation of the object (Ungerleider & Haxby, 1994). The synthesis of the physical properties of an object occurs in area TE, which is thought to store object representations and is involved in object recognition (Mishkin et al., 1983). Therefore, relatively early in visual processing, properties such as color, size, and texture are processed, followed by visual form, and eventually object recognition occurs within the more anterior portions of the ventral visual stream (Mishkin et al., 1983; Ungerleider & Haxby, 1994). Figure 3-3. The ventral visual stream of the macaque (modified from Ungerleider & Haxby, 1994).

PAGE 60

48 A major component of the ventral visual stream is the fusiform gyrus, an elongated structure starting about 3 cm posterior to the temporal pole and running roughly 8 cm posteriorly, nearly to the occipital pole. The fusiform gyrus is the structure in which Chao et al. (1999) and Ishai et al. (1999) have found a medial vs. lateral localization for nonliving vs. living things, respectively. Yet, there remains a lack of agreement regarding the function of the fusiform gyrus in the semantic network. One view argues that the fusiform gyrus is primarily involved in low level perceptual processing of attributes, such as color perception, shape analysis, and object form (Kanwisher, 1997; Malech et al., 1995). An alternative view is that the fusiform gyrus is responsible for higher level visual perceptual semantic processing and object identification according to category domain (Chao & Martin, 1999; Murtha et al., 1999; Thompson-Schill, 1999). The aim of the present study is to elucidate the role of the fusiform gyrus in visual semantic processing. The implication of the hierarchical nature of object identification for semantic processing within the ventral visual stream is simple: attributes such as color, size, texture, and form are likely to be processed in the more posterior regions of this processing stream, i.e., fusiform gyrus. Since category is closely related to the actual identification of objects, category is likely to be processed in the more anterior portion of the ventral visual stream, i.e., fusiform gyrus. Global and Local Features An additional characteristic in processing visual form in the human ventral cortex must be introduced at this point. The distinction between local and global features is important in visual processing of pictures, particularly when investigating the role of visual attributes in the organization of semantic information. Global visual attributes refer to the basic shape of an object (e.g., an object's outline); local visual attributes refer

PAGE 61

49 to the smaller details of an object's visual features that comprise the larger configuration. For example, an airplane can be recognized by its outline or global form, but airplanes also have local features such as windows, doors, etc. Evidence suggests that global visual features are processed in the ventral visual stream of the right hemisphere and local visual features are processed in the ventral visual stream of the left hemisphere (Delis et al., 1992; Doyen & Milner, 1991). In addition, Navon’s (1977) global precedence hypothesis posits that the global level is processed before the local level, consistent with the hierarchical processing of information in the visual system. Navon (1977) reported that individuals responded faster to global than local features, and that incongruent global information interfered with local processing, but not vice versa. Recent electrophysiological and neuroimaging studies of global/local feature analysis investigating temporal stages and lateralized differences between global and local processing support a global precedence effect (Han et al., 2000; Tanaka et al., 2000), a right hemisphere bias for global features and a left hemisphere bias for local features at early stages of processing (Evans et al., 2000; Fink et al., 1997), and an interaction between spatial attention and hierarchical analysis (Fink et al., 1997; Han et al., 2000). Results of my master’s thesis (Wierenga, 2001) suggest that visual processing of global and local features is involved in semantic processing of line drawings and leads to hemispheric differences in functional activity during confrontation naming of different categories of objects according to the characteristic physical features of members in a category. Specifically, the right posterior fusiform gyrus was activated during processing of animals and vehicles, which are characterized by their global form. Conversely, the

PAGE 62

50 left posterior fusiform gyrus was activated during processing of implements, which are characterized by local details. Convergent evidence for the global/local distinction is found in the patient literature. Patient studies have investigated visuospatial abilities in patients with focal brain lesions and diffuse brain disease and report dissociations in global/local feature processing (Delis et al., 1988; Delis et al., 1992; Doyen & Milner, 1991; Massman et al., 1993). Individuals with right hemisphere damage demonstrated selective impairment in processing the global level of stimuli whereas individuals with left hemisphere damage demonstrated poor performance in processing the local level of stimuli (Delis et al., 1992). Doyen and Milner (1991) found that patients with left temporal lobe lesions were less affected by interference from global features, supporting the role of the right temporal cortex in global processing of visual information. Retrieval of Semantic Knowledge Since aging may differentially implicate representation of semantic information and the retrieval of words from semantic stores, attention must also be paid to the mechanisms underlying retrieval of semantic knowledge, which are dependent on frontal lobe structures (Raymer & Rothi, 2000). Regions of the left lateral and medial prefrontal cortex are recruited during retrieval of semantic information (Binder et al., 1997; Demonet et al., 1992; Vandenberghe et al., 1996). However, there is current debate regarding the specific role of the frontal lobes in accessing semantic information, particularly concerning the role of the left inferior frontal gyrus. Several roles of semantic processing have been attributed to the left lateral inferior prefrontal cortex, including semantic analysis, effortful retrieval, maintenance, and/or control of semantic information (Fiez, 1997). Support for semantic analysis is provided

PAGE 63

51 by evidence of left inferior prefrontal cortex activity during tasks that involve several possible responses such as verb generation or word stem completion tasks (Gabrieli et al, 1998; Petersen et al., 1988). Decreased activity in the left prefrontal cortex associated with semantic priming has also been interpreted to suggest that this region is involved in semantic analysis (Gabrieli et al., 1996). Grabowski et al. (1998) reported activity of the left inferior frontal gyrus during a naming task of animals, tools, and faces. Since no categorical differentiation was noted, they interpreted these results as evidence of general semantic retrieval. Gabrieli et al. (1998) suggest that left prefrontal activity during semantic tasks represents semantic working memory and hypothesize that left inferior prefrontal activity reflects the degree to which semantic information must be held temporarily in working memory to perform a particular task. They argue that this theory is consistent with evidence of increased left prefrontal activity during tasks that require a greater amount, longer duration, or more selection of semantic information. Left inferior frontal activity in low constraint tasks (verb generation and stem completion) involving high selection demands provides support for the role of this region in semantic selection (Barch et al., 2000; Gabrieli et al., 1998). Thompson-Schill et al. (1997) argue that the left inferior frontal gyrus (LIFG) is involved in selection of some relevant feature of semantic knowledge from a set of competing alternatives rather than semantic retrieval per se. Accordingly, they suggest that the degree of activity of the LIFG during retrieval of semantic information is modulated by the selection demands of the semantic task, such that greater LIFG activity is associated with greater selection demands. In response to Thompson-Schill et al.’s (1997) theory of selection of relevant semantic features, Wagner et al (2001) suggested that the role of LIFG in semantic

PAGE 64

52 processing involves controlled semantic retrieval, consistent with the prefrontal cortex’s role in cognitive control. The left inferior prefrontal cortex is assumed to guide recovery of semantic knowledge in situations when prepotent or automatic responses are not appropriate to the task demands. Thus, Wagner et al. (2001) suggest that the left inferior prefrontal cortex may be involved in semantic retrieval even when selection against competing responses is not required. Wagner et al. (1999) attempted to investigate controlled semantic retrieval in a low-selection task similar to that used by Thompson-Schill et al. (1997), but they varied the associative strength between cue and correct target. Findings support the role of the anterior and ventral extent of the left inferior prefrontal cortex in controlled semantic retrieval. In addition, greater left inferior prefrontal activity was demonstrated as semantic retrieval demands increased (4 vs. 2 targets). However, target stimuli included both concrete and abstract words, which differ in imageability and thus add a confounding variable to this study. The medial frontal cortex may also be of interest when examining semantic processing, since this region plays a role in initiating spontaneous language production (Crosson et al., 1999; Damasio & Anderson, 1993). In particular, the pre-supplementary and supplementary motor areas (pre-SMA/SMA) are involved in internally generated language. There is evidence of a medial/lateral distinction in terms of internally versus externally guided language, respectively. Pre-SMA is more involved in internally guided word generation whereas the inferior frontal sulcus is more involved in externally guided word generation (Crosson et al., 2001). More specifically, evidence suggests that the anterior cingulate cortex (pre-SMA) evaluates the demand or need for cognitive control by monitoring for the occurrence of response conflict in language generation (Barch et

PAGE 65

53 al., 2000; Carter et al., 2000; MacDonald et al., 2000). Findings that the anterior cingulate cortex and left inferior frontal cortex are more active during tasks with high conflict are consistent with previous studies that demonstrate the involvement of the anterior cingulate cortex in low constraint tasks (Barch et al., 2000; Carter et al., 2000; Thompson-Schill et al., 1997). However, the anterior cingulate cortex was the only region identified that responded more to weakas compared to strong-verb associates. Production of a weakover a strong-verb associate is thought to elicit more competition between alternative responses and does not involve automatic processing. In addition, Barch et al. (2000) report an interaction between noun type (high vs. low constraint) and verb response (weakvs. strong-associate) seen only in the anterior cingulate cortex, which suggests that the degree of conflict elicited by the production of weak-verb associates is modulated by the relative constraint of the associated noun. Taken together, these findings support the role of the anterior cingulate cortex in evaluation of conflict monitoring. Lastly, it must be noted that the basal ganglia and thalamus also play a role in semantic processing, particularly in the initiation of search strategies to retrieve lexical items. Recent evidence suggests that the left pre-SMA-dorsal caudate nucleus-ventral anterior thalamic loop is involved in semantic word generation and plays a role in determining the likelihood that a particular response will be made (Crosson et al, 2003). This is consistent with the role of pre-SMA in conflict monitoring and in internally guided word generation (Barch et al., 2000; Crosson et al, 2001). In summary, recent findings of the involvement of the LIFG in semantic processing clearly demonstrate that this region is involved in retrieval or access of semantic

PAGE 66

54 information. However, the exact characterization of the role of this region is still not clear, in part due to the difficulty of differentiating between processes involved in semantic access. For example, retrieval of semantic information likely involves a strategic search of semantic stores and selection among alternative responses. Definitions of constructs such as cognitive control and selection remain elusive and are operationalized differently across studies (Thompson-Schill et al., 1997; Wagner et al., 2001). Medial prefrontal regions are involved in initiation of internally guided language generation and conflict monitoring. A complete analysis of strategic processes involved in accessing semantic information includes the interaction between medial and lateral regions of the left prefrontal cortex. Further research is needed to determine whether these neural substrates of executive language functions change with age.

PAGE 67

CHAPTER 4 HYPOTHESES Hypothesis 1 Older adults will show an increase in activity relative to younger adults in medial and/or lateral frontal regions of the left hemisphere during the naming task. Based on the existing literature, we hypothesize that healthy older adults will show more activity where mechanisms are impaired to compensate for reductions in efficiency. Accordingly, if word retrieval difficulties stem from problems with executive functions of language, then we expect increased activity in the frontal regions related to these functions. Medial frontal structures, especially those at the border of pre-SMA and the rostral cingulate zone, are involved in initiation of language and monitoring conflict between competing responses (Barch et al., 2000; Carter et al., 2000; Crosson et al., 1999). Lateral frontal cortex, including the inferior frontal gyrus and cortex along the inferior frontal sulcus, is involved in selection and execution of a semantic response appropriate to internal and external constraints (Crosson et al., 2003; Crosson et al., 1999; Thompson-Schill et al., 1997; Wagner et al., 2001). Thus, a difference in level or extent of activity for older vs. younger persons in either medial or lateral frontal structures would indicate a change in the neural substrates for executive functions. If such changes in frontal activity are seen in older compared to younger persons, it would indicate that future studies should focus on manipulating the implicated functions to firmly establish the precise functional-structural relationship accounting for word-finding difficulty in older persons. 55

PAGE 68

56 Hypothesis 2 An alternative hypothesis to Hypothesis 1 is that activity for older participants will be significantly less lateralized in the posterior fusiform gyrus and/or more diffuse in the anterior fusiform gyrus, characterized by less medial vs. lateral distinction, than for younger persons. If word retrieval difficulties result from the deterioration of information in lexical-semantic stores, we expect differences in the pattern of activity in the temporal lobe during confrontation naming in younger and older adults. Potential changes in the temporal lobe are somewhat more complicated than for the frontal lobe. It is expected that in an intact semantic system in the posterior inferior temporal cortex (feature-based cortex), lateralization of processing for semantic categories will be based upon the degree to which processing of global (right hemisphere) vs. local (left hemisphere) visual information is necessary in identification of objects within the category. In the anterior fusiform gyrus, medial vs. lateral activity bilaterally will be determined by the distinction of living vs. nonliving things (see Hypothesis 3). Activity from cognitive processes can become more diffuse and less lateralized with aging (Cabeza, 2001). If semantic substrates in the inferior temporal lobe become less efficient with aging, then during confrontation naming, we might expect to see either a breakdown in lateralization of processing for global and local features in the posterior fusiform gyrus and/or broad activity within the anterior fusiform gyrus without respect to whether items are living or nonliving things. Hypothesis 3 Based on findings from my master’s thesis, we expect to find neural activity in the temporal lobe indicative of both attribute and category (Wierenga, 2001). More specifically, we expect a double dissociation between processing of visual-semantic

PAGE 69

57 features in the posterior portion of the fusiform gyrus and processing of categories in the anterior portion of the fusiform gyrus in younger adults. In addition, it is likely that global and local visual attributes will place greater processing demands on different hemispheres. Therefore, we expect that visual attributes are processed in the posterior portion of the fusiform gyrus, and lateralization of processing for semantic categories will be based upon the degree to which processing of global (right hemisphere) vs. local (left hemisphere) visual information is necessary to distinguish between objects within the category. This right/left dissociation is consistent with the literature on processing of global and local features (Delis et al., 1992; Doyen & Milner, 1991; Fink et al., 1997). However, in the anterior portion of the fusiform gyrus where object identification is performed, we expect a categorical distinction; living things will activate the lateral anterior fusiform gyrus and nonliving things will activate the medial fusiform gyrus consistent with previous reports (Chao et al., 1999; Ishai et al., 1999). We expect to demonstrate this dissociation between the posterior and anterior fusiform gyrus in the way that vehicles are processed (according to attribute or category). Vehicles are members of the superordinate category of nonliving things but share a visual feature with living things. Specifically, both animals and vehicles are identified on the basis of global form in contrast to implements, which are distinguished by local features. Note that in the inferior temporal cortex, tools and vehicles should cluster together in the anterior portions of this region (similar category) but animals and vehicles should cluster together in the posterior portion of this region (similar features). Kraut et al. (2002) report a similar dissociation between attribute and category in the premotor cortex. They found that tools, which are nonliving things, and fruits, which are living things, shared an

PAGE 70

58 area of activity in the premotor cortex because both kinds of items are manipulable. In other words, Kraut et al. (2002) found that processing specific items from living and nonliving categories activated the neural substrate for a common feature. We expect that these distinctions in the inferior temporal cortex can be applied to both younger and older participants, though the degree to which they can be applied will depend on the origin of the naming problems in older adults. More specifically, the hypotheses are: Hypothesis 3a In the right posterior fusiform gyrus, processing of both animals and vehicles will evoke significantly greater activity than processing of tools because identification of animals and vehicles, but not tools, is based on processing of global visual form. Hypothesis 3b In the left posterior fusiform gyrus, processing of tools will evoke significantly greater activity than processing of animals and vehicles because identification of tools, but not animals or vehicles, is based on processing of local visual details. Hypothesis 3c In the anterior fusiform gyrus bilaterally, animals will activate the lateral portion of the fusiform gyrus more than tools or vehicles because animals, but not tools or vehicles, are living things. Hypothesis 3d In the anterior fusiform gyrus bilaterally, both tools and vehicles will activate the medial portion of the fusiform gyrus more than animals because tools and vehicles, but not animals, are nonliving things.

PAGE 71

CHAPTER 5 PILOT STUDY Participants Twenty (5 male, 15 female) healthy young adults (age = 21-30 years, M = 24.6 years; education M = 16.9 years) with normal or corrected-to-normal vision participated in the pilot study. All participants were native English speakers who were recruited from the University of Florida faculty, staff, and students and from the Gainesville, Florida community. All participants were strongly right-handed (Edinburgh Handedness Inventory, Oldfield, 1971). Informed consent was obtained from participants according to guidelines established by the Health Science Center Institutional Review Board at the University of Florida. Experimental Design and Task The pilot study examined the interaction of category and visual attributes in semantic organization by investigating semantic category identification as a function of parametric spatial frequency content (Wierenga et al., 2003). Spatial frequency content, a primary perceptual feature for recognizing objects, has been found to differentially contribute to category identification (Vannucci et al., 2001). Low spatial frequency content refers to the course form or contour of an object’s shape; it is similar to global form. High spatial frequency content refers to the fine internal details of objects, similar to local visual information. The spatial frequency of pictures of objects can be manipulated to preserve global form while varying the amount of local details. This is done by parametrically removing high spatial frequency information from the images. 59

PAGE 72

60 Therefore, this technique allows us to investigate the degree to which global and local visual attributes contribute to the identification of animals, tools and vehicles. Specifically, we can test our assumption that vehicles and animals are similar because they both rely on global form for identification, whereas tools are different because local details are necessary for identification. We hypothesized that both animals and vehicles are identified by low spatial frequency information (course form or contour) whereas tools require more high spatial frequency information (details) for identification. Stimuli 75 gray-scale photographs of real-life objects belonging to animals, tools, and vehicles (25 photographs/category) were included. Photographs were equated for size and items in the three categories were balanced for familiarity and frequency in the English language (Coltheart, 1981; Kucera & Francis, 1967). Each photograph was spatially filtered at 9 levels by removing high spatial frequency content (i.e., details) from the original image using the Gaussian Blur filter (Adobe Photoshop 7.0) in increments of 3 pixels. For example, the 9th level of filtering corresponds to a blur radius of 27 pixels and the 3rd level of filtering corresponds to a blur radius of 9 pixels. Each stimulus was presented in canonical orientation at the center of the computer monitor. Design and Task Each photograph was presented for 200 ms at 10 resolution levels (9 filtered plus original) of high-frequency spatial filtering in ascending order (by adding high spatial frequency content) (see Figure 5-1). The presentation order of the objects was randomized for each participant. A training phase preceded the experimental phase in which different photographs were used. Participants were instructed to press one of two buttons according to whether they identified the object or not. After the identification

PAGE 73

61 response, subjects were asked to name the object and were given feedback on their naming accuracy for each object. Figure 5-1. Schemata of an ascending sequence of 9 filtering levels of a stimulus in each category for the pilot study. Results A threshold value was established for each object per subject according to the level of spatial filtering at which the item was initially correctly identified. Reaction times were recorded for the initial correct response for each item. A repeated-measures ANOVA was conducted on threshold values for category at 3 levels (animals, tools, vehicles). A significant effect for category was found [F(2,38)=81.54, p = .000]. Planned contrasts revealed that each category differed significantly from the other two categories, indicating that identification of the three categories is differentially affected by spatial frequency content. Identification of animals required less high spatial

PAGE 74

62 frequency information (details, internal elements) than tools [t(19) = 10.52, p = .000] and vehicles [t(19) = 8.37, p = .000] for identification. In contrast, tools required a greater amount of high spatial frequency content than vehicles [t(19) = -5.71, p = .000] for identification. Taken together, findings suggest that vehicles were intermediate between animals and tools in the amount of local detail required for identification (Figure 5-2). Based on total percent accuracy at each filtering level for each category, animals and vehicles were identified earlier (at resolution levels of less high spatial frequency content) than tools (Figure 5-3). Figure 5-2. Average threshold level of spatial filtering for identification of animals, tools, and vehicles. Figure 5-3. Percentage of correct responses for animals, tools, and vehicles at each level of spatial filtering.

PAGE 75

63 A repeated-measures ANOVA was conducted on reaction time for initial correct response for category at 3 levels (animals, tools, vehicles). No significant effect for category was found [F(2,38) = .64, p = .532]. Based on reaction time, results suggest that the three categories did not differ in terms of level of difficulty in identification of objects and support the role of spatial frequency content in category identification. Results of the pilot study indicate that identification of animals, tools, and vehicles is differentially affected by spatial frequency content. This finding suggests that the contribution of global form vs. local detail to the recognition of objects differs by semantic category. Differences between categories indicate that perceptual (visual) features have an important role in semantic processing. Current results form a basis with which to examine the neural substrates of semantic information in regards to the dimensions of category and visual attributes. Specifically, the results informed our selection of stimuli to be used in the FMRI experiment and enabled us to experimentally manipulate the categories to equate animals and vehicles according to global detail necessary for identification in contrast to tools.

PAGE 76

CHAPTER 6 METHODS Participants Twenty (10 male, 10 female) neurologically normal young adults (age 20-34 years, M = 25.1 years, education M = 15.85 years) and twenty (10 male, 10 female) neurologically normal older adults (age 68-84 years, M = 74.9 years, education M = 15.65 years) participated. The age range for older adults was chosen based on evidence that impairments in word retrieval become more pronounced in the seventh decade (Ivnik et al., 1996). Older and younger participants did not differ significantly according to level of education, [t(38) = -.28, p = .779] (see Table 6-1). All participants were native English speakers who were recruited from the University of Florida faculty, staff, and students, the Malcom Randall VAMC and from the Gainesville, Florida community. All participants were strongly right-handed (Edinburgh Handedness Inventory, Oldfield, 1971). Potential participants were excluded if they reported a history of neurological disease, dementia or mild cognitive impairment, cardiovascular disease, uncontrolled hypertension, Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition Axis 1 diagnosis, learning disability, attention deficit disorder, substance abuse, or poor visual acuity. In addition, participants were excluded if they had a cardiac pacemaker, metal implants, metal aneurysm clips, or metal in their body other than dental fillings, or if they were taking psychoactive prescription medications. Female participants were excluded if they were pregnant or trying to become pregnant. Current medications were recorded for each participant. Participants were instructed to abstain from caffeine the day of the scan. 64

PAGE 77

65 Informed consent was obtained from participants according to guidelines established by the Health Science Center Institutional Review Board at the University of Florida. Participants were paid $25 for participation. Table 6-1. Demographic Variables for Younger and Older Adults. Younger Adults (n=20) Older Adults (n=20) Demographics M SD M SD Age 25.10 (4.23) 74.85 (4.55) Education 15.85 (2.03) 15.65 (2.43) Gender (women/men) 10/10 10/10 Although data will only be reported on the 40 subjects mentioned above, 57 participants were enrolled in the study. However, seven participants were excluded from the FMRI experiment on the basis of neuropsychological testing performance that suggested the presence of mild cognitive impairment. Of the 50 participants that underwent FMRI, data from 10 participants were excluded due to motion or susceptibility artifact or poor image quality. Experimental Design and Tasks Dementia Screening Prior to being enrolled in the FMRI study, participants completed a few brief cognitive tests to establish the presence or absence of dementia or mild cognitive impairment. Mild cognitive impairment (MCI) refers to a transitional state between the cognitive changes of normal aging and Alzheimer’s disease (AD), in which individuals experience memory loss to a greater degree than normal for their age but do not meet criteria for clinically probable AD (Petersen et al., 2001). MCI is a precursor to pathological aging and is a high-risk condition for the development of clinically probable AD. Individuals with MCI progress to clinically probable AD at an accelerated rate

PAGE 78

66 compared to healthy age-matched individuals. Criteria for MCI include: memory complaint, preferably corroborated by an informant, impaired memory function for age and education, preserved general cognitive functioning, intact activities of daily living, and not meeting criteria for dementia (Petersen et al., 2001). The Mini-Mental State Exam (MMSE) was administered to assess for the presence of general cognitive impairments, and individuals who scored below 27 out of 30 were excluded from the study (Folstein et al., 1975). The MMSE is a brief test that screens for cognitive impairment and has been found to be sensitive to the presence of dementia. It consists of a variety of items that assess orientation to time and place, attention and concentration, language, constructional ability, and immediate and delayed recall (Spreen & Strauss, 1998). In order to assess memory function, the Hopkins Verbal Learning Test (HVLT) was administered. The HVLT is a brief verbal list learning test that assesses verbal memory and learning (Shapiro et al., 1999). It consists of a list of 12 nouns with four words drawn from each of three semantic categories. The test involves three learning trials, in which the examiner reads the word list and the examinee responds with as many words as possible from immediate memory, a delayed recall trial and a recognition trial. Individuals who obtain a Total recall score (trials 1-3) of less than 15 were excluded from the study, based on reports that using a cut-off score of 14.5 for Total recall resulted in 87% sensitivity and 98% specificity for dementia (Hogervorst et al., 2002). Assessment of Language Several neuropsychological tests were administered to each participant to assess naming, executive functions of language, and semantic processing. These tests included the Boston Naming Test (BNT), the Delis-Kaplan Executive Functions System (D

PAGE 79

67 KEFS) Verbal Fluency Test, the Test of Language Competence (TOLC-E) Ambiguous Sentences subtest, and the Pyramids and Palm Trees Test. These tests were chosen to examine semantic and executive language functions of the participants and to determine whether group differences exist in these domains independent of neuroimaging results. Boston Naming Test (BNT). The Boston Naming Test is a 60-item test that examines naming ability (Kaplan et al., 1983). It consists of 60 line drawings ranging from simple, high-frequency words to rare words, although the majority of items are low-frequency words. Line drawings are presented one at a time. Participants are instructed to name each picture and may be prompted by a phonemic or semantic cue. Verbal Fluency Test from the Delis-Kaplan Executive Function System (D-KEFS). The Delis-Kaplan Executive Function System (D-KEFS) is a comprehensive battery that assesses verbal and nonverbal executive functions (Delis et al., 2004). The D-KEFS Verbal Fluency Test is comprised of three conditions: Letter Fluency, Category Fluency, and Category Switching. The Letter Fluency task requires the examinee to say words that begin with a specified letter as quickly as possible in three trials of 60 seconds each. For the Category Fluency task, the examinee is required to say words that belong to a designated semantic category as quickly as possible in two trials of 60 seconds each. Lastly, the Category Switching condition provides a means of evaluating an individual’s ability to alternate between saying words from two different semantic categories as quickly as possible for 60 seconds. Overall, the D-KEFS Verbal Fluency Test assesses the ability to generate words fluently according to phonemic and semantic constraints. Ambiguous Sentences subtest from the Test of Language Competence-Expanded Edition (TOLC-E). The Ambiguous Sentences subtest requires multiple interpretations of

PAGE 80

68 sentences and evaluates the ability to recognize and interpret alternative meanings of selected lexical and structural (surface vs. underlying) ambiguities (Wiig & Secord, 1998). This test was found to be sensitive to “frontal” language impairments involving metalinguistics, lexical-semantic manipulation, and language strategy and integration in a group of patients with chronic nonthalamic subcortical lesions who performed within normal limits on standard aphasia batteries (Copland et al., 2000). The Pyramids and Palm Trees Test. The Pyramids and Palm Trees Test assesses the ability to access detailed semantic representations from pictures or words. The test requires an individual to retrieve conceptual information from the test items and perform semantic associations by focusing on specific semantic properties of a target. FMRI Naming Task Participants alternated between an overt picture naming task and a passive viewing task during eight functional imaging runs (Figure 6-1). During the visual naming task, 20 grayscale photographs of animals, 20 grayscale photographs of tools or implements, and 20 grayscale photographs of vehicles were presented. Each photograph was presented twice for a total of 120 naming trials during the scanning session. Photographs were equated for size and resolution and members of each category were matched for frequency of occurrence in the English language (Kucera & Francis, 1967). The photographs were selected based the pilot study (see Chapter 5) that investigated the role of high spatial frequency content (internal details) on object identification. Selected photographs of animals and vehicles did not differ significantly in terms of spatial frequency content necessary for object identification [t(38) = .19, p = .850] but both animals [t(38) = 4.26, p = .000] and vehicles [t(38) = 4.10, p = .000] differed significantly from tools, such that the chosen tools require significantly greater high spatial frequency

PAGE 81

69 content than animals or vehicles for identification. Incorporation of vehicles was chosen to experimentally examine category and attribute in semantic organization, since vehicles share a visual attribute with animals (global form) but are categorically similar to tools (nonliving). By choosing the stimuli on an empirical basis we ensure that the degree to which animals and vehicles are identified on the basis of global form is equivalent, whereas identification of tools is more dependent on local details. A training phase preceded the experimental phase in which different photographs were used. Pictures were presented one at a time for 3400 ms each, in an event-related format, and participants named the picture aloud. An event-related design was chosen to allow for overt responding so that performance accuracy could be assessed. In order to interpret differences in patterns of functional activity in older and younger adults, it is necessary to relate functional activity to behavioral performance (D’Esposito et al., 2003). Between trials, participants were instructed not to think any words to themselves, to rest quietly, and to look at abstract patterns derived by pixelating photographs from the three categories using Adobe PhotoShop 7.0. Pixelating a line drawing clumps pixels into a solid grayscale color in polygon shapes of a predetermined size, in this case 40 pixels.

PAGE 82

70 Figure 6-1. Sample portion of a pseudorandomized event-related run alternating picture naming with passive viewing of pixilated images. Intertrial intervals were pseudorandomly varied between 13600 ms (8 images), 15300 ms (9 images), 17000 ms (10 images) and 18700 ms (11 images) to mitigate low frequency artifacts (Zarahn et al., 1997) and allow the hemodynamic response to return to normal before the participant speaks again, preventing contamination of the latter part of the hemodynamic response by movement during the subsequent response. Experimental runs began and ended with a rest interval. There were 15 trials in an experimental run (5 trials from each category), and 8 runs were administered for a total of 120 trials. Each 15-trial run was 323s in length and acquired 188 functional images for each slice. Stimuli were projected onto a translucent screen above the participant’s head via the Integrated Functional Imaging System (IFIS) using E-Prime Version 1 software. Overt verbal responses were monitored using a bidirectional dual microphone (Commander XG, Resonance Technology, Inc.). Microphone output was run through the penetration panel and connected to a Solo 2500 LS Laptop Computer (Gateway) with Cool Edit software in the scanner control room that recorded verbal responses from each scanning

PAGE 83

71 run. These responses were scored for accuracy and reaction time off-line (Gaiefsky et al., 2002). The picture naming task was chosen for the following reasons: (1) to maintain consistency with previous research (Chao et al., 1999), (2) employ a task that probes both semantic representation, particularly the functioning of the fusiform gyrus, and strategic processes involved in accessing semantic information, i.e., the functioning of the left prefrontal cortex (Chao et al., 1999; Murtha et al., 1999; Thompson-Schill et al., 1997), and (3) examine word-finding performance in older adults. The control task was also chosen to maintain consistency with previous research (Chao et al., 1999). Viewing of pixelated pictures controls for basic visual processing without engaging semantic processes of interest for the current study. The use of perceptual tasks during control states is recommended in order to interrupt the ongoing neural activity during rest states that may involve the same brain regions activated during word retrieval (Binder et al., 1999). FMRI Motor Task Characteristics of the blood oxygenation level-dependent (BOLD) signal may differ between younger and older adults due to changes in neurovascular coupling as a result of vascular compromise in older adults. This presents a potential confound when investigating age-related differences in cognitive activity based on the BOLD response. In order to reduce the likelihood of misattributing aging effects on the BOLD response to changes in cognitive processing, a motor task was used to serve as a control measure of the effect of aging on the BOLD response. We adapted a finger flexion-extension task that has shown a positive linear relationship between movement rate and FMRI signal change in the primary motor cortex (Rao et al., 1996). The modified motor task involves

PAGE 84

72 pressing a button with the index finger of the right hand three times per event in synchronization with a visually presented flashing star. The finger tapping rate was set at one tap every 566 ms during a 1700 ms active period. During the baseline task, participants viewed a static red star. Four runs were administered for a total of 28 activation events. Movements were performed for 1700 ms (activation) followed by variable intervals of rest: 13600 ms (8 images), 15300 ms (9 images), 17000 ms (10 images) and 18700 ms (11 images). The total length of each imaging run was 156 s and 92 images were acquired per run. Image Acquisition Images were acquired on a 3T Allegra head-only scanner (Siemens) using the standard quadrature radio frequency head coil. T1-weighted scout scans were acquired in three planes to determine location of sagittal functional images. The head was aligned such that the interhemispheric fissure was within 1 of vertical. Head motion was minimized using foam padding. Before functional imaging sequences, a time-of-flight MR angiogram (TE = 6.15 ms; TR = 23 ms; FA = 30; FOV = 20 cm; matrix = 256 X 192) was acquired with the same slices used for functional images to check for large vessel effects. Structural images were also acquired prior to functional images for 128 X 1.3 mm thick transverse slices, using a 3D echo planar acquisition (TE = 4.13 ms; TR = 2000 ms; NEX = 1; FOV = 24 cm; FA = 8; matrix size = 256 X 192). Functional images were acquired for the whole brain with 28 4-5 mm thick sagittal slices (TE = 30 ms; TR = 1700 ms; FA = 70; FOV = 240; 1 echo).

PAGE 85

73 Image Analysis Functional images were analyzed and overlaid onto structural images with the Analysis of Functional Neuroimaging (AFNI) program from the Medical College of Wisconsin (Cox, 1996). To minimize the effects of head motion, time series images were spatially registered in 3-dimensional space. The average functional intensity of each acquired slice for every subject was normalized to the mean intensity of all slices for all subjects to control for differences in functional intensity between subjects. Images were visually inspected for gross artifacts and quality control procedures were applied to the data to detect residual motion or susceptibility artifact. As mentioned above, data from 10 of the 50 subjects who participated were eliminated because of motion artifact or poor image quality. All of the remaining 40 subjects demonstrated an adequate hemodynamic response in the left sensory-motor cortex during the finger tapping task for inclusion in group data analyses. Eight imaging runs for the naming task were orthogonalized for linear trends (Birn et al., 2001). Prior to analysis, imaging runs were concatenated into a single time series. For the naming task, hemodynamic response functions (HRFs) for the active task were deconvolved from the baseline state on a voxel-by-voxel basis for the time series according to the images corresponding to stimulus presentation. Deconvolution was continued for 13 images (21.1 seconds) allowing the hemodynamic response to return to baseline. A single HRF was deconvolved separately for each category (e.g., animals, vehicles, tools) and for all responses collapsed across conditions in the naming task. Magnitude of response for each condition was operationally defined as area under the curve of the HRF. Area under the curve was calculated by adding the deconvolved image intensity at each deconvolved time point, with the exception of the first two images, for

PAGE 86

74 each condition in the picture naming task. The first two images following stimulus presentation, during which the participant responded overtly, were excluded to eliminate stimulus-correlated motion artifact (see Carter et al., 2000). Deconvolution estimates the hemodynamic response elicited from the task by empirically determining the waveform and matching it to the data to find the optimum shape (Ward, 1998). Deconvolution utilizes the input stimulus functions (which indicate when the stimuli occurred) and the measured FMRI signal data to estimate the hemodynamic response. Thus, the data determine the functional form of the estimated response. One of the advantages of deconvolution is that the form of the hemodynamic response is empirically determined, with the only a priori assumptions being the time at which the response begins and ends. Because the shape of the hemodynamic response in a given area for a particular task can vary considerably from one individual to another (Aguirre et al., 1998), deconvolution should offer greater sensitivity between subjects. Because deconvolution requires varying intervals of a baseline state between trials, it is also possible to control for cyclical artifacts, such as physiological responses that can occur at the same frequency as experimental task-control state cycles at a constant frequency (Zarahn et al., 1997). Anatomic and functional images were interpolated to volumes with 1 mm3 voxels, co-registered, and converted to stereotaxic space of Talairach and Tournoux (1988) using AFNI. HRFs within the 1 mm3 voxels were spatially smoothed using a 3 mm full-width half-maximum (FWHM) Gaussian filter to compensate for variability in structural and functional anatomy across participants.

PAGE 87

75 Group Analyses of the Naming Task A priori analyses were tested with a statistical threshold of p < .005 and a contiguity threshold of volume > 200 l. These analyses included a 2 (group) x 3 (category) repeated measures ANOVA to examine the effect of age and the interaction of age and semantic category on the pattern of brain activity in frontal and inferior temporal cortices associated with word retrieval during the naming task. Area under the HRF curve was the dependent variable. To compare semantic category, pairwise t-tests were performed for animals versus tools, vehicles versus tools, and animals versus vehicles to examine differences in intensity of the HRF between categories in the fusiform gyrus. When a cluster of significant activity was found, a mask consisting of the voxels from the cluster was constructed and applied to functional images of the individual subjects such that a single average time course for the cluster was extracted for each subject. These averaged time courses were derived to determine the nature of differences in hemodynamic responses, which is especially important in comparisons when age is a factor. Repeated measures ANOVAs were then performed to examine differences in the time course of the HRF within brain regions for the comparisons of age and category. All other comparisons in regions not hypothesized on an a priori basis to be involved in word retrieval differences were conducted at p < .001 and a contiguity threshold of volume > 200 l with the exception that the main effect of category was tested with a statistical probability threshold of p < .0001 in order to reduce the amount of significant volumes to an interpretable number. Behavioral Analyses Student’s t-tests were conducted to compare naming accuracy and response time between the older and younger adults on the overt FMRI picture naming task. Within

PAGE 88

76 subjects repeated measures ANOVAs were performed to compare accuracy and response time for animals, tools, and vehicles independent of age group. Group comparisons of performance on tests of memory (HVLT) and language (BNT, D-KEFS Verbal Fluency, Pyramids and Palm Trees Test, Ambiguous Sentences) were conducted using student’s t-tests. Repeated measures ANOVAs were conducted to examine performance across trials on the HVLT and across time intervals on the D-KEFS Verbal Fluency test between older and younger adults.

PAGE 89

CHAPTER 7 RESULTS Behavioral Results Naming Performance during FMRI Performance on the overt naming task during FMRI was compared between the 20 older and 20 younger adults using student’s t-tests for accuracy and response time. Accuracy rates for the younger (92%) and older subjects (89%) were similar [t(38) = 1.71, p = .096] (Table 7-1). However, response times were significantly slower for older adults (1669 ms) compared to younger adults (1421 ms) [t(38) = 3.32. p = .002]. Additionally, there was an interaction between age and accuracy for each category [F(2, 76) = 4.10, p = .020]. Follow-up pairwise t-tests reveal that older adults responded less accurately for animals than younger adults [t(38) = 3.25, p = .002]. However, accuracy did not differ as a result of category when performance was collapsed across participants [F(2, 78) = 2.74, p = .071]. Additionally, there was a significant interaction between age and category response time [F(2, 76) = 7.45, p = .001]. Follow-up pairwise t-tests reveal that older adults responded more slowly to animals [t(38) = -4.15, p = .000] and vehicles [t(38) = -3.68, p = .001] than younger adults. Similarly, response time differed significantly between categories when performance was collapsed across subjects [F(2,78) = 19.00, p = .000]. Follow-up pairwise t-tests reveal that participants responded more quickly to tools than vehicles [t(38) = 4.26, p = .024] and responded more quickly to vehicles than animals [t(38) = 2.36, p = .024]. Since there were no significant differences in accuracy rates between older and younger adults across category, 77

PAGE 90

78 differences in FMRI activity for age more likely represent age-related changes in the neural substrates underlying word retrieval rather than differences in performance. Similarly, since naming accuracy did not differ for animals, tools, or vehicles collapsed across subjects, level of performance is unlikely to have influenced FMRI comparisons of semantic category. Table 7-1. Performance on the FMRI Naming Task for Younger and Older Adults. Total Animals Tools Vehicles M SD M SD M SD M SD Younger 92.0 (5.6) 92.9 (6.3) 93.0 (6.0) 91.8 (7.2) Accuracy (%correct) Older 89.0 (5.4) 85.2 (8.5) 91.2 (7.1) 90.8 (6.1) Younger 1421 (2544) 1449 (2565) 1385 (3180) 1429 (2267) Response Time (ms) Older 1669 (2172) 1772 (2360) 1531 (2259) 1696 (2329) Neuropsychological Testing Performance Performance on the memory and language measures was compared between older and younger adults. Older adults recalled significantly fewer words on both the Hopkins Verbal Learning Test immediate recall [Total score: t(38) = -5.01, p = .000] and delayed recall [Delay score: t(38) = -3.49, p = .001]. Results from a 2 (group) x 4 (trial) repeated measures ANOVA for HVLT performance indicate a significant effect for trial [F(3,57) = 92.61, p = .000], suggesting that word recall improves across trials 1-3 and decreases on the delay trial. However, the interaction of age and trial on memory performance was not significant [F(3,57) = .22, p = .879], suggesting that the pattern of list-learning does not change with age (Table 7-2). In contrast to previous findings, there was no significant age-related difference in naming accuracy on the Boston Naming Test [t(38) = -0.19, p = .854]. Older adults

PAGE 91

79 performed significantly worse on the TOLC-E Ambiguous Sentences subtest than the younger adults [t(38) = -2.80, p = 0.008]. However, older adults performed significantly better on the Pyramid and Palm Trees Test than younger adults [t(38) = 3.13, p = .003], although the average performance of the groups differed by only 2 points. On the D-KEFS Verbal Fluency Test, older adults generated significantly fewer category members overall than younger adults [Category Fluency: t(38) = -3.84, p = .000] but the total number of words generated to three lexical cues (i.e., begins with letter “F”) did not differ between younger and older adults [Lexical Fluency: t(38) = -1.86, p = .071]. Results from a 2 (group) x 4 (time interval) repeated measures ANOVA for lexical retrieval indicate a significant within subject difference for trial [F(3,57) = 107.41, p = .000] suggesting that word generation to lexical cues decreases over time. No significant interaction of age and trial [F(3, 57) = 0.94, p = .425] was found on lexical fluency, suggesting that the pattern of word retrieval does not change with age. Results from a 2 (group) x 4 (time interval) repeated measures ANOVA for category generation also indicate a significant within subject difference for trial [F(3,57) = 157.90, p = .000], with a decreasing pattern similar to lexical retrieval. No age by interval of category generation interaction was found [F(3,57) = 1.10, p = .338]. On the category switching task, which requires alternating exemplar generation between two categories, older adults performed significantly worse on Total Accuracy [t(38) = 3.93, p = .000]. Of note, older adults demonstrated greater repetition errors [t(38) = 2.49, p = .017] and set loss errors, although the difference in set loss errors did not reach significance [t(38) = -1.66, p = .106], across all word generation tasks (lexical and semantic retrieval and category switching).

PAGE 92

80 Table 7-2. Raw Scores on the Neuropsychological Tests for Younger and Older Adults. Younger Adults (n=20) Older Adults (n=20) Measures M SD M SD p-valuesa Global Cognition Mini Mental State Exam 29.30 (0.92) 28.55 (1.39) Learning and Memory Hopkins Verbal Learning Test Immediate Total Recall 29.65 (2.78) 24.25 (3.95) .000 Delayed Recall 10.65 (1.53) 8.70 (1.98) .001 Language Boston Naming Test (60-item) 55.55 (4.89) 55.30 (3.51) .854 TOLC-E Ambiguous Sentences 34.40 (4.54) 28.90 (7.52) .008 Pyramids and Palm Trees Test 49.60 (1.88) 51.05 (.89) .003 DKEFS Verbal Fluency Test Letter Fluency (FAS Total) 47.65 (11.37) 41.05 (11.11) .071 Category Fluency (Total) 45.45 (6.22) 37.25 (7.25) .000 Total Switching Accuracy 14.7 (2.72) 11.4 (2.72) .000 Total Repetition Errors 2.25 (2.31) 4.65 (3.63) .017 Total Set Loss Errors 1.35 (1.69) 2.25 (1.74) .106 ap-value associated with student’s t-tests comparing younger and older adults Taken together, age-related differences in performance on language tests suggests that older adults have greater difficulty on internally generated tasks that place increased demands on executive functions involved in accessing and manipulating verbal information. In contrast, older adults performed comparably to younger adults on externally constrained language tests that require providing a single word response. These behavioral results suggest that lexical-semantic knowledge is relatively preserved in healthy aging, but higher-order functions involved in selecting, retrieving and manipulating lexical-semantic information may deteriorate with age. FMRI Results All the FMRI analyses for the word retrieval task that have to do with age will be presented first. Once that is accomplished, we will address the effects that are due to category independent of age. A priori comparisons will precede exploratory analyses. A

PAGE 93

81 priori analyses were tested with a statistical threshold of p < .005 and a contiguity threshold of volume > 200 l, and a posteriori analyses were tested with a statistical threshold of p < .001 and a contiguity threshold of volume > 200 l, with the exception of the main effect of category, which was tested with p < .0001 to reduce the number of activated clusters to a level of interpretability. To address Hypothesis 1, which predicts greater left frontal activity during naming in older adults, a comparison of intensity of activity between younger and older adults was performed with a 2 (group) x 3 (category) voxel-wise repeated measures ANOVA with area under the curve (AUC) of the hemodynamic response function (HRF) as the dependent variable. Area under the curve of the hemodynamic response is a measure of the intensity of activity based on the total raw signal change from baseline over a specified number of images following a given stimulus. The baseline is calculated by the deconvolution analysis and represents an estimation of the magnetic resonance signal under the control condition, based on the input stimulus vectors which indicate when the stimuli occurred and the number of images specified to model the hemodynamic response (Kaundinya Gopinath, personal communication March 17, 2004). Region of interest (ROI) analyses were then conducted to examine the temporal characteristics of the HRF by extracting a single average time course in clusters of significant activity between older and younger adults. Derived HRFs are based on the average raw signal change from baseline across subjects. To examine common areas of activity for word retrieval, student’s t-tests were then performed with a dependent variable of AUC for the naming task (collapsed across categories) against the passive viewing baseline for younger and older adults.

PAGE 94

82 To address Hypothesis 2, which predicts that neural substrates in the fusiform gyrus for semantic information differ between older and younger adults, the interaction between age and semantic category was examined resulting from the 2 (group) x 3 (category) repeated measures ANOVA with AUC as the dependent variable. Lastly, to address Hypothesis 3, which predicts that semantic information is represented according to category and visual attribute in the fusiform gyrus, comparisons of each of the categories (animals, tools, and vehicles) directly to one another were performed with pairwise t-tests to examine differences in intensity of the HRF between categories in the fusiform gyrus. ROI analyses were subsequently conducted on clusters of significant activity to examine the characteristics of the HRF in regions of the fusiform gyrus activated by the comparison of categories. To identify areas not a priori hypothesized to be involved in semantic processing, the main effect of category resulting from the 2 (group) x 3 (category) repeated measures ANOVA based on AUC was examined at a statistical threshold of p < .0001. Comparison of Age on Picture Naming The first hypothesis (Hypothesis 1, Chapter 4) stated that activity in the medial and/or lateral frontal cortices of the left hemisphere will be greater in older than younger participants during word retrieval due to compensation for reduced efficiency of functioning. This assumption is based on literature indicating that word retrieval problems in older adults are due to deterioration of executive language processes. The main effect for age resulting from a 2 (group) x 3 (category) repeated measures ANOVA with area under the curve (AUC) as the dependent variable revealed that older adults, when compared to younger adults, showed significantly greater activity in both medial and lateral frontal cortices (Table 7-3). Given the a priori nature of our hypotheses about

PAGE 95

83 the role of the frontal lobes in word retrieval, a threshold volume of 200 l and a statistical threshold of p < .005 were chosen. Greater activity in the older adults was found medially in the left and right rostral cingulate zone ventral to the supplementary motor area (BA 32) and the right anterior cingulate ventral to the genu of the corpus collosum, extending anteriorly to the posterior boundary of the forceps of the corpus collosum. Additionally, older adults showed greater activity compared to younger adults in the posterior portion of the supplementary motor area (BA 6) and motor area (BA 4) bilaterally, although this cluster of activity extended laterally along the right sensory-motor cortex. A more stringent statistical threshold (p < .001) decomposes this cluster into its components (see Table 7-4). Laterally, activity was localized to Broca’s area homologue (BA 45) in the right hemisphere and an anterior region of the right inferior frontal gyrus.

PAGE 96

84 Table 7-3. A Priori Volumes of Tissue in Medial and Lateral Frontal Cortices (> 200 l) Showing Significant Activity during Naming for Older Compared to Younger Adults, p < .005. Older > Younger Region Anatomic Area (peak talairach coordinates) volume in l Medial Frontal Cortex Rostral cingulate zone Supplementary motor area (extending laterally) Anterior cingulate (subgenual) L and R BA 32 (-1, -6, 48) 402 l L and R BA 6, 4 (0, -29, 56) 7209 l R BA 32, 12 (8, 27, -11) 749 l Lateral Frontal Cortex Broca’s area homologue Inferior frontal gyrus R BA 45 (39, 28, -1) 520 l R BA 10 (51, 44, 1) 234 l Note. BA = Brodmann's Area (according to Talairach & Tournoux, 1988), L = Left, R = Right. Given that characteristics of the BOLD signal may differ between individuals due to potential age-related changes in cerebrovascular dynamics and neurovascular coupling, D’Esposito et al. (2003) recommend examining other characteristics of the HRF that may reflect differences in neural activity between younger and older adults in addition to intensity of activity. Therefore, to supplement the comparisons based on area under the curve analyses, we examined the temporal characteristics of the BOLD HRFs in older and younger adults during word retrieval using region of interest (ROI) techniques. A mask of each cluster in the frontal lobes showing differences in activity between younger and older adults at p < .005 was generated and the HRF, based on raw signal change from

PAGE 97

85 baseline, for each participant was derived and then averaged across group. The averaged hemodynamic response functions (HRFs) for each group and cluster were entered into a 2 (group) x 11 (image number) repeated measures ANOVA to investigate differences in the time course of the HRF between older and younger adults. The first two images were excluded from the group x time repeated measures ANOVA to remain consistent with the group x category repeated measures ANOVA for area under the curve, in which we eliminated the first two images to reduce motion artifact. Because of our a priori interest in medial and lateral frontal regions associated with language production in healthy and aphasic individuals, the current discussion focuses on the bilateral rostral cingulate zone and Broca’s homologue in the right hemisphere. Results of the HRF analyses for the other frontal regions are presented in Appendix A. The 2 (group) x 11 (image number) interaction was significant for the bilateral rostral cingulate zone (BA 32) [F(10, 350) = 4.88, p = .000] and Broca’s homologue in the right hemisphere (BA 45) [F(10, 380) = 6.97, p = .000] (see Figure 7-1). A visual examination of the response curves reveals different time courses between groups. In both regions, the average HRF of the older adults exhibits a longer rise to peak amplitude and a greater delay in returning to baseline than for the younger adults. In the bilateral rostral cingulate zone, the amplitude of signal change for the average HRF in older and younger adults is similar, but the younger adults demonstrate a greater undershoot. However, in right Broca’s area homologue, the amplitude of signal change for the average HRF for the older adults is considerably larger than for the younger adults, but the younger adults exhibit a larger undershoot.

PAGE 98

86 Figure 7-1. Regions of the a) medial (rostral cingulate zone) and b) lateral (Broca’s homologue in the right hemisphere) frontal cortex activated by older adults relative to younger adults during picture naming, red = p < .005; yellow = p < .001, along with corresponding hemodynamic response functions (HRFs) for older and younger adults for these areas. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) group difference was found based on follow-up student’s t-tests. Although not hypothesized a priori, older adults showed greater activity compared to younger adults in the right sensory-motor cortex involving the precentral gyrus and postcentral gyrus at a statistical threshold of p < .001. A significant difference in activity was also found in the right posterior perisylvian region including the superior temporal gyrus at p < .001 (see Table 7-4).

PAGE 99

87 Table 7-4. A Posteriori Volumes of Tissue (> 200 l) Showing Significant Activity during Naming for Older Compared to Younger Adults, p < .001. Older > Younger Region Anatomic Area (peak talairach coordinates) volume in l Sensory-Motor Cortex Precentral gyrus Postcentral gyrus R BA 4 (56, -9, 31) 285 l R BA 4 (37, -16, 59) 260 l R BA 4 (18, -22, 64) 258 l R BA 3 (33, -31, 56) 262 l Perisylvian Cortex Superior temporal gyrus L BA 22, 41, 42 (-55, -11, -3) 265 l Note. BA = Brodmann's Area (according to Talairach & Tournoux, 1988), L = Left, R = Right. In addition to the between-group comparisons previously mentioned, withingroup (older, younger) comparisons of intensity of activity (AUC) for picture naming to baseline passive viewing were conducted with student’s t-tests. Within-group analyses of word retrieval were performed to investigate common areas of activity for each group. Naming was collapsed across category to gain power in identifying areas involved in word retrieval for each group. Since the main effect of age from the 2 (group) x 3 (category) repeated measures ANOVA only reveals regions in which the intensity of activity differs between older and younger adults, within-group t-tests of picture naming compared to baseline provide more information about areas activated during word

PAGE 100

88 retrieval for each group. Compared to a baseline of passive viewing, younger and older adults demonstrated similar activity during picture naming in Broca’s area and the left sensory-motor strip at a threshold volume of 200 l and a statistical threshold of p < .005 (Figure 7-2). An examination of relative volume size in the frontal regions activated by both groups reveals that the extent of activity in Broca’s area was comparable (990 l for older adults and 909 l for younger adults) although activity for older adults extends more anteriorly compared to younger adults. In contrast, older adults activated the left sensory-motor strip to a substantially greater extent than younger adults, whose activity was limited to the left precentral gyrus (11481 l for older adults and 896 l for younger adults). The involvement of Broca’s area in older and younger adults during picture naming is not surprising, given the role of this region in language production, and was not detected by the 2 (group) x 3 (category) repeated measures ANOVA because this region was activated to a similar degree in both groups. Figure 7-2. Activity in left frontal cortex, including Broca’s area, for older adults (red) and younger adults (blue) during picture naming compared to baseline (overlap = green, p < .005, cluster > 200 l). Comparison of the Interaction between Age and Semantic Category The second hypothesis (Hypothesis 2, Chapter 4) stated that activity in the posterior and anterior inferior temporal cortex for older participants will be significantly more

PAGE 101

89 diffuse and/or less lateralized than for younger persons for semantic category. This assumption is based on literature that suggests that word retrieval difficulties in older persons are due to deterioration of information in lexical-semantic stores. A 2 (group) x 3 (category) voxel-wise repeated measures ANOVA with AUC as the dependent variable did not yield a significant interaction effect at a statistical threshold of p < .005 and a cluster volume threshold of 200 l in the inferior temporal cortex. The lack of an interaction effect in the inferior temporal cortex suggests that the inferior temporal substrates for the representation of semantic information remain largely intact in healthy aging and do not likely contribute to word retrieval problems in older adults. Since no group x category differences were found in the inferior temporal cortex, we collapsed across groups for semantic comparisons. Furthermore, there were no areas that met the exploratory statistical threshold of p < .001 outside the inferior temporal lobe for the interaction of age and category. Direct Comparisons of Semantic Category The third hypothesis (Hypothesis 3, Chapter 4) states that there will be dissociations within the inferior temporal cortex, particularly the fusiform gyrus, for processing category and visual attributes of objects. More specifically, the hypotheses are that in the posterior fusiform gyrus (where analysis of features takes place), processing of semantic categories will be greater in the right hemisphere for processing global visual information (animals and vehicles) compared to local information, whereas activity will be greater in the left hemisphere for processing local visual information (tools) compared to global form. In the anterior fusiform gyrus (where object identity is finally resolved), medial vs. lateral activity bilaterally will be determined by the

PAGE 102

90 distinction between living vs. nonliving things. In the anterior inferior temporal cortex bilaterally, animals (living) will activate the lateral portion of the fusiform gyrus more than tools or vehicles (nonliving), whereas tools and vehicles will activate the medial portion of the fusiform gyrus bilaterally more than animals. Since there was no interaction between age and semantic category in the inferior temporal cortex, we interpret the main effect for semantic category across all participants. To elucidate differences in the intensity of activity between categories within the fusiform gyrus, paired t-tests were conducted to compare animals vs. tools, animals vs. vehicles, and tools vs. vehicles. Given the a priori nature of this investigation, a statistical threshold of p < .005 and a cluster threshold volume of 200 l was used (Table 7-5). For the comparison of animals vs. tools, a significant cluster of activity for animals greater than tools was found in the right hemisphere in the lateral fusiform gyrus (Figure 7-3). A cluster of significant activity for tools greater than animals was lateralized to the left hemisphere in the medial fusiform gyrus (Figure 7-3). For the comparison of tools vs. vehicles, clusters of significant activity for vehicles greater than tools were found in the medial fusiform gyrus, with two large clusters extending from the parahippocampal gyrus along the collateral sulcus through the fusiform gyrus bilaterally, and a smaller cluster located in the posterior medial fusiform gyrus of the left hemisphere (Figure 7-4). For the comparison of animals vs. vehicles, there were no significant clusters of activity for animals greater than vehicles in the inferior temporal cortex. Clusters of significant activity for vehicles greater than animals extended along the medial fusiform gyrus bilaterally including the parahippocampal gyrus and collateral sulcus (Figure 7-5).

PAGE 103

91 Table 7-5. A Priori Volumes of Tissue in the Inferior Temporal Cortex (> 200 l) Showing Differential Responses to Animals, Tools, and Vehicles for Pairwise Comparisons, p < .005. Region Selectivity Hemisphere Peak talairach coordinates (x, y, z) Volume Inferior Temporal Cortex Lateral fusiform gyrus A > T R (36, -42, -20) 558 l Medial fusiform gyrus T > A L (-25, -59, -11) 332 l V > T L (-25, -74, -4) 341 l V > T R (25, -36, -8) 4414 l V > T L (-28, -39, -8) 3069 l V > A R (31, -37, -9) 13329 l V > A L (-27, -41, -9) 12589 l Note. A = Animals, T = Tools, V = Vehicles, L = Left, R = Right. To determine the time course of differences seen in the fusiform gyrus for semantic representation of animals, tools and vehicles, a mask of each cluster of activity resulting from the paired t-tests was generated and the HRF for each category and participant was derived and then averaged across subjects for each category. The averaged HRFs for each category were entered into a 2 (category) x 11 (image number) repeated measures ANOVA. A visual examination of the time course for each category in the seven clusters in the fusiform gyrus reveals similar shapes across category, although amplitudes vary between categories. Significant category x time interactions were found in each cluster of activity. For the HRF analyses, the 2 (category) x 11 (time) interaction was significant for the comparison of animals greater than tools in the right lateral fusiform gyrus [F(10, 390) = 10.57, p = .000], for the comparison of tools greater than animals in the left medial fusiform gyrus [F(10, 380) = 2.78, p = .003], for the comparison of vehicles greater than tools in the left posterior medial fusiform gyrus [F(10, 380) = 10.68, p =

PAGE 104

92 .000] and bilateral medial fusiform gyrus extending anteriorly to the collateral sulcus and parahippocampal gyrus [left: F(10, 390) = 17.01, p = .000; right: F(10, 390) = 28.02, p =.000], and for the comparison of vehicles greater than animals in the bilateral medial fusiform gyrus [left: F(10, 390) = 16.07, p = .000; right: F(10, 390) = 14.89, p =.000]. Figure 7-3. Lateral and medial fusiform gyrus activated by the comparison of animals and tools, red = p < .005; yellow = p < .001 for animals greater than tools, dark blue = p < .005; light blue = p < .001 for tools greater than animals, along with corresponding hemodynamic response functions (HRFs) for animals and tools for these areas. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) category difference was found based on follow-up paired t-tests.

PAGE 105

93 Figure 7-4. Activity in the fusiform gyrus bilaterally for the comparison of vehicles to tools, dark blue = p < .005; light blue = p < .001 for vehicles greater than tools, red = p < .005; yellow = p < .001 for tools greater than vehicles, along with corresponding hemodynamic response functions (HRFs) for vehicles and tools for these areas. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) category difference was found based on follow-up paired t-tests.

PAGE 106

94 Figure 7-5. Activity in the fusiform gyrus bilaterally for the comparison of vehicles to animals, dark blue = p < .005; light blue = p < .001 for vehicles greater than animals, along with corresponding hemodynamic response functions (HRFs) for vehicles and animals for these areas. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) category difference was found based on follow-up paired t-tests. Taken together, a medial-lateral differentiation for tools and animals, respectively, was found in the fusiform gyrus, consistent with previous findings (Chao et al., 1999, Ishai et al., 2001). When animals and tools were compared, animals showed greater activity in the right lateral fusiform gyrus whereas tools showed greater activity in the left medial fusiform gyrus. An anterior-posterior distinction between category and attribute was not found in the fusiform gyrus; rather, lateralization according to visual attribute was not limited to only posterior regions. Somewhat surprising is the finding that activity for vehicles dominates the fusiform gyrus in both the left and right hemispheres when

PAGE 107

95 compared to either animals or tools. However, activity for vehicles compared to animals and tools tends to reside within the medial aspect of the fusiform gyrus bilaterally. The bilateral medial activity for vehicles may represent both categorical and attribute processing. Bilateral activity suggests that vehicles are an intermediary category between animals and tools in terms of the amount of global form or local details needed for identification, despite our attempt to experimentally manipulate the categories to equate animals and vehicles. Yet, since vehicles are nonliving, activity of the medial fusiform is consistent with previous findings for category. Overall, these findings suggest that we must revise our model of visual semantic processing. A simpler model can be proposed. Visual attributes are clearly an important factor in the organization of semantic information and are lateralized in the brain according to global form (right hemisphere) and local details (left hemisphere). Since processing visual attributes is not restricted to the posterior portion of the fusiform gyrus, but occurs throughout the fusiform gyrus, this information likely assists with object identification, and may be processed in more of an interactive manner throughout the course of object recognition than originally thought. An exploratory analysis of other areas of activity not specified a priori for the main effect of category resulting from the 2 (age) x 3 (category) repeated measures ANOVA reveals clusters of activity in the left inferior parietal lobe and angular gyrus, left posterior middle temporal gyrus, right middle occipital gyrus, and right inferior frontal sulcus at p < .0001 (see Table 7-6, Figure 7-6). A more stringent p-value was chosen for this comparison to reduce the amount of activated clusters to an interpretable number. A table of the 21 clusters of activity not a priori hypothesized to be involved in semantic processing at p < .001 is included in Appendix B. The large number of activated regions

PAGE 108

96 at p < .001 demonstrates that categories are a robust source of differences in cortical activity. To examine differences in HRF signal characteristics across time between the three categories in the five identified clusters, ROI analyses were performed on each cluster, and averaged HRFs for each category (animals, tools, vehicles) were entered into a 3 (category) x 11 (image number) repeated measures ANOVA. A visual examination of the time course for each category in each cluster reveals similar shapes across category, although amplitudes differ significantly between categories. Significant category x time interactions were found in each cluster of activity. Tools show greater activity than vehicles and animals in the left inferior parietal lobe [F(20,780) = 6.62, p = .000] and left middle temporal gyrus [F(20,780) = 7.67, p = .000]. Animals and vehicles elicit greater activity than tools in the right inferior frontal sulcus [F(20,780) = 2.38, p = .001]. Activity for vehicles is greater than animals and tools in the right middle occipital gyrus [F(20,780) = 9.75, p = .000]. Dominance of activity for tools in the left inferior parietal lobe likely represents knowledge of the skilled motor movements needed to use such objects (i.e., praxis). Activation of the inferior frontal sulcus to a greater degree by animals and vehicles may be due to input from the ventral visual stream, since this area has been implicated in object working memory (Belger et al., 1998).

PAGE 109

97 Table 7-6. A Posteriori Volumes of Tissue (> 200 l) Showing Differential Responses to Animals, Tools, and Vehicles, p < .0001. Region Selectivity Hemisphere Peak talairach coordinates (x, y, z) Volume Lateral Temporal Cortex Middle temporal gyrus T > V > A L (-51, -55, -2) 1344 l Parietal Cortex Inferior parietal lobe T > V > A L (-57, -33, 35) 2157 l Angular gyrus T > V > A L (-27, -72, 30) 257 l Occipital Cortex Middle occipital gyrus V > A > T R (32, -85, 19) 479 l Frontal Cortex Inferior frontal sulcus A = V > T R (40, 14, 27) 278 l Note. A = Animals, T = Tools, V = Vehicles, L = Left, R = Right. Figure 7-6. Activity in the a) right inferior frontal sulcus, b) left inferior parietal lobe and left middle temporal gyrus associated with the main effect of category across participants, red = p < .0001; yellow = p < .00001. Notably, even in regions outside the inferior temporal cortex, areas that respond maximally to tools are lateralized to the left hemisphere, whereas areas in which the hemodynamic response for animals is greater or equivalent to either tools or vehicles are lateralized to the right hemisphere. With the exception of the fusiform gyrus and middle occipital gyrus, areas in which activity for vehicles is greater than animals are lateralized

PAGE 110

98 to the left hemisphere, suggesting that local details are more important for the identification of vehicles than animals. In contrast, areas in which activity for vehicles is greater than tools are lateralized to the right hemisphere, suggesting that global form is more important for the identification of vehicles than tools. This is consistent with our pilot data that suggest that vehicles may represent an intermediary category that share visual properties with both animals and tools. Furthermore, this provides additional support for the role of features in the organization of semantic information. Although the lateral temporal lobe was not specified as a region of interest in our original hypotheses, previous literature exists regarding its role in category-specific effects. Because distinctions in the lateral temporal lobe have been made for knowledge of movement in experiments similar to ours (Beauchamp et al., 2003; Chao et al., 1999), we relaxed our statistical threshold to p < .005 to explore the possibility of activity related to the attribute of object movement in the lateral temporal lobe. In particular, Beauchamp et al. (2003) and Chao et al. (1999) report a dissociation between animate and inanimate motion that breaks down according to superior and inferior regions of the lateral temporal lobe, respectively. Comparisons of each of the categories (animals, tools, and vehicles) directly to one another were performed with pairwise t-tests to examine differences in intensity of the HRF between categories in the lateral temporal lobe (see Table 7-7). For the comparison of animals vs. tools, a significant cluster of activity for animals greater than tools was found in the right posterior middle temporal gyrus (MTG). Clusters of significant activity for tools greater than animals were found in the right inferior temporal gyrus and left posterior MTG (Figure 7-7). For the comparison of tools vs. vehicles, a cluster of significant activity for tools greater than vehicles was found in the left MTG.

PAGE 111

99 Activity greater for vehicles than tools was found in the right lateral MTG bordering on the angular gyrus. The left MTG and right posterior inferior temporal gyrus were also activated for vehicles compared to animals. Table 7-7. A Posteriori Volumes of Tissue in the Lateral Temporal Lobe (>200 l) Showing Differential Responses to Animals, Tools, and Vehicles for Pairwise Comparisons, p < .005. Region Selectivity Hemisphere Peak talairach coordinates (x, y, z) Volume Lateral Temporal Lobe MTG A > T R (42, -69, 2) 1028 l T > V L (-43, -66, 1) 567 l MTG/ITG T > A L (-50, -57, -6) 5058 l V > A L (-52, -53, -2) 1977 l MTG/angular gyrus V > T R (41, -79, 21) 2274 l ITG T > A R (57, -54, -10) 265 l V > A R (48, -53, -7) 285 l Note. MTG = middle temporal gyrus, ITG = inferior temporal gyrus, A = Animals, T = Tools, V = Vehicles, L = Left, R = Right. Figure 7-7. Left and right middle temporal gyrus activated for the comparison of animals to tools, red = p < .005; yellow = p < .001 for animals greater than tools; dark blue = p < .005; light blue = p < .001 for tools greater than animals. These findings are relatively consistent with Beauchamp et al. (2003); however, the superior-inferior distinction appears to occur more ventrally in our study. Beauchamp et

PAGE 112

100 al. (2003) report a dissociation between animate movement in the superior temporal sulcus and inanimate movement in the middle temporal gyrus. In contrast, we found a dissociation between animals in the middle temporal gyrus and tools in the inferior temporal gyrus, although this inferior-superior distinction was only found in the right hemisphere because activity for animals is entirely lateralized to the right hemisphere. Both tools and vehicles show greater activity than animals in the left MTG/ITG region and right ITG. However, tools show greater activity than vehicles in the left MTG and vehicles show greater activity in the right MTG compared to tools. Taken together, these findings suggest that visual features likely influence activity in the middle temporal gyrus for animals, tools, and vehicles. In contrast, the left and right inferior temporal gyrus is selective for nonliving categories and possibly represents inanimate motion. The degree to which the lateralization of these findings results from visual attributes rather than functional attributes (type of movement) is difficult to determine. Although the lateral temporal cortex is traditionally thought to represent object movement, current findings suggest that distinctions between the attributes of object movement and visual features, such as global form or local details, may be necessary to understand the role of this region in representing semantic information.

PAGE 113

CHAPTER 8 DISCUSSION Frontal-Executive Changes in Word Retrieval with Age A major finding of the current study is that older adults demonstrate a larger frontal network of activity during word retrieval than younger adults. Not only do older adults show greater activity in left hemisphere regions traditionally associated with language (i.e., bilateral rostral cingulate zone and SMA), but they demonstrate less lateralization of activity characterized by increased right frontal activity. In particular, older adults show greater activity in Broca’s homologue (BA 45) in the right hemisphere, an anterior region of the right inferior frontal gyrus, and the right anterior cingulate. Increased frontal activity in older adults compared to younger adults provides support for our first hypothesis, which states that activity in the medial and/or lateral frontal cortices of the left hemisphere will be greater in older relative to younger adults, but extends differences in activity to include the right hemisphere. This finding suggests that word retrieval difficulties in older adults are due to a deterioration of the neural substrates for executive language processes. Given that both medial (bilateral rostral cingulate zone and SMA, and right anterior cingulate) and lateral (right inferior frontal gyrus) regions showed increased activity in older adults, we are unable to specify the degree to which executive functions associated with these regions (i.e., initiation and selection of competing responses vs. retrieval of semantic information) contribute to the changes seen in older adults. Further research is needed to more clearly distinguish between the role of initiation and selection in word retrieval difficulties experienced by older adults. Given 101

PAGE 114

102 that Crosson et al. (2001) reported a shift in activity from medial (pre-SMA) to lateral (inferior frontal sulcus) regions based on internally vs. externally constrained word generation tasks in younger adults, it is possible that the bilateral rostral cingulate zone activity in older adults reflects a change in the role that external constraint has on naming associated with aging. In other words, perhaps picture naming requires greater internal generation in older adults, similar to the level of internal generation needed during word generation in younger adults. Notably, both the younger and older adults activated Broca’s area to a similar degree, indicating that while older adults engage additional right hemisphere regions, they continue to rely on traditional language regions of the left hemisphere. Although the increased right frontal activity in the older adults was not predicted, the decreased lateralization seen in the older adults is consistent with Cabeza’s (2001) Hemispheric Asymmetry Reduction in Old Adults (HAROLD) model. Recruitment of contralateral areas in the right hemisphere may reflect a compensatory mechanism for reduced efficiency of functioning. This interpretation is supported by the fact that older and younger adults did not differ in accuracy of naming during the FMRI experiment, but older adults took longer to respond, suggesting increased engagement in search strategies. Several studies have reported that additional bilateral activity in the prefrontal cortex for older adults corresponds to higher performance in older adults (Cabeza, 2002; Cabeza et al., 2002; Reuter-Lorenz et al., 2000; Reuter-Lorenz et al., 2002; Rosen et al., 2002). Engagement of neural substrates of executive functions in the right hemisphere may assist in resolving difficult task demands. Additionally, studies of language rehabilitation post-left CVA (cerebral vascular accident) suggest that the right hemisphere’s ability to

PAGE 115

103 perform functions previously localized to the left hemisphere assists in the restoration of language ability (Cao et al., 1999; Crosson et al., under review; Heiss et al., 1999; Richards et al., 2001). An alternative interpretation of increased right frontal activity in aging suggests that perhaps older adults are less able to use the right basal ganglia to suppress right frontal activity during language production and the right frontal activity interferes with word production. The only area of activity elicited by the interaction of age and category resulting from the 2 (group) by 3 (category) repeated measures ANOVA was a small area (337 l) in the body of the right caudate nucleus (Maximum Intensity xyz = 22, -7, 28) at p < .005. The HRF analysis indicates that younger adults demonstrate a greater amplitude of response for all three categories compared to older adults in this region [main effect of group: F(1, 35) = 4.11, p = .050]. Although the activity in the right caudate was below our established criterion for a posteriori analyses, previous literature suggests that it is difficult to show reliable activity in the caudate and thalamus (Crosson et al., 2003; Ojemann et al., 1998; Palmer et al., 2001; Rosen et al., 2000; Warburton et al., 1996). The reason for decreased reliability in these structures is uncertain, but sensitivity in this subcortical area appears more limited than in cortex. There may be tasks that activate the caudate more strongly than naming. However, activity of the right caudate nucleus may explain differences in frontal activity between older and younger adults. Decreased right caudate activity in the older adults for picture naming can be interpreted in light of recent neuroimaging data on word generation. Crosson et al. (2003) described a loop consisting of the dominant (left) pre-SMA, dorsal caudate

PAGE 116

104 nucleus, and ventral anterior thalamus that is involved in generating words, but not in generating nonsense syllables. They concluded that the dominant pre-SMA—dorsal caudate—ventral anterior thalamic loop is involved in biasing selection toward a particular pre-existing lexical representation during word generation when multiple choices are possible. Crosson et al. (2003) also found robust activity in the nondominant (right) basal ganglia in the absence of significant nondominant frontal activity during word generation but not during nonsense syllable generation. These data were interpreted as evidence that the right basal ganglia mediate suppression of right frontal activity by the left pre-SMA during word generation to prevent interference with left-hemisphere processes. The proposed roles of the basal ganglia (i.e., maintaining response biases and suppressing right frontal activity) are consistent with conceptualizations of the basal ganglia as facilitating desired and suppressing undesired behaviors (e.g., Gerfen, 1992; Mink, 1996; Penny & Young, 1986). Although the role of the pre-SMA—dorsal caudate—ventral anterior thalamic loop has not been verified in object naming, the decrease in right caudate activity for older adults as compared to younger adults may result in reduced suppression of right frontal activity and therefore contribute to the increase in right frontal activity in older adults during the naming task. Despite Crosson et al.’s (2003) position that the right basal ganglia’s suppression of right frontal activity serves to prevent right frontal structures from interfering with language production, whether this process serves the same function has not been studied in aging. In fact, reduced suppression of right hemisphere activity may be adaptive in aging. The role of the nondominant basal ganglia may change with age to allow the expression of right frontal structures to compensate for reduced functioning. In spite of

PAGE 117

105 these alternative interpretations, the question remains whether the right hemisphere contributes to word finding or whether right hemisphere activity reflects problems with focusing attention to select a word and a failure of right hemisphere suppression that underlies word retrieval problems. Differences in performance between older and younger adults on the neuropsychological language tasks provide additional support for the deterioration of executive functions in older adults. Compared to younger adults, older adults performed significantly worse on the lexical fluency task and category switching task in the D-KEFS Verbal Fluency Test and on Ambiguous Sentences from the TOLC-E. Both tasks require the ability to search lexical-semantic stores and maintain and manipulate lexical-semantic information online. The increase in repetition errors and set loss errors across all word generation tasks (lexical and semantic retrieval and category switching) suggests difficulty with cognitive control. Additionally, older adults demonstrated a decreased ability to learn new information (as assessed by the HVLT). Taken together, performance on these measures may reflect global differences in the neural substrates of frontally-mediated cognitive processes that are not restricted to executive functions of language but include learning and memory. Given that neuroanatomical correlates of aging are most pronounced in the frontal lobe, changes in functional activity during picture naming in frontal structures are not surprising and lend support to the frontal hypothesis of aging (West, 1996). The ability to discriminate changes in activity that represent differences in cognitive processes from changes in activity that represent physiological changes due to aging is a challenge for FMRI research, especially since neuroanatomical changes of

PAGE 118

106 aging may implicate neurovascular coupling. We attempted to decrease the possibility of confounding compromised vascular responses with changes in cognitive processing by including only subjects who demonstrated normal activity in the left sensory-motor cortex during a finger tapping task. However, the examination of the averaged hemodynamic response function in regions of interest is crucial to inform interpretations of age-related changes in activity. The HRFs for the bilateral rostral cingulate zone and Broca’s area homologue in the right hemisphere differed significantly between groups. Interpretation of these differences is somewhat challenging since our ability to compare differences in rise time to peak of the HRFs was compromised by motion artifact in the first two images due to speaking. Nevertheless, older adults showed a more extended HRF and slower return to baseline in each area. The extended HRF seen in older adults in both regions may reflect slower processing or recruitment of resources. Peak amplitude was similar between groups in the rostral cingulate zone, but the amplitude in Broca’s homologue was larger in older adults compared to younger adults, which suggests that a different deployment of resources may occur for older adults to perform comparably to younger adults. The comparisons of activity between age groups were performed using area under the hemodynamic curve as the dependent variable. Because area under the curve averages signal intensity over the length of the HRF, it tends to equate the normative hemodynamic curve for aging to the normative hemodynamic curve for younger adults. For example, since intensity values corresponding to an undershoot act to cancel out the average intensity in the AUC calculation, it is possible that an HRF with a quick rise to peak and fall to baseline with an extended undershoot may have the same average

PAGE 119

107 intensity value as an HRF with an extended rise and fall and minimal undershoot. We chose AUC as the dependent variable to reduce the likelihood of attributing physiological differences in the HRF to cognitive differences between groups. However, there is no perfect solution to ensure that age comparisons of cognition are not complicated by physiological changes that may affect the BOLD signal. Current findings highlight the need for the continued development of analysis techniques to accurately characterize differences in the BOLD response between younger and older adults. Temporal Substrates of Semantic Representation Differences in activity between older and younger groups in word retrieval were primarily restricted to the frontal lobes and perisylvian language regions. The lack of an age-associated difference in the inferior temporal cortex indicates that older and younger adults do not differ in terms of how the posterior substrates for semantic functions operate. Therefore, the current results do not support our second hypothesis, which states that activity for older adults will be significantly less lateralized in the posterior fusiform gyrus and/or more diffuse in the anterior fusiform gyrus, characterized by less medial vs. lateral distinction than for younger adults. Findings do not suggest that word retrieval difficulties in older persons are due to deterioration of information in the neural substrates for semantic processes. In contrast, semantic processes appear well-preserved in the present group of relatively high functioning older adults. Since older and younger adults did not demonstrate differences in activity for semantic categories in the inferior temporal cortex, the groups could be combined to investigate the organization of semantic information in the inferior temporal cortex in 40 participants. Results confirm the role of the fusiform gyrus in semantic representation and extend our knowledge of the organizing dimensions of semantic information.

PAGE 120

108 However, the hypothesized dissociation between the posterior and anterior fusiform gyrus according to processing visual attributes and category, respectively, was not found. In contrast, results suggest a somewhat less complex model of semantic representation. Partial support was found for our hypothesis regarding the anterior fusiform gyrus, which states that animals will activate the lateral portion of the fusiform gyrus bilaterally more than tools or vehicles because animals, but not tools or vehicles, are living things while both tools and vehicles will activate the medial portion of the fusiform gyrus bilaterally more than animals because tools and vehicles, but not animals, are nonliving things. Animals activated an anterior region in the right lateral fusiform gyrus compared to tools and tools activated an anterior region of the left medial fusiform gyrus compared to animals. However, the medial/lateral distinction for tools and animals, respectively, was not found bilaterally. This partially replicates the findings of Chao et al. (1999) and Ishai et al. (1999). Notably, vehicles activated a large region of the medial fusiform gyrus bilaterally extending the length of the fusiform gyrus anteriorly to the collateral sulcus and parahippocampal gyrus in comparison to both animals and tools. The dominance of activity for vehicles in both the left and right fusiform gyrus compared to animals and tools was not expected. Our hypothesis regarding the role of the posterior fusiform gyrus in processing global and local visual attributes was also partially supported. Specifically, we hypothesized that in the right posterior fusiform gyrus, processing of both animals and vehicles would evoke significantly greater activity than processing of tools because identification of animals and vehicles, but not tools, is based on processing of global visual form and in the left posterior fusiform gyrus, processing of tools would evoke

PAGE 121

109 significantly greater activity than processing of animals and vehicles because identification of tools, but not animals or vehicles, is based on processing of local visual form. Our results suggest that the right-left lateralization for processing global and local form, respectively, is not restricted to the posterior region of the fusiform gyrus, but instead guides semantic representation throughout the entire fusiform gyrus. This is evident in the comparison of animals and tools, in which animals activate the right fusiform gyrus and tools activate the left fusiform gyrus. In fact, this left-right distinction is not restricted to the fusiform gyrus but includes the lateral temporal cortex as well. Animals showed greater activity than tools in the right middle temporal gyrus, whereas tools and vehicles exhibited greater activity in the left middle temporal gyrus compared to animals. Interpretation of the activity for vehicles is slightly more difficult. Bilateral activity for vehicles compared to both animals and tools suggests that vehicles are processed by both global and local visual attributes, despite our attempt to experimentally manipulate the members of the categories based on pilot data. In fact, the functional imaging data are consistent with our pilot data and suggest that vehicles comprise an intermediary category between tools and animals in that both global and local details are utilized in identification. The greater extent of activity for vehicles compared to animals and tools suggests that perhaps vehicles require more processing resources because their visual characteristics overlap with both tools and animals. If this is the case, the anterior extent of activity bordering on the parahippocampal gyrus for vehicles may represent iterative feedforward and back-propagation processing consistent with spread of activation models.

PAGE 122

110 There are several possible explanations for the differences between the current findings and those reported by Chao et al. (1999) and Ishai et al. (1999). Analysis procedures may account for different results. Namely, we performed group analyses on 40 individuals, adding to our power to detect differences, whereas the previous studies employed single-subject analysis techniques. Methodological differences may also account for the disparity of results related to category. The prior studies did not account for global and local features of their stimuli to distinguish category from attribute. Perhaps our manipulation of categories according to the degree to which animals and vehicles depend on global form and tools depend on local details for identification increased the influence of features in semantic processing such that visual form became a dominant factor for object identification and exerted a lateralized bias on a system that may otherwise represent category knowledge bilaterally. Although the focus of the current investigation was on the role of the fusiform gyrus in semantic representation, it must be noted that activity for semantic knowledge was distributed throughout the brain, including the right inferior frontal sulcus, left inferior parietal lobe and lateral temporal cortex. Tools activated the left inferior parietal lobe compared to vehicles and animals. This region has been associated with knowledge of the use of tools and lesions to this area often result in apraxia, the loss of skilled motor programs needed to use objects (Heilman & Valenstein, 1993). Animals activated the right inferior frontal sulcus compared to tools and vehicles. The prefrontal cortex has been divided according to a ventral portion, responsible for object memory, and a dorsal portion, responsible for the spatial memory, similar to the distinction between the ventral visual stream and dorsal spatial stream in the posterior cortex (Belger et al., 1998; Wilson

PAGE 123

111 et al., 1993). Therefore, processing of global form may extend as far anteriorly as the right inferior frontal sulcus given that both animals and vehicles activated this region more than tools. To summarize the current findings of semantic representation, results support the matrix theory and suggest inclusion of the role of global and local visual attributes in semantic organization. Because visual attribute was examined independent of semantic category (i.e., nonliving category shared similar attributes with living category), the current study was able to dissociate the roles of category and attribute to inform a model of visual semantic organization. As such, we interpret the data to suggest that the fusiform gyrus may have more than one organizing principle of semantics; the semantic system is organized according to category domain and visual attribute. Categories are organized by a medial vs. lateral distinction in the fusiform gyrus and visual attribute is lateralized to the right or left hemisphere according to degree of global versus local detail needed for object identification. Although object recognition has traditionally been explained as a result of hierarchical processing (Humphreys et al., 1994) consistent with sequential modular models, current results raise the possibility that processing within the fusiform gyrus occurs in a distributed manner involving co-activation of multiple components of semantic knowledge, such as visual attribute and category, to derive the name of an object. Of note, selectivity in the fusiform gyrus for animals, tools and vehicles reflected differences in the degree of activity between categories rather than all-or-none responses. In other words, regions selective for one category were activated to a lesser degree by the other categories. Consistent with previous reports (Ishai et al.,

PAGE 124

112 1999), this provides further support that object knowledge is distributed within the fusiform gyrus. Conclusion Identification of the underlying mechanisms of word retrieval problems in aging is critical to understanding the nature of cognitive aging and possible contributors to cognitive decline. Results show that compared to younger adults, older adults show more widespread activity in the frontal lobes with decreased lateralization during word retrieval. The lack of an age-related difference in the inferior temporal cortex indicates that older and younger adults do not differ in terms of how the presumed substrates for semantic functions operate. Results also support the role of features (visual attribute) and category in the organization of semantic information. Findings indicate that categories are processed in the lateral and medial regions of the fusiform gyrus according to whether they are living (animals) or nonliving (tools, vehicles), respectively. In contrast, visual attributes of global form (animals) are processed more by the right fusiform gyrus and local details (tools) are processed more by the left fusiform gyrus. When both attributes are relevant to processing (vehicles), cortex from both left and right fusiform gyri is active. Taken together, these findings suggest that a deterioration of executive functions, such as selecting, retrieving, and manipulating lexical-semantic information, underlies age-related changes in word retrieval while the organization of semantic information remains relatively preserved in aging. Further research is needed to examine the specific components of executive functions responsible for age-related word finding problems. Elucidation of the contribution of executive functions to word-finding problems will inform the development of effective interventions to promote efficient communication in healthy older adults as well as individuals with more severe language impairments

PAGE 125

113 resulting from stroke or Alzheimer’s disease. Similarly, current results inform our model of semantic representation, and this model may be applied to the study of semantic memory impairments associated with mild cognitive impairment and Alzheimer’s disease.

PAGE 126

APPENDIX A ADDITIONAL HRF ANALYSES FOR MAIN EFFECT OF AGE IN FRONTAL CORTICES For the HRF analyses, the 2 (group) x 11 (image number) interaction was significant for the bilateral supplementary motor area [F(10, 380) = 5.30, p = .000], right inferior frontal gyrus [F(10, 360) = 2.53, p = .006] and right anterior cingulate [F(10, 380) = 7.55, p = .000] Figure A-1. Region of the bilateral supplementary motor area activated by older adults relative to younger adults during picture naming, red = p < .005, yellow = p < .001, along with corresponding hemodynamic response functions (HRFs) for older and younger adults. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) group difference was found based on follow-up student’s t-tests. 114

PAGE 127

115 Figure A-2. Region of the right inferior frontal gyrus activated by older adults relative to younger adults during picture naming, red = p < .005, yellow = p < .001, along with corresponding hemodynamic response functions (HRFs) for older and younger adults. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) group difference was found based on follow-up student’s t-tests. Figure A-3. Region of the right anterior cingulate gyrus activated by older adults relative to younger adults during picture naming, red = p < .005, yellow = p < .001, along with corresponding hemodynamic response functions (HRFs) for older and younger adults. Note: the first two images are excluded from the HRF to reduce motion artifact. Asterisks indicate images in which a significant (p < .05) group difference was found based on follow-up student’s t-tests. Although the hemodynamic response functions (HRFs) in the bilateral supplementary motor area and right inferior frontal gyrus represent the typical, although

PAGE 128

116 more gradual, shape of an event-related hemodynamic response, the HRF in the right anterior cingulate does not appear like the normal event-related response. This difference is difficult to interpret, but the gradual increase in the HRF for younger adults may reflect a degree of anticipation of trials that does not occur in older adults.

PAGE 129

APPENDIX B EXPLORATORY ANALYSIS FOR MAIN EFFECT OF CATEGORY Table B-1. A Posteriori Volumes of Tissue (>200 l) Showing Differential Responses to Animals, Tools, and Vehicles for Main Effect of Category, p < .001. Region Hemisphere Peak talairach coordinates (x, y, z) Volume Frontal Cortex Inferior frontal sulcus R (40, 14, 27) 1207 l Middle frontal gyrus L (-24, -8, 60) 525 l Anterior cingulate R (4, 39, -1) 345 l Cingulate gyrus L (-7, -35, 43) 313 l Limbic Areas Posterior cingulate R (16, -50, 19) 344 l Insula L (-39, -13, 8) 228 l Sensory-Motor Cortex Post-central gyrus R (60, -18, 37) 2474 l L (-26, -44, 59) 328 l Pre-central gyrus L (-45, -7, 21) 218 l Temporal Cortex Middle temporal gyrus L (-51, -55, -2) 2493 l Inferior temporal gyrus R (43, -70, 2) 499 l Parietal Cortex Inferior parietal lobe L (-57, -33, 35) 6797 l R (33, -39, 41) 293 l Angular gyrus L (-27, -72, 30) 1817 l L (-48, -68, 22) 419 l Occipital Cortex Middle occipital gyrus R (32, -85, 19) 1554 l R (27, 90, 2) 365 l L (-28, -87, 6) 319 l Lingual gyrus R (20, -75, -11) 203 l Cuneus R (5, -89, 17) 214 l Subcortical Structures Caudate nucleus L (-4, 9, 1) 404 l 117

PAGE 130

118 Note. Italics indicate regions also significant at p < .0001 and reported previously in text.

PAGE 131

LIST OF REFERENCES Aguirre, G.K., Zarahn, D., & D'Esposito, M. (1998). The variability of human BOLD hemodynamic responses. NeuroImage, 8, 360-369. Albert, M.S, Heller, H.S, & Milberg, W. (1988). Changes in naming ability with age. Psychology and Aging, 3, 173-8. Alexander, M.P., Benson, D.F., & Stuss, D.T. (1989). Frontal lobes and language. Brain and Language, 37, 656-691. Au, R., Joung, P., Nicholas, M., Obler, L.K., Kass, R., & Albert, M.L. (1995). Naming ability across the adult life span. Aging and Cognition, 2, 300-311. Bckman, L., Ginovart, N., Dixon, R.A., Robins Wahlin, T., Walhin, A., Halldin, C., & Farde, L. (2000). Age-related cognitive deficits mediated by changes in the striatal dopamine system. American Journal of Psychiatry, 157, 635-637. Bckman, L., & Nilsson, L.B. (1996). Semantic memory functioning across the adult life span. European Psychologist, 1, 27-33. Balota, D.A., Dolan, P.O., & Duchek, J.M. (2000). Memory changes in healthy older adults. In: E.Tulving, & F.I.M. Craik. (Eds.), The Oxford handbook of memory (pp. 395-409). New York: Oxford University Press. Barch, D.M, Braver, T.S., Saab, F.W., & Noll, D.C. (2000). Anterior cingulate and the monitoring of response conflict: Evidence from an FMRI study of overt verb generation. Journal of Cognitive Neuroscience, 12, 298-309. Barry, C., & McHattie, J.V. (1995). Problems naming animals: Category-specific anomia or a misnomer? In R. Campbell & M.A. Conway (Eds.), Broken memories: Case studies in memory impairment. New York: Blackwell. Bartzokis, G., Cummings, J.L., Sultzer, D., Henderson, V.W., Nuechterlein, K.H., & Mintz, J. (2003). White matter structural integrity in healthy aging adults and patients with Alzheimer disease: A magnetic resonance imaging study. Archives of Neurology, 60, 393-398. Bayles, K.A., & Kaszniak, A.W. (1994). Communication and cognition in normal aging and dementia. Boston: A College-Hill Publication. 119

PAGE 132

120 Beauchamp, M.S., Lee, K.E., Haxby, J.V., & Martin, A. (2003). FMRI responses to video and point-light displays of moving humans and manipulable objects. Journal of Cognitive Neuroscience, 15, 991-1001. Belger, A., Puce, A., Krystal, J.H., Gore, J.C., Goldman-Rakic, P., & McCarthy, G. (1998). Dissociation of mnemonic and perceptual processes during spatial and nonspatial working memory using fMRI. Human Brain Mapping, 6, 14-32. Bentourkia, M., Bol, A., Ivanoiu, A., Labar, D., Sibomana, M., Coppens, A., Michel, C., Cosnard, G., & De Volder, A.G. (2000). Comparison of regional cerebral blood flow and glucose metabolism in the normal brain: Effect of aging. Journal of Neurological Sciences, 181, 19-28. Binder, J.R., Frost, J.A., Hammeke, T.A., Bellgowan, P.S.F., Rao, S.M., & Cox, R.W. (1999). Conceptual processing during the conscious resting state: A functional MRI study. Journal of Cognitive Neuroscience, 1, 80-93. Binder, J.R., Frost, J.A., Hammeke, T.A., Cox, R.W., Rao, S.M., & Prieto, T. (1997). Human brain language areas identified by functional magnetic resonance imaging. The Journal of Neuroscience, 17, 353-362. Birn, R.M., Saad, Z.S., & Bandettini, P.A. (2001). Spatial heterogeneity of the nonlinear dynamics in the FMRI BOLD response. NeuroImage, 14, 817-826. Bowles, N.L. (1994). Age and rate of activation in semantic memory. Psychology and Aging, 9, 414-429. Bowles, N.L., & Poon, L.W. (1985). Aging and retrieval of words from semantic memory. Journal of Gerontology, 40, 71-77. Brosseau, J., & Cohen, H. (1996). The representation of semantic categories in aging. Experimental Aging Research, 22, 381-391. Brown, A.S., & Mitchell, D.B. (1991). Age differences in retrieval consistency and response dominance. Journal of Gerontology: Psychological Sciences, 46, 332-339. Bryan, J., & Luszcz, M. (2000). Measurement of executive function: considerations for detecting adult age differences. Journal of Clinical and Experimental Neuropsychology, 22, 40-55. Buckner, R.L., Snyder, A.Z., Sanders, A.L., Raichle, M.E., & Morris, J.C. (2000). Functional brain imaging of young, nondemented, and demented older adults. Journal of Cognitive Neuroscience, 12, 24-34. Cabeza, R. (2001). Cognitive neuroscience of aging: Contributions of functional neuroimaging. Scandinavian Journal of Psychology, 42, 277-286.

PAGE 133

121 Cabeza, R. (2001). Functional neuroimaging of cognitive aging. In R. Cabeza, & A. Kingstone (Eds.), Handbook of functional neuroimaging of cognition (pp. 331-377). Cambridge, Mass.: MIT Press. Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: the HAROLD model. Psychology and Aging, 17, 85-100. Cabeza, R., Anderson, N.D., Locantore, J.K., & McIntosh, A.R. (2002). Aging gracefully: Compensatory brain activity in high-performing older adults. NeuroImage, 17, 1394-1402. Cabeza, R., Grady, C.L., Nyberg, L., McIntosh, A.R., Tulving, E., Dapur, S., Jennings, J.M., Houle, S., & Craik, F.I.M. (1997). Age-related differences in neural activity during memory encoding and retrieval: a positron emission tomography study. Journal of Neuroscience, 17, 391-400. Cameli, L., & Phillips, N.A. (2000). Age-related differences in semantic priming: evidence from event-related brain potentials. Brain and Cognition, 43, 69-73. Cao, Y., Vikingstad, E.M., George, K.P., Johnson, A.F., & Welch, K.M.A. (1999). Cortical language activation in stroke patients recovering from aphasia with functional MRI. Stroke, 30, 2331-2340. Caramazza, A. (1998). The interpretation of semantic category-specific deficits: what do they reveal about the organization of conceptual knowledge in the brain? Neurocase, 4, 265-272. Caramazza, A., Hillis, A.E., Rapp, B.C., & Romani, C. (1990). The multiple semantics hypothesis: Multiple confusions? Cognitive Neuropsychology, 7, 161-189. Caramazza, A., & Shelton, J.R. (1998). Domain-specific knowledge systems in the brain: the animate-inanimate distinction. Journal of Cognitive Neuroscience, 10, 1-34. Carbonnel, S., Charnallet, A., David, D., & Pellat, J. (1997). One or several semantic system(s)? Maybe none: evidence from a case study of modality and category specific "semantic" impairment. Cortex, 33, 391-417. Carter, C., Macdonald, A.M., Botvinick, M., Ross, L., Stenger, V.A., Noll, D., & Cohen, J.D. (2000). Parsing executive processes: strategic vs.. evaluative functions of the anterior cingulated cortex. Proceedings of the National Academy of Science, 97, 1944-1948. Cato, M.A., Crosson, B., Gokcay, D., Soltysik, D., Wierenga, C., Gopinath, K., Himes, N., Belanger, H., Bauer, R.M., Fischler, I.S., Gonzalez-Rothi, L., & Briggs, R.W (2004). Processing words with emotional connotation: An FMRI study of time course and laterality in rostral frontal and retrosplenial cortices. Journal of Cognitive Neuroscience, 16, 167-177.

PAGE 134

122 Cato, M.A., Moore, A.B., & Crosson, B. (2001). Elucidation of semantic organization in the brain using functional neuroimaging: A review. Topics in Language Disorders, 21, 60-74. Chao, L., Haxby, J., & Martin, A. (1999). Attribute-based neural substrates in temporal cortex for perceiving and knowing about objects. Nature Neuroscience, 2, 913-919. Chao, L., & Martin, A. (1999). Cortical regions associated with perceiving, naming, and knowing about colors. Journal of Cognitive Neuroscience, 11, 25-35. Chao, L., & Martin, A. (2000). Representation of manipulable man-made objects in the dorsal stream. NeuroImage, 12, 478-484. Coffey, C.E., Wilkinson, W.E., Parashos, I.A., Soady, S.A.R., Sullivan, R.J., Patterson, L.J., Figiel, G.S., Webb, M.C., Spritzer, C.E., & Djang, W.T. (1992). Quantitative cerebral anatomy of the aging human brain: a cross-sectional study using magnetic resonance imaging. Neurology, 42, 527-536. Coltheart, M. (1981). The MRC Psycholinguistic Database. Quarterly Journal of Experimental Psychology, 33A, 497-505. Coltheart, M., Inglis, L., Cupples, L, Michie, P., & Budd, W. (1998). A semantic subsystem of visual attributes. Neurocase, 4, 353-370. Copland, D.A., Chenery, H.J., & Murdoch, B.E. (2000). Persistent deficits in complex language function following dominant nonthalamic subcortical lesions. Journal of Medical Speech-Language Pathology, 8, 1-14. Cox, R.W. (1996). AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Computers in Biomedical Research, 29, 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, semantic, and phonological processes. Journal of the International Neuropsychological Society, 9, 1061-1077. Crosson, B., Cato, M.A, Sadek, J., Gokcay, D., Bauer, R., Fischler, I., Maron, L., Gopinath, K., Auerbach, E., Browd, S., & Briggs, R. (2002). Semantic monitoring of words with emotional connotation during FMRI: Contribution of anterior left frontal cortex. Journal of the International Neuropsychological Society, 8, 607-622. Crosson, B., Cato, M.A., Sadek, J.R., & Lu, L. (2000). Organization of semantic knowledge in the human brain: toward a resolution in the next millennium. Brain and Cognition, 42, 146-148.

PAGE 135

123 Crosson, B., Moberg, P.J., Boone, J.R., Gonzalez Rothi, L.J., & Raymer, A. (1997). Category-specific naming deficit for medical terms after dominant thalamic/capsular hemorrhage. Brain and Language, 60, 407-442. Crosson, B., Moore, A.B., Gopinath, K., White, K.D., Wierenga, C.E., Gaiefsky, M.E., Fabrizio, K.R., Peck, K.K., Soltysik, D., Milstead, C., Briggs, R.W., Conway, T.W., & Rothi, L.J.G. (under review). Role of the right and left hemispheres in recovery of function during treatment of intention in aphasia. Crosson, B., Sadek, J.R., Bobholz, J.A., Gokcay, D., Mohr, C.M., Leonard, C.M., Maron, L., Auerbach, E.J., Browd, S.R., Freeman, A.J., & Briggs, R.W. (1999). Activity in the paracingulate and cingulate sulci during word generation: An FMRI study of functional anatomy. Cerebral Cortex, 9, 307-316. Crosson, B., Sadek, J.R., Maron, L., Gokcay, D., Mohr, C.M., Auerbach, E.J., Freeman, A.J., Leonard, C.M., & Briggs, R.W. (2001). Relative shift in activity from medial to lateral frontal cortex during internally versus externally guided word generation. Journal of Cognitive Neuroscience, 13, 272-283. Crosson, B., Singletary, F., Richards, K., Koehler, S., & Rothi, L.J.G. (2000). Use of nonsymbolic gestures of the left hand to enhance word finding in nonfluent aphasia. Paper presented at the Second National Department of Veterans Affairs Rehabilitation Research and Development Conference, Washington, DC. Damasio, A.R., & Anderson, SW (1993). The frontal lobes. In K.M. Heilman, & E. Valenstein (Eds.), Clinical neuropsychology (pp. 409-460). New York: Oxford University Press. Damasio, H., Grabowski, T.J., Tranel, D., Hichwa, R.D., & Damasio, A.R. (1996). A neural basis for lexical retrieval. Nature, 380, 499-505. Damasio, A.R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed neural systems. Proceedings of the National Academy of Science U.S.A., 90, 4957-4960. Daut, J., Maier-Rudolph, W., von Beckerath, N., Mehrke, G., Gunther, K., &Goedel-Meinen, L. (1990). Hypoxic dilation of coronary arteries is mediated by ATP-sensitive potassium channels. Science, 247, 1341-1344. Delis, D., Kiefner, M., & Fridlund, A. (1988). Visuospatial dysfunction following unilateral brain damage: Dissociations in hierarchical and hemispatial analysis. Journal of Clinical and Experimental Neuropsychology, 10, 421-431. Delis, D.C., Kramer, J.H., Kaplan, E., & Holdnack, J. (2004). Reliability and validity of the Delis-Kaplan Executive Function System: An update. Journal of the International Neuropsychological Society, 10. 301-303.

PAGE 136

124 Delis, D., Massman, P., Butters, N., Salmon, D., Shear, P., Demadura, T., & Filoteo, J. (1992). Spatial Cognition in Alzheimer's Disease: Subtypes of Global-Local Impairment. Journal of Clinical and Experimental Neuropsychology, 14, 463-477. Demonet, J.F., Chollet, F., Ramsay, S., Cardebat, D., Nespoulous, J.F., Wise, R., Rascoll, A., & Frackowiak, R. (1992). The anatomy of phonological and semantic processing in normal subjects. Brain, 115, 1753-1768. De Renzi, E., & Lucchelli, F. (1994). Are semantic systems separately represented in the brain? The case of living category impairment. Cortex, 30, 3-25. D’Esposito, M., Deouell, L.Y., & Gassaley, A. (2003). Alterations in the BOLD FMRI signal with ageing and disease: A challenge for neuroimaging. Nature Reviews Neuroscience, 4, 863-872. D’Esposito, M., Zarahn, E., Aguirre, G.K., & Rypma, B. (1999). The effect of normal aging on the coupling of neural activity to the BOLD hemodynamic response. NeuroImage, 10, 6-14. D’Esposito, M., Zarahn, E., Aguirre, J., & Rypma, B. (1999). The variability of the BOLD hemodynamic responses in young and elderly subjects: Implications for FMRI studies of normal aging. NeuroImage, 10, 6-14. Devlin, J.T., Gonnerman, L.M., Andersen, E.S., & Seidenberg, M.S. (1998) Category-specific semantic deficits in focal and widespread brain damage: A computational account. Journal of Cognitive Neuroscience, 10, 77-94. Devlin, J.T., Russell, R.P., Davis, M.H, Price, C.J., Moss, H.E., Fadili, M.J., & Tyler, L.K. (2001). Is there an anatomical basis for category-specificity? Semantic memory studies in PET and FMRI. Neuropsychologia, 40, 54-75. DiGirolamo, G.J., Kramer, A.F., Barad, V., Cepeda, N.J., Weissman, D.H., Milham, M.P., Wszalek, T.M., Cohen, N.J., Banich, M.T., Webb, A., Belopolsky, A.V., & McAuley, E. (2001). General and task-specific frontal lobe recruitment in older adults during executive processes: A FMRI investigation of task-switching. NeuroReport, 12, 2065-2071. Doyon, J., & Milner, B. (1991). Right temporal-lobe contribution to global visual processing. Neuropsychologia, 29, 343-360. Dunn, L.M., & Dunn, E.S. (1981). Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service. Ellis, A., & Young, A. (1988). Human cognitive neuropsychology. Hove, UK: Lawrence Erlbaum Associates Ltd.

PAGE 137

125 Eustache, F., Desgranges, B., Jacques, V., & Platel, H. (1998). Preservation of the attribute knowledge of concepts in normal aging groups. Perceptual and Motor Skills, 87, 1155-1162. Evans, M.A., Shedden, J.M., Hevenor, S.J., & Hahn, M.C. (2000). The effect of variability of unattended information on global and local processing: evidence for lateralization at early stages of processing. Neuropsychologia, 38, 225-239. Eysenck, M.W., & Keane, M.T. (1995). Cognitive psychology. Hove, UK: Psychology Press Ltd. Fang, H.C.H. (1976). Observations on aging characteristics of cerebral blood vessels, macroscopic and microscopic featuers. In S. Gerson, & R.D. Terry (Eds.), Neurobiology of aging. NY: Raven Press Farah, M.J., & McClelland, J.L. (1991). A computational model of semantic memory impairment: Modality specificity and emergent category specificity. Journal of Experimental Psychology: General, 120, 339-357. Farah, M.J., & Wallace, M.A. (1992). Semantically-bounded anomia: implications for the neural implementation of naming. Neuropsychologia, 30, 609-621. Fiez, J.A. (1997). Phonology, semantics, and the role of the left inferior prefrontal cortex. Human Brain Mapping, 5, 79-83. Fink, G.R., Halligan, P.W., Marshall, J.C., Frith, C.D., Frackowiak, R.S., & Dolan, R.J. (1997). Neural mechanisms involved in the processing of global and local aspects of hierarchically organized visual stimuli. Brain, 120, 1779-1791. Fink, G.R., Marshall, J.C., Halligan, P.W., & Dolan, R.J. (1999). Hemispheric asymmetries in global/local processing are modulated by perceptual salience. Neuropsychologia, 37, 31-40. Fodor, J.A. (1983). The modularity of mind. Cambridge, Mass.: MIT Press. Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). ‘Mini-mental State’. A practical method for grading the cognitive status of patients for the clinician. Journal of Psychiatric Research, 12, 189-198. Forman, S.D., Cohen, J.D., Fitzgerald, M., Eddy, W.F., Mintun, M.A., & Noll, D.C. (1995). Improved assessment of significant activation in functional magnetic resonance imaging (FMRI): Use of a cluster-size threshold. Magnetic Resonance in Medicine, 33, 636-647. Freedman, L., Snow, W.G., & Millikin, C. (1995). Anomia in Alzheimer’s disease. Journal of the International Neuropsychological Society, 1, 386.

PAGE 138

126 Friston, K.J., Price, C.J., Fletcher, P., Moore, C., Frackowiak, R.S.J., & Dolan, R.J. (1996). The trouble with cognitive subtraction. NeuroImage, 4, 97-104. Furuta, A., Ishii, N., Nishihara, Y., & Horie, A. (1991). Medullary arteries in aging and dementia. Stroke, 22, 442-446. Fuster, J.M. (1989). The prefrontal cortex (2nd ed.). New York: Raven Press. Gabrieli, J.D. (1996). Memory systems analyses of mnemonic disorders in aging and age-related diseases. Proceedings of the National Academy of Science USA, 93, 13534-13540. Gabrieli, J.D., Desmond, J.E., Demb, J.B. & Wagner, A.D. (1996). Functional magnetic resonance imaging of semantic memory processes in the frontal lobes. Psychological Science, 7, 278-283. Gabrieli, J.D., Poldrack, R.A., & Desmond, J.E. (1998). The role of left prefrontal cortex in language and memory. Proceedings of the National Academy of Science, 95, 906-913. Gaiefsky, M.E., Crosson, B., White, K.D., Peck, K.K., Moore, A.B., Wierenga, C., Gopinath, K.S., Soltysik, D., & Briggs, R. (2002, November). Imaging Overt Language Production in Aphasia Using FMRI. Poster session presented at the 32nd annual meeting of the Society for Neuroscience, Orlando, FL. Gainotti, G. (2000). What the locus of brain lesion tells us about the nature of the cognitive defect underlying category-specificity disorders: a review. Cortex, 36, 539-59. Gainotti, G., & Silveri, M.C. (1996). Cognitive and anatomical locus of lesion in a patient with a category-specific semantic impairment for living beings. Cognitive Neuropsychology, 13, 357-389. Gerfen, C.R. (1992). The neostriatal matrix: Multiple levels of compartmental organization in the basal ganglia. Annual Review of Neuroscience, 15, 285-320. Goldman-Rakic, P.S. (1984). Modular organization of prefrontal cortex. Trends in Neuroscience, 7, 419-424. Goldman-Rakic, P.S., & Brown, R.M. (1981). Regional changes of monoamines in cerebral cortex and subcortical structures of aging rhesus monkeys. Neuroscience, 6, 177-187. Gonnerman, L.M., Andersen, E.S., Devlin, J.T., Kempler, D., & Seidenberg, M.S. (1997). Double dissociation of semantic categories in Alzheimer's Disease. Brain and Language, 57, 254-279.

PAGE 139

127 Good, C.D., Johnsrude, I.S., Ashburner, J., Henson, R.N.A., Friston, K.J., & Frackowiak, R.S.J. (2001). A voxel-based morphometric study of ageing in 465 normal adult human brains. NeuroImage, 14, 21-36. Goodglass, H., & Wingfield, A. (1993). Selective preservation of a lexical category in aphasia: dissociations in comprehension of body parts and geographical place names following focal brain lesion. Memory, 1, 313-328. Grabowski, T.J., Damasio, H., & Damasio, A.R. (1998). Premotor and prefrontal correlates of category-related lexical retrieval. NeuroImage, 7, 232-243. Grady, C.L. (2000). Functional brain imaging and age-related changes in cognition. Biological Psychology, 54, 259-281. Grady, C.L., Maisog, J.M., Horwitz, B., Ungerleider, L.G., Mentis, M.J., Salemo, J.A., Pietrini, P., Wagner, E., & Haxby, J.V. (1994). Age-related changes in cortical blood flow activation during visual processing of faces and location. Journal of Neuroscience, 14, 1450-1462. Grady, C.L., McIntosh, A.R., Bookstein, F., Horwitz, B., Rapoport, S.I., & Haxby, J.V. (1998). Age-related changes in regional cerebral blood flow during working memory for faces. NeuroImage, 8, 409-425. Grady, C.L., McIntosh, A.R., Horwitz, B., Maisog, J.M., Ungerleider, L.G., Mentis, M.J., Pietrini, P., Shapiro, M.B., & Haxby, J.V. (1995). Age-related reductions in human recognition memory due to impaired encoding. Science, 269, 218-221. Gunning-Dixon, F.M., Head, D., McQuain, J., Acker, J.D., & Raz, N. (1998). Differential aging of the human striatum: A prospective MR imaging study. American Journal of Neruoradiology, 19, 1501-1507. Gunning-Dixon, F., & Raz, N. (2000). The cognitive correlates of white matter abnormalities in normal aging: A quantitative review. Neuropsychology, 14, 224-232. Gur, R.E., Cowell, P., Turetsky, B.I., Gallagher, R., Cannon, T., Bilker, W., & Gur, R.C. (1998). A follow-up magnetic resonance imaging study of schizophrenia relationship of neuroanatomical changes to clinical and neurobehavioral measures. Archives of General Psychiatry, 55, 145-152. Gur, R.C., Gur, R.E., Orbist, W.D., Skolnick, B.E., & Reivich, M. (1987). Age and regional cerebral blood flow at rest and during cognitive activity. Archives of General Psychiatry, 44, 617-621. Guttmann, C.R.G., Jolesz, F.A., Kikinis, R., Killiany, R.J, Moss, M.B., Sandor, T., & Albert, M.S. (1998). White matter changes with normal aging. Neurology, 50, 972-978.

PAGE 140

128 Han, S., Liu, W., Yund, E.W., & Woods, D.L. (2000). Interactions between spatial attention and global/local feature selection: an ERP study. NeuroReport, 11, 2753 –2758. Hart, J., & Gordon, B. (1992). Neural subsystems for object knowledge. Nature, 359, 60-64. Haug, H. (1997). The aging human cerebral cortex: Morphometry of areal differences and their functional meaining. In S.U. Dani, A. Hori, & G.F. Walter (Eds.), Principles of neural aging, Amsterdam: Elsevier Publishing. Haug, H., & Eggers, R. (1991). Morphometry of the human cortex cerebri and corpus striatum during aging. Neurobiology of Aging, 12, 336-338. Hauk, O., Johnsrude, I., & Pulvermller, F. (2004). Somatotopic representation of action words in human motor and premotor cortex. Neuron, 41, 301-307. Head, D., Buckner, R.L., Shimony, J.S., Williams, L.E., Akbudak, E., Conturo, T.E., McAvoy, M., Morris, J.C., & Snyder, A.Z. (2004). Differential vulnerability of anterior white matter in nondemented aging with minimal acceleration in dementia of the Alzheimer Type: Evidence from diffusion tensor imaging. Cerebral Cortex, 14, 410-423. Hedden, T., & Gabrieli, D.E. (2004). Insights into the ageing mind: A view from cognitive neuroscience. Nature Reviews Neuroscience, 5, 87-97. Heilman, K.M., & Valenstein, E. (1993). Clinical neuropsychology (3rd ed.). New York: Oxford University Press. Heiss, W.D., Kessler, J., Thiel, A., Ghaemi, M, & Karbe, H. (1999). Differential capacity of left and right hemispheric areas for compensation of poststroke aphasia. Annals of Neurology, 45, 430-438. Hesselmann, V., Weber, O.Z., Wedekind, C., Drings, T., Schulte, O., Kugel, H., Krug, B., Klug, N., & Lackner, K.J. (2001). Age related signal decrease in functional magnetic resonance imaging during motor stimulation in humans. Neuroscience Letters, 308, 141-144. Hillis, A.E., & Caramazza, A. (1991). Category-specific naming and comprehension impairment: a double dissociation. Brain, 114, 2081-2094. Hillis, A., & Rapp, B. (1995). Constraining claims about theories of semantic memory: More on unitary versus multiple semantics. Cognitive Neuropsychology, 12, 175-186. Hillis, A., Rapp, B., Romani, C., & Caramazza, A. (1990). Selective impairments of semantics in lexical processing. Cognitive Neuropsychology, 7, 191-243.

PAGE 141

129 Hogervorst, E., Combrind, M., Lapuerta, P., Rue, J., Swales, K., & Budge, M. (2002). The Hopkins Verbal Learning Test and screening for dementia. Dementia and Geriatric Cognitive Disorders, 13, 13-20. Huettel, S.A., Singerman, J.D., & McCarthy, G. (2001). The effects of aging upon the hemodynamic response measured by functional MRI. NeuroImage, 13, 161-175. Humphreys, G.W., Riddoch, M.J., & Quinlan, P.T. (1988). Cascade processes in picture identification. Cognitive Neuropsychology, 5, 67-103. Ishai, A., Ungerleider, L., Martin, A., Schouten, J., & Haxby, J. (1999). Distributed representation of objects in the human ventral visual pathway. Proceedings of the National Academy of Science, 96, 9379-9384. Ivnik, R.J., Malec, J.F., Smith, G.E., Tangalos, E.G., & Petersen, R.C. (1996). Neuropsychological norms above age 55: COWA, BNT, MAE Token, WRAT-R Reading, AMNART, STROOP, TMT, and JLO. The Clinical Neuropsychologist, 10, 262-278. Johannes, S., Wieringa, B., Matzke, M., & Munte, T. (1996). Hierarchical visual stimuli: electrophysiological evidence for separate left hemispheric global and local processing mechanisms in humans. Neuroscience Letters, 210, 111-114. Kanwisher, N., McDermott, J., & Chun, M. (1997). The fusiform face area: A module in human extrastriate cortex specialized for face perception. The Journal of Neuroscience, 17, 4302-4311. Kaplan, E.F., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test (2nd ed.). Philadelphia: Lea & Febiger. Kawamura, J., Terayama, Y., Takashima, S., Obara, K., Pavol, M.A., Meyer, J.S., Mortel, K.F., & Weathers, S. (1993). Leuko-araiosis and cerebral perfusion in normal aging. Experimental Aging Research, 19, 225-240. Kempler, D., Teng, E.L., Dick, M., Taussig, I.M., & Davis, D.S. (1998). The effects of age, education, and ethnicity on verbal fluency. Journal of the International Neuropsychological Society, 4, 531-538. Kempler, D. & Zelinski, E.M. (1994). Language in dementia and normal aging. In F.A. Huppert, C. Brayne, & D.W. O’Connor (Eds.), Dementia and normal aging (pp. 331-365). Cambridge, UK: Cambridge University Press. Keys, B.A., & White, D.A. (2000). Exploring the relationship between age, executive abilities, and psychomotor speed. Journal of the International Neuropsychological Society, 6, 76-82. King, K.F., Foo, T.K.F., & Crawford, C.R. (1995). Optimized gradient waveforms for spiral scanning. Magnetic Resonance in Medicine, 34, 156-160.

PAGE 142

130 Knesevich, J.W., LaBarge, E. & Edwards, D. (1986). Predictive value of the Boston Naming Test in mild senile dementia of the Alzheimer type. Psychiatry Research, 19, 155-161. Kraut, M.A., Moo, L.R., Segal, J.B., & Hart, J. (2002). Neural activation during an explicit categorization task: categoryor feature-specific effects? Brain Research Cognitive Brain Research, 13, 213-220. Krejza, J., Mariak, Z., Walecki, J., Szydlik, P., Lewko, J., & Ustymowicz, A. (1999). Transcranial color Doppler sonography of basal cerebral arteries in 182 healthy subjects: Age and sex variability and normal reference values for blood flow parameters. American Journal of Roentgenology, 172, 213-218. Kuchera, H. & Francis, W. (1967). Computational Analysis of Present-Day American English. Providence: Brown University Press. Laver, G.D., & Burke, D.M. (1993). Why do semantic priming effects increase in old age? A meta-analysis. Psychology and Aging, 8, 34-43. La Voie, D., & Light, L.L. (1994). Adult age differences in repetition priming: A meta-analysis. Psychology and Aging, 9, 539-553. Laws, K., Evans, J., Hodges, J., & McCarthy, R. (1995). Naming without knowing and appearance without associations: evidence for constructive processes in semantic memory? Memory, 3, 409-433. Le Clec'H., G., Dehaene, S., Cohen, L., Mehler, E., Dupoux, J., Poline, J., Lehericy, S., van de Moortele, P., & Le Bihan, D. (2000). Distinct cortical areas for names of numbers and body parts independent of language and input modality. NeuroImage, 12, 381-391. Lezak, M.D. (1995). Neuropsychological assessment (3rd ed.). New York: Oxford University Press. Light, L. (1992). The organization of memory in old age. In F.I.M. Craik & T.A. Salthouse (Eds.), The handbook of aging and cognition. Hillsdale, NJ: Erlbaum. Linnville, S. (1995). Are there age-related processing differences in semantic priming? Perceptual and Motor Skills, 80, 585-586. Logan, J.M, Sanders, A.L., Snyder, A.Z., Morris, J.C., & Buckner, R.L. (2002). Under-recruitment and nonselective recruitment: Dissociable neural mechanisms associated with aging. Neuron, 33, 827-840. Lu, L.H., Crosson, B., Nadeau, S.E., Heilman, K.M., Gonzalez-Rothi, L.J., Raymer, A., Gilmore, R.L., Bauer, R.M., & Roper, S.N. (2002). Category-specific naming deficits for objects and actions: Semantic attribute and grammatical role hypotheses. Neuropsychologia, 40, 1608-1621.

PAGE 143

131 Lucas, J.A., Ivnik, R.J., Smith, G.E., Bohac, D.L., Tangalos, E.G., Graff-Radford, N.R., & Petersen, R.C. (1998). Mayo’s older Americans normative studies: Category fluency norms. Journal of Clinical and Experimental Neuropsychology, 20, 194-200. Luria, A.R. (1980). Higher cortical function in man. New York: Basic Books. MacDonald A.W., Cohen, J.D., Stenger, V.A., & Carter, C.S. (2000). Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science, 288, 1835-1838. Macovski, A. (1985). Volumetric NMR imaging with time-varying gradients. Magnetic Resonance Imaging in Medicine, 2, 29-40. Madden, D.J., & Hoffman, J.M. (1997). Application of positron emission tomography to age-related cognitive changes. In K.R.R. Krishnan, & P.M. Doraiswamy (Eds.), Brain imaging of clinical psychiatry (pp. 575-613). New York: Marcel Dekker. Madden, D.J., Langley, L.K., Denny, L.L., Turkington, T.G., Provenzale, J.M., Hawk, T.C., & Coleman, R.E. (2002). Adult age differences in visual word identification: Functional neuroanatomy by positron emission tomography. Brain and Cognition, 49, 297-321. Madden, D.J., Turkington, T.G., Coleman, R.E., Provenzale, J.M., DeGrado, T.R., & Hoffman, J.M. (1996). Adult age differences in regional cerebral blood flow during visual word identification: Evidence from H215O PET. NeuroImage, 3, 127-142. Madden, D.J., Turkington, T.G., Provenzale, J.M., Hawk, T.C., Hoffman, J.M., & Coleman, R.E. (1997). Selective and divided visual attention: age-related changes in regional cerebral blood flow measured by H15O PET. Human Brain Mapping, 5, 115-135. Maher, L., & Ochipa, C. (1997). Management and treatment of limb apraxia. In L.J.G. Rothi and K.M. Heilman (Eds.), Apraxia: The neuropsychology of action (pp. 75-91). Hove: Psychology Press. Malech, R., Reppas, J., Benson, R., Kwong, K., Jiang, H., Kennedy, W., Ledden, P., Brady, T., Rosen, B. & Tootell, B. (1995). Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex. Proceedings of the National Academy of Science, 92, 8135-8139. Martin, A., Haxby, J., Lalonde, F., Wiggs, S., & Ungerleider, L. (1995). Discrete cortical regions associated with knowledge of color and knowledge of action. Science, 270, 102-105. Martin, A.,Wiggs, C.L., Ungerleider, L.G., & Haxby, J.V. (1996). Neural correlates of category-specific knowledge. Nature, 379, 649-652.

PAGE 144

132 Masawa, N., Yoshida, Y., Yamada, T., Joshita, T., Sato, S., & Mihara, B. (1994). Morpometry of structural preservation of tunica media in aged and hypertensive human intracerebral arteries. Stroke, 25, 122-127. Massman, P., Delis, D., Filoteo, J., Butters, N., Salmon, D., & Demadura, T. (1993). Mechanisms of spatial impairment in Alzheimer's Disease subgroups: Differential breakdown of directed attention to global-local stimuli. Neuropsychology, 7, 172-181. Mayr, U., & Kliegl., R. (2000). Complex semantic processing in old age: Does it stay or does it go? Psychology and Aging, 15, 29-43. McCarthy, R.A., & Warrington, E.K. (1988). Evidence for modality-specific meaning systems in the brain. Nature, 344, 428-430. McClelland, J.R., & Rumelhart, D.E. (1985). Distributed memory and the representation of general and specific information. Journal of Experimental Psychology General, 114. 159-197. McCrae, R.R., Arenberg, D., & Costa, P.T. (1987). Declines in divergent thinking with age: Cross-sectional, longitudinal, and cross-sequential analyses. Psychology and Aging, 2, 130-137. Menon, R.S., Ogawa, S., Hu, X., Strupp, J.P., Anderson, P., & Ugurbil, K. (1995). BOLD bsed functional MRI at 4 Tesla includes a capillary bed contribution: Echo-planar imaging correlates with previous optical imaging using intrinsic signals. Magnetic Resonance in Medicine, 33, 453-459. Mesulam, M.M. (2000). Aging, Alzheimer’s Disease and Dementia. In Principles of behavioral and cognitive neurology (pp. 439-516). London: Oxford University Press. Meuller, E.A., Moore, M.M., Kerr, D.C.R., Sexton, G., Camicioli, R.M, Howieson, D.B., Quinn, J.F., & Kaye, J.A. (1998). Brain volume preserved in healthy elderly through the eleventh decade. Neurology, 51, 1555-1562. Milberg, W., Mcglinchey-Berroth, R., Duncan, K.M, & Higgins, J.A. (1999). Alterations in the dynamics of semantic activation in Alzheimer’s disease: Evidence for the Gain/Decay hypothesis of a disorder of semantic memory. Journal of the International Neuropsychological Society, 5, 641-658. Mink, J.W. (1996). The basal ganglia: Focused selection and inhibition of competing motor programs. Progress in Neurobiology, 50, 381-425. Mishkin, M., & Appenzeller, T. (1987). The anatomy of memory. Scientific American, 256, 80-89.

PAGE 145

133 Mishkin, M., Ungerleider, L., & Macko, K. (1983). Object vision and spatial vision: two cortical pathways. Trends in Neuroscience, 6, 414-417. Mitrushina, M., & Satz, P. (1995). Repeated testing of normal elderly with the Boston Naming Test. Aging: Clinical and Experimental Research, 7, 123-127. Miyamoto, T., Katayama, J., & Koyama, T. (1998). ERPs, semantic processing and age. International Journal of Psychophysiology, 29, 43-51. Moberg, M., Ferraro, F., & Petros T.V. (2000). Lexical properties of the Boston Naming Test stimuli: Age differences in word naming and lexical decision latency. Applied Neuropsychology, 7, 147-153. Moody, D.M., Brown, W.R., Challa, V.R., Ghazi-Birry, H.S., & Reboussin, D.M. (1997). Cerebral microvascular alterations in aging, leukoaraiosis, and Alzheimer’s disease. Annals of the New York Academy of Science, 826, 103-116. Moore, C. & Price, C. (1999). A functional neuroimaging study of the variables that generate category-specific object processing differences. Brain, 122, 943-962. Mummery, C.J., Patterson, K., Hodges, J.R., & Price, C.J. (1998). Functional neuroanatomy of the semantic system: divisible by what? Journal of Cognitive Neuroscience, 10, 766-777. Murtha, S., Chertkow, H., Beauregard, M., & Evans, A. (1999). The neural substrate of picture naming. Journal of Cognitive Neuroscience, 11, 399-423. Myerson, J., Hale, S., Chen, J. & Lawrence, B. (1997). General lexical slowing and the semantic priming effect: The roles of age and ability. Acta Psychologica, 96, 83-101. Nagasawa, S., Handa, H., Okumura, A., Naruo, Y., Moritake, K., & Hayashi, K. (1979). Mechanical properties of human cerebral arteries. Part 1: Effects of age and vascular smooth muscle activation. Surgical Neurology, 12, 297-304. Navon, D. (1977). Forest before trees: The precedence of global features in visual processing. Cognitive Psychology, 9, 353-383. Neils, J., Baris, J.M., Carter, C., Dell’aira, A., Nordloh, S.J., Weiler, E., & Weisiger, B. (1995). Journal of Speech and Hearing Research, 38, 1143-1149. Noll, D.C., Cohen, J.D., Meyer, C.H., & Schneider, W.J. (1995). Spiral k-space MR imaging of cortical activation. Magnetic Resonance Imaging, 5, 49-56.

PAGE 146

134 Ojemann, J.G., Buckner, R.L., Akbudak, E., Snyder, A.Z., Ollinger, J.M, McKinstry, R.C., Rosen, B.R., Petersen, S.E., Raichle, M.E., & Conturo, T.E. (1998). Functional MRI studies of word-stem completion: Reliability across laboratories and comparison to blood flow imaging with PET. Human Brain Mapping, 6, 203-215. Oldfield, R.C. (1971). The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologia, 9, 97-113. Paivio, A. (1971). Imagery and verbal processes. New York: Holt, Rinehart & Winston. Paivio, A. (1986). Mental comparisons involving abstract attributes. Memory and Cognition, 2, 199-208. Palmer, E.D., Rosen, H.J., Ojemann, J.G., Buckner, R.L., Kelley, W.M., & Petersen, S.E. (2001). An event-related fMRI study of overt and covert word stem completion. NeuroImage, 14, 182-193. Park, D.C., Lautenschlager, G., Hedden, T., Davidson, N.S., Smith, A.D., & Smith, P.K. (2002). Models of visuospatial and verbal memory across the adult life span. Psychology and Aging, 17, 299-320. Park, D.C., Smith, A.D., Lautenschlager, G., Earles, J.L., Frieske, D., Zwahr, M, & Gaines, C.L. (1996). Mediators of long-term memory performance across the life span. Psychology and Aging, 11, 621-637. Penney, J.B., Jr. & Young, A.B. (1986). Striatal inhomogeneities and basal ganglia function. Movement Disorders, 1, 3-15. Petersen, R.C., Doody, R., Kurz, A., Mohs, R.C., Morris, J.C., Rabins, P.V., Ritchie, K., Rossor, M., Thal, L., & Winblad, B. (2001). Current concepts in mild cognitive impairment. Archives of Neurology, 58, 1985-1992. Petersen, S., Fox, P.T., Posner, M.I., Mintur, M., & Raichle, M.E. (1988). Positron emission tomographic studies of the cortical anatomy of single-word processing. Nature, 331, 585-589. Petersen, S., Van Mier, H., Fiez, J., & Raichle, M. (1998). The effects of practice on the functional anatomy of task performance. Proceedings of the National Academy of Science, 95, 853-860. Poldrack, R.A., Wagner, A.D., Prull, M.W., Desmond, J.E., Glover, G.H, & Gabrieli, J.D.E. (1999). Functional specialization for semantic and phonological processing in the left inferior prefrontal cortex. NeuroImage, 10, 15-35. Price, C., Moore, C., Humphreys, G., Frackowiak, R., & Friston, K. (1996). The neural regions sustaining object recognition and naming. Proceedings of the Royal Society in London, 263, 1501-1507.

PAGE 147

135 Pulvermuller, F. (2001). Brain reflections of words and their meaning. Trends in Cognitive Sciences, 5, 517-524. Pulvermuller, F., Harle, M., & Hummel, F. (2001). Walking or talking? Behavioral and neurophysiological correlates of action verb processing. Brain and Language, 78, 143-168. Randolph, C., Braun, A.R., Goldberg, T.E., & Chase, T.N. (1993). Semantic fluency in Alzheimer’s, Parkinson’s, and Huntington’s disease: dissociation of storage and retrieval failures. Neuropsychology, 7, 82-88. Rao, S.M., Bandettini, P.A., Binder, J.R., Bobholz, J.A., Hammeke, T.A., Stein, E.A., & Hyde, J.S. (1996). Relationship between finger movement rate and functional magnetic resonance signal change in human primary motor cortex. Journal of Cerebral Blood Flow and Metabolism, 16, 1250-1254. Rapp, B., Hillis, A., & Caramazza, A. (1993). The role of representations in cognitive theory: More on multiple semantics and the agnosias. Cognitive Neuropsychology, 10, 235-249. Rapp, P.R., Deroche, P.S., Mao, Y., & Burwell, R.D. (2002). Neuron number in the parahippocampal region is preserved in aged rats with spatial learning deficits. Cerebral Cortex, 12, 1171-1179. Rasmussen, T., Schliemann, T., Sorensen, J.C., Zimmer, J., & West, M.J. (1996). Memory impaired aged rats: no loss of principal hippocampal and subicular neurons. Neurobiology and Aging, 17, 143-147. Raymer, A.M., & Rothi, L.J.G. (2000). The semantic system. In S. E. Nadeau, L. J. G. Rothi, & B. Crosson (Eds.), Aphasia and language: Theory to practice (pp. 108-132). New York: The Guilford Press. Raz, N. (2000). Aging of the brain and its impact on cognitive performance: Integration of structural and functional findings. In F.I.M. Craik & T.A. Salthouse (Eds.), The handbook of aging and cognition (2nd ed.) (pp. 1-90). Mahwah, NJ: Lawrence Erlbaum Associates. Raz, N., Gunning-Dixon, F., Head, D., Dupuis, J.H., & Acker, J.D. (1998). Neuroanatomical correlates of cognitive aging: Evidence from structural magnetic resonance imaging. Neuropsychology, 12, 95-114. Raz, N., Gunning, F.M., Head, D., Dupuis, J.H., McQuain, J., Briggs, S.D., Loken, W.J., Thornton, A.E., & Acker, J.D. (1997). Selective aging of the human cerebral cortex observed in vivo: Differential vulnerability of the prefrontal gray matter. Cerebral Cortex, 7, 268-282.

PAGE 148

136 Raz, N., Gunning, F.M., Head, D., Dupuis, J.H., McQuain, J.M., Briggs, S.D., Thornton, A.E., Loken, W.J., & Acker, J.D. (1997). Selective aging of human cerebral cortex observed in vivo: Differential vulnerability of the prefrontal gray matter. Cerebral Cortex, 7, 268-282. Raz, N., Gunning-Dixon, F., Head, D., Rodrigue, K.M., Williamson, A., & Acker, J.D. (2004). Aging, sexual dimorphism, and hemispheric asymmetry of the cerebral cortex: replicability of regional differences in volume. Neurobiology of Aging, 25, 377-396. Resnick, S.M., Pham, D.L., Kraut, M.A., Zonderman, A.B., & Davatzikos, C. (2003). Longitudinal magnetic resonance imaging studies of older adults: a shrinking brain. The Journal of Neuroscience, 23, 3295-3301. Reuter-Lorenz, P. (2002). New visions of the aging mind and brain. Trends in Cognitive Science, 6, 394. Reuter-Lorenz, P.A., Jonides, J., Smith, E.E., Hartley, A., Miller, A., Marshuetz, C., & Koeppe, R.A. (2000). Age differences in the frontal lateralization of verbal and spatial working memory revealed by PET. Journal of Cognitive Neuroscience, 12, 174-187. Richards, K., Singletary, F., Rothi, L.J.G., Koehler, S., & Crosson, B. (2002). Activation of intentional mechanisms through utilization of nonsymbolic movements in aphasia rehabilitation. Journal of Rehabilitation Research and Development, 39, 7-16. Rosen, A.C., Prull, M.W., Gabrieli, J.D., Stoub, T., O’Hara, R., Friedman, L., Yesavage, J.A., & deToledo-Morrell, L. (2003). Differential associations between entorhinal and hippocampal volumes and memory performance in older adults. Behavioral Neuroscience, 117, 1150-1160. Rosen, A.C., Prull, M.W., O’Hara, R., Race, E.A., Desmond, J.E., Glover, G.H., Yesavage, J.A., & Gabrieli, J.D.E. (2002). Variable effects of aging on frontal lobe contributions to memory. NeuroReport, 13, 2425-2428. Rosen, H.J., Ojemann, J.G., Ollinger, J.M., & Petersen, S.E. (2000). Comparison of brain activation during word retrieval done silently and aloud using fMRI. Brain and Cognition, 42, 201-217. Ross, M.H., Yurgelun-Todd, D.A., Renshaw, P.F., Maas, L.C., Mendelson, J.H., Mello, N.K., Cohen, B.M., & Levin, J.M. (1997). Age-related reduction in functional MRI response to photic stimulation. Neurology, 48, 173-178. Rubin, D.C. (1999). Frontal-striatal circuits in cognitive aging: Evidence for caudate involvement. Aging, Neuropsychology, and Cognition, 6, 241-259.

PAGE 149

137 Rypma, B., & D’Esposito, M. (2000). Isolating the neural mechanisms of age-related changes in human working memory. Nature Neuroscience, 3, 509-515. Rypma, B., Prabhakaran, V., Desmond, J.E., & Gabrieli, J.D.E. (2001). Age differences in prefrontal cortical activity in working memory. Psychology and Aging, 16, 371-384. Sacchett, C., & Humphreys, G. (1992). Calling a squirrel a squirrel, but a canoe a wigwam: a category-specific deficit for artefactual objects and body parts. Cognitive Neuropsychology, 9, 73-86. Salat, D.H., Kaye, J.A, & Janowksy, J.S. (2001). Selective preservation and degeneration within the prefrontal cortex in aging and Alzheimer disease. Archives of Neurology, 58, 1403-1408. Salthouse, T.A. (1996). The processing-speed theory of adult age differences in cognition. Psychology Reviews, 103, 403-428. Schacter, D.L., Savage, C.R., Alpert, N.M., Rauch, S.L., & Alpert, M.S. (1996). The role of hippocampus and frontal cortex in age-related memory changes: A PET study. NeuroReport, 7, 1165-1169. Schaie, K.W. (1996). Intellectual development in adulthood: The Seattle Longitudinal Study. Cambridge: Cambridge University Press. Shapiro, A.M., Benedict, R.H., Schretlen, D., & Brandt, J. (1999). Construct and concurrent validity of the Hopkins Verbal Learning Test-revised. Clinical Neuropsychology, 13, 348-358. Sheridan, J., & Humphreys, G. (1993). A verbal-semantic category-specific recognition impairment. Cognitive Neuropsychology, 10, 143-184. Silveri, M., & Gainotti, G. (1988). Interaction between vision and language in category-specific semantic impairment. Cognitive Neuropsychology, 5, 677-709. Snodgrass, J.G. & Vanderwart, M. (1980). A standardized set of 260 pictures: Norms for name agreement, familiarity, and visual complexity. Journal of Experimental Psychology: Human Learning and Memory, 6, 174-215. Spieler, D., & Balota, D. (2000). Factors influencing word naming in younger and older adults. Psychology and Aging, 15, 225-231. Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests: Administration, norms, and commentary (2nd ed.). New York: Oxford University Press.

PAGE 150

138 Stebbins, G.T., Carrillo, M.C., Dorfman, J., Dirkson, C., Desmond, J.E., Turner, D.A., Bennett, D.A., Wilson, R.S., Glover, G., & Gabrieli, J.D.E. (2002). Aging effects on memory encoding in the frontal lobes. Psychology and Aging, 17, 44-55. Stuss, D.T., & Alexander, M.P. (2000). Executive functions and the frontal lobes: a conceptual view. Psychological Research, 63, 289-298. Stuss, R.T., & Levine, B. (2002). Adult clinical neuropsychology: lessons from studies of the frontal lobes. Annual Reviews in Psychology, 53, 401-33. Talairach, J., & Tournoux, P. (1988). Co-Planar stereotaxic atlas of the human brain. New York: Thiem Medical Publishers. Tanaka, H., & Fujita, I. (2000). Global and local processing of visual patterns in macaque monkeys. NeuroReport, 11, 2881-2884. Taoka, T., Iwasaki, S., Uchida, H., Fukusumi, A., Nakagawa, H., Kichikawa, K., Takayama, K., Yoshioka, T., Takewa, M., & Ohishi, H. (1998). Age correlation of the time lag in signal change on EPI-FMRI. Journal of Computer Assisted Tomography, 22, 514-517. Terry, R.D. (2000). Cell death or synaptic loss in Alzheimer disease. Journal of Neuropathology and Experimental Neurology, 59, 1118-1119. Thompson-Schill, S.L., Aguirre, G.K., D'Esposito, M., & Farah, M.J. (1999). A neural basis for category and modality specificity of semantic knowledge. Neuropsychologia, 37, 671-676. Thompson-Schill, S.L., D’Esposito, M., Aguirre, G.K., & Farah, M. (1997). Role of left inferior prefrontal cortex in retrieval of semantic knowledge: A reevaluation. Proceedings of the National Academy of Science, USA, 94, 14792-12797. Tisserand, D.J., Pruessner, J.C., Sanz Arigita, E.J., van Boxtel, M.P.J., Evans, A.C., Jolles, J., & Uylings, H.B.M. (2002). Regional frontal cortical volumes decrease differentially in aging: An FMRI study to compare volumetric approaches and voxel-based morphometry. NeuroImage, 17, 657-669. Tombaugh, T.N., & Hubley, A.M. (1997). The 60-item Boston Naming Test: Norms for cognitively intact adults aged 25 to 88 years. Journal of Clinical and Experimental Neuropsychology, 19, 922-932. Troyer, A., Moscovitch, M. & Winocur, G. (1997). Clustering and switching as two components of verbal fluency: Evidence from younger and older healthy adults. Neuropsychology, 11, 138-146. Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.), Organization of memory (pp 382-404). New York: Academic Press.

PAGE 151

139 Ungerleider, L.G., & Haxby, J.V. (1994). 'What' and 'where' in the human brain. Current Opinion in Neurobiology, 4, 157-165. Ungerleider, L., & Mishkin, M. (1982). Two cortical visual systems. In D.J. Ingle, M.A. Goodale, & R.J.W. Mansfield (Eds.), Analysis of Visual Behavior. Cambridge, Mass.: MIT Press. Vandenberghe, R., Price, C., Wise, R., Josephs, O. & Frackowiak, R. (1996). Functional anatomy of a common semantic system for words and pictures. Nature, 383, 254-256. Van der Werf, Y.D., Tisserand, D.J., Visser, P.J., Hofman, P.A., Vuurman, E., Uylings, H.B., & Jolles, J. (2001). Thalamic volume predicts performance on tests of cognitive speed and decreases in healthy aging: A magnetic resonance imaging-based volumetric analysis. Cognitive Brain Research, 11, 377-385. Van Gorp, W.G., Satz, P., Kiersch, M.E., & Henry, R. (1986). Normative data on the Boston Naming Test for a group of normal older adults. Journal of Clinical and Experimental Neuropsychology, 8, 702-705. Van Essen, D., Anderson, C., & Felleman, D. (1992). Information processing in the primate visual system: An integrated systems perspective. Science, 255, 419-422. Vannucci, M., Viggiano, M.P., & Argenti, F. (2001). Identification of spatially filtered stimuli as function of the semantic category. Brain Research Cognitive Brain Research, 12, 475-478. Volkow, N.D., Ding, Y.S., Fowler, J.S., Wang, G.J., Logan, J., Gatley, S.J., Hitzemann, R., Smith, G., Fields, S.D., & Gur, R. (1996). Dopamine transporters decrease with age. Journal of Nuclear Medicine, 37, 554-559. Volkow, N.D., Logan, J., Fowler, J.S., Wang, G.J., Gur, R.C., Wong, C., Felder, C., Gatley, S.J., Ding, Y.S., Gitzemann, R., & Pappas, N. (2000). Association between age-related decline in brain dopamine activity and impairment in frontal and cingulate metabolism. American Journal of Psychiatry, 157, 75-80. Wagner, A.D., Pare-Blagoev, E.J., Clark, J. & Poldrack, R.A. (2001). Recovering meaning: left prefrontal cortex guides controlled semantic retrieval. Neuron, 31, 329-338. Warburton, E., Wise, R.J.S., Price, C.J., Weiller, C., Hadar, U., Ramsay, S., & Frackowiak, R.J.S. (1996). Noun and verb retrieval by normal subjects: Studies with PET. Brain, 119, 159-179. Ward, D. (1998). Deconvolution of FMRI timeseries data. In R. Cox MCW AFNI User Manual, pp. 1-24. Unpublished manuscript.

PAGE 152

140 Warrington E.K., & McCarthy, R. (1994). Multiple meaning systems in the brain: A case for visual semantics. Neuropsychologia, 32, 1465-1473. Warrington, E.K., & McCarthy, R. (1983). Category specific access dysphasia. Brain, 106, 859-878. Warrington, E.K., & McCarthy, R. (1987). Categories of knowledge: Further fractionations and an attempted integration. Brain, 110, 1273-1296. Warrington, E., & Shallice, T. (1984). Category specific semantic impairments. Brain, 107, 82 9-854. Welch, L.W., Doineau, D., Johnson, S., & King, D. (1996). Educational and gender normative data for the Boston Naming Test in a group of older adults. Brain and Language, 53, 260-266. West, M.J. (1993). Regionally specific loss of neurons in the aging human hippocampus. Neurobiology and Aging, 14, 287-289. West, R.L. (1996). An application of prefrontal cortex function and theory to cognitive aging. Psychological Bulletin, 120, 272-292. Wierenga, C.E. (2001). Attribute Versus Category-Specific Organization of the Visual Semantic System: An FMRI Investigation. Unpublished master’s thesis, University of Florida, Gainesville, Florida. Wierenga, C.E., Benjamin, M., Perlstein, W.M., & Crosson, B. (2003, March). Semantic Category Identification as a Function of Parametric Spatial Frequency Content. Poster session presented at the annual meeting of the Cognitive Neuroscience Society, New York City, NY. Wiig, E.H., & Secord, W. (1989). Test of Language Competence: Expanded. New York: Psychological Corporation. Wilson, F.A., Scalaidhe, S.P., & Goldman-Rakic, P.S. (1993). Dissociation of object and spatial processing domains in primate prefrontal cortex. Science, 260, 1955-1958. Zarahn, E., Aguirre, G.K., & D'Esposito, M. (1997). Empirical analysis of BOLD FMRI statistics: I. Spatially unsmoothed data collected under null-hypothesis conditions. NeuroImage, 5, 179-197.

PAGE 153

BIOGRAPHICAL SKETCH Christina Elizabeth Wierenga graduated from Calvin College in 1998 with a double major in psychology and English. Christina began her doctoral training in the Department of Clinical and Health Psychology at the University of Florida in 1999 with a concentration in the area of neuropsychology. She received her Master of Science degree in clinical psychology in 2001 from the University of Florida. During her graduate studies, Christina has pursued her interests in language and the semantic system through involvement in various research projects investigating healthy and impaired language functions using functional magnetic resonance imaging (fMRI). Christina has been accepted to the Internship in Clinical Psychology at the University of California at San Diego/San Diego VA Healthcare System where she will continue to develop her knowledge of clinical neuropsychology. 141