THE CONTRIBUTION OF TH E BASAL GANGLIA TO EX PRESSIVE LANGUAGE PERFORMANCE By CHARLES ELLIS, JR 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
Copyright 2005 by Charles Ellis, Jr.
This document is dedicated to my parents the late Hazel L. Joiner and Charles Ellis, Sr.
ACKNOWLEDGMENTS The completion of this dissertation required the encouragement, support and guidance of research committee members, family and friends. I would like to acknowledge the contributions of the many individuals who made this project possible. First, I would like to extend my heartfelt thanks to my research committee. As my graduate advisor and research chair, Dr. John C. Rosenbek provided the direction, insight and encouragement necessary to complete a project of this nature. Dr. Rosenbek demonstrated excellence in mentorship and scholarship while surpassing each of those qualities with his diplomacy and goodwill. Dr. Bruce Crossonâ€™s extensive research of the role of subcortical mechanisms in cognitive functioning served as the inspiration for this project. Dr. Crossonâ€™s systematic approach to research excellence and mentorship will serve as a model for all of my future endeavors. Dr. Leslie J. Gonzalez-Rothi served as a constant source of encouragement and an example of clinical and scientific excellence. I am forever grateful for her ability to constantly stimulate my thoughts beyond my normal capabilities and her challenge to look beyond the current state of published science. At the same time, Dr. Gonzalez-Rothi reminds us all to focus our work in a manner that will positively influence the lives of those we serve. Finally, I am extremely thankful to have Dr. Mary Ellen Young as a research mentor and committee member. Dr. Young challenged me to understand all research methodologies, especially qualitative research and to consider its use as a primary research method as opposed to a supplement to quantitative methods. iv
I am thankful for the participants in this project, particularly our nationâ€™s veterans who remind us that â€œFreedom is never really freeâ€. I am extremely grateful to the VA Office of Academic Affairs for awarding me a VA HSR&D Pre-Doctoral Fellowship for the completion of this project. Also, special thanks to Yvonne Rogalski for assistance with the language analysis of this project. On a personal level, I have been particularly fortunate to have Dr. Richard K. Peach as a long-time mentor and friend. Dr. Peach provided me with support and confidence at a time when I was desperately searching for direction. He has been an integral source of support and encouragement during my doctoral program. I am especially thankful to have had Dr. Maude Rittman as a VA mentor and friend. Dr. Rittman has served as a model for research excellence and mentorship. Dr. Rittman played a critical role in my understanding of the subtleties of scientific writing and programmatic research. I want to thank Dr. Anna Moore, Nan Musson MS, Dr. Michael Okun, Dr. Ramon Rodriguez, Dr. Diane Kendall and Dr. Craig A. Boylstein for insightful guidance during this process. I am forever grateful to my best friends Jarrod and Tess Benton for unconditional love and support throughout the completion of this project and my doctoral progress. Most importantly, I would like to acknowledge and thank my wife Micheala and daughter Hailee for being my primary source of support, encouragement and love throughout my doctoral program and the completion of this project. They serve as a constant source of sunshine on my cloudiest days and are a reminder that a truly fulfilled life consists of much more than academic pursuits. v
TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES .............................................................................................................ix LIST OF FIGURES .............................................................................................................x ABSTRACT .......................................................................................................................xi CHAPTER 1 INTRODUCTION........................................................................................................1 2 REVIEW OF THE LITERATURE..............................................................................5 The Basal Ganglia.........................................................................................................5 Basal Ganglia Anatomy and Circuitry..........................................................................6 Basal Ganglia Models and Signal Enhancement..........................................................8 Fronto-Basal Ganglia Connections and Intentional Behavior....................................12 Fronto-Basal Ganglia Connections and Word Retrieval............................................15 Parkinsonâ€™s Disease and Language............................................................................17 Language Assessment of Language Disorders of the Basal Ganglia.........................23 Discourse Studies........................................................................................................25 Research Questions and Hypotheses..........................................................................31 Research Questions.....................................................................................................31 Hypotheses..................................................................................................................32 3 METHOD...................................................................................................................34 Subjects.......................................................................................................................34 Test Instruments..........................................................................................................35 Mini Mental Status Exam (MMSE)....................................................................35 Boston Naming Test (BNT)................................................................................35 Wechsler Memory Scale â€“ Logical Memory I (WMS-LMI)..............................35 Procedure....................................................................................................................36 Standardized Assessments...................................................................................36 Narrative Discourse Data Collection...................................................................36 Analysis of Narrative Data..................................................................................37 Transcription and segmentation...................................................................37 vi
Productivity analysis....................................................................................39 Coherence analysis.......................................................................................39 Cohesion analysis.........................................................................................40 Scoring reliability.........................................................................................40 4 RESULTS...................................................................................................................42 Demographic Comparisons........................................................................................42 Cognitive and Language Comparisons.......................................................................43 Dysarthria Ratings......................................................................................................43 Scoring Reliability......................................................................................................43 Productivity.................................................................................................................44 Narrative Production Time..................................................................................44 Verbal Cuing.......................................................................................................45 Communication Units..........................................................................................47 Word Production.................................................................................................48 Coherence...................................................................................................................50 Cohesion.....................................................................................................................52 Cohesive Ties......................................................................................................52 Total Percent Correct Cohesive Ties...................................................................55 Parkinsonâ€™s Disease Within Group Comparisons.......................................................57 PD Demographic Data.........................................................................................57 PD Cognitive and Language Comparisons..........................................................58 PD Dysarthria Ratings.........................................................................................58 Narrative Comparisons........................................................................................59 5 DISCUSSION.............................................................................................................60 Productivity.................................................................................................................60 Total Narrative Comparisons..............................................................................61 Three-Minute Comparisons.................................................................................65 Coherence...................................................................................................................67 Cohesion.....................................................................................................................71 Cohesive Ties......................................................................................................71 Percent Correct Cohesive Ties............................................................................71 Intra-Group Comparisons...........................................................................................77 General Discussion.....................................................................................................78 6 TREATMENT CONSIDERATIONS........................................................................81 General Considerations...............................................................................................81 Models of Basal Ganglia Function and Treatment Development..............................83 Exo-evoked Cuing Strategies.....................................................................................84 Concept Manipulation................................................................................................84 Final Considerations...................................................................................................85 7 CONCLUSIONS........................................................................................................86 vii
APPENDIX A INSTRUCTIONS FOR NARRATIVE TASK...........................................................88 B TRANSCRIPTION FORM........................................................................................89 C DYSARTRIA STAGES.............................................................................................90 D GLOBAL COHERENCE RATINGS.........................................................................91 E DEFINITION OF COHESIVE MARKERS AND COHESIVE ADEQUACY.........92 F DEMOGRAPHIC DATA AND COGNITIVE SCORES FOR ALL SUBJECTS.....93 LIST OF REFERENCES...................................................................................................94 BIOGRAPHICAL SKETCH...........................................................................................104 viii
LIST OF TABLES Table page 4-1. Demographic Data for Parkinsonâ€™s Subjects and Controls......................................42 4-2. Cognitive and Language Measures for Parkinsonâ€™s Subjects and Controls.............43 4-3. Intraclass Correlation Coefficients for Scoring........................................................44 4-4. Mean Narrative Production Time (Minutes)............................................................45 4-5. Mean Number of Communication Units..................................................................47 4-6. Mean Words.............................................................................................................49 4-7 Percentage High Coherence.......................................................................................51 4-8. Cohesive Ties...........................................................................................................53 4-9. Percentage Correct Cohesive Ties............................................................................56 4-10. Comparison of Demographic Data for PD Subjects...............................................58 4-11. Cognitive and Language Measures for Parkinsonâ€™s Subjects and Controls...........58 4-12. Comparison of Narrative Measures for PD Subjects..............................................59 ix
LIST OF FIGURES Figure page 4-1. Average Narrative Production..................................................................................45 4-2. Total Number of Verbal Cues for Each Group During Each 1-minute Interval......46 4-3. Average Number of Communication Units Produced for Each Narrative...............47 4-4. Average Number of Communication Units 1-minute intervals.............................49 4-5. Average Number of Words Produced for Each Narrative.......................................49 4-6. Percent High Coherence for Each Narrative............................................................51 4-7. Percent High Coherence 1-minute intervals..........................................................52 4-8. Average Number of Cohesive Ties Produced for Each Narrative...........................53 4-9. Percentage Distribution of Each Cohesive Tie type for Each Group.......................54 4-10. Average Number of Communication Units 1-minute intervals.............................54 4-11. Average Percent Correct Cohesive Ties for Each Narrative....................................56 4-12. Average Percent Correct Cohesive Ties 1-minute intervals..................................56 4-13. Distribution of Incomplete and Erroneous Cohesive Ties.......................................57 x
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 THE CONTRIBUTION OF THE BASAL GANGLIA TO EXPRESSIVE LANGUAGE PERFORMANCE By Charles Ellis, Jr. August 2005 Chair: John C. Rosenbek Major Department: Rehabilitation Science The basal ganglia is a structure known for its role in motor functioning. Disorders of the basal ganglia are typically accompanied by motor deficits and expressive language skills are generally thought to be preserved. Studies of expressive language in individuals with basal ganglia disorders have traditionally used standardized language measures of language form and their results have been mixed. As a result, it has been generally concluded that the basal ganglia has no role in expressive language. However, recent models of basal ganglia functioning suggest a specific role for the basal ganglia in language form and language use. Therefore, the belief that language integrity exists in individuals with basal ganglia disorders has been based primarily on results from studies using language measures of language form and not language use. The analysis of discourse has been hypothesized as a methodology to identify language form and use deficits that may exist following neurological injury or disease. This study tested the hypothesis that individuals with disease of the basal ganglia exhibit language form and xi
use differences in narrative discourse compared to individuals who have no evidence of the disease. Results indicated differences in language use although no significant differences were observed in language form. These results suggest that the basal ganglia has an executive role in expressive language use and that individuals with basal ganglia disorders can exhibit deficits in language use while maintaining relatively unimpaired language form. xii
CHAPTER 1 INTRODUCTION The presence of language disorders due to stroke or diseases of the basal ganglia continue to be a source of debate. Although language disorders following basal ganglia stroke and Parkinsonâ€™s disease (PD) have been described, many believe that the basal ganglia is not involved in language. Some suggest that there is no evidence that the basal ganglia itself has a direct role in language and that white matter pathways near the basal ganglia are primarily critical for language disorders (Alexander, Naeser, & Palumbo, 1987). Nadeau & Crosson (1997) proposed that linguistic deficits following striato-capsular stroke predominately result from cortical hypoperfusion and infarction that is not detected by imaging studies. Their results have been supported by imaging studies (Hillis et al., 2002, 2004; Love, Swinney, Wong, & Buxton, 2002, & Weiller, Willmes, Reiche, & Thron, 1993) in which it was also hypothesized that most cases of language disorders following basal ganglia stroke are most likely due to concurrent cortical hypoperfusion. In contrast, Copland, Chenery, & Murdoch (2000a) suggested that language deficits do in fact occur following basal ganglia disorders and are the result of a disruption of the cortico-striato-pallido-thalamo-cortico system. Copland et al. (2000a) assessed the language skills in subjects with chronic nonthalamic subcortical lesions (NS) with a test battery designed to assess a broad range of language functions. Results revealed that the NS subjects differed from matched controls on the Boston Naming Test, word list generation, recreating sentences, interpreting ambiguous or figurative passages and in 1
2 providing definitions, synonyms, and antonyms. They concluded that deficits in more cognitively demanding aspects of language functions occur after such lesions. In addition, most of the tasks performed involved some degree of cognitive-linguistic flexibility in dealing with novel situations, devising linguistic strategies, and organizing and monitoring responses. As a result the authors suggested that the observed language deficits resembled a frontal lobe type language disorder. Further, the authors noted that standard aphasia batteries are frequently insensitive to such deficits. Additional studies by Copland, Chenery, & Murdoch (2000b) also report word retrieval difficulties following basal ganglia damage. In these studies, individuals with basal ganglia disorders presented with word retrieval deficits reportedly due to a failure to sustain lexical activation via attentional processes. They concluded that these deficits were due to a disruption of fronto-striatal mechanisms. Copland et al. (2000b) also reported qualitatively different language impairments in individuals with basal ganglia damage (Parkinsonâ€™s disease & basal ganglia lesions) compared to those with cortical lesions. As a result of these findings, Copland et al. (2000a, 2000b) and Copland (2003) concluded that the basal ganglia does play a significant role in generative language skills. A number of studies including individuals with Parkinsonâ€™s disease, a disease of the basal ganglia, have also noted deficits in expressive language (Berg, Bjornram, Hartelius, Lasko, & Johnels, 2003; Illes, 1989; Illes, Metter, Hanson, & Iritani, 1988; Murray & Lenz, 2001; Murray, 2000). However, the underlying cause of these language deficits remains unclear. Illes et al. (1988) suggested that expressive language deficits occurring in individuals with PD could result from an adaptation to the progressive disease process. More specifically, Illes hypothesized that observations of syntactic
3 simplification were due to specific linguistic deficits. However, the author also suggested that the reductions in syntactic complexity might be related to motor adjustments for utterance length, indicating a motoric cause of observed expressive language differences. In contrast, Murray (2000) concluded that expressive language deficits in individuals with PD are most likely related to multiple neuropsychological and motor speech changes resulting from the disease process. As a result, the underlying cause of expressive language deficits remains uncertain. Further, some suggest that much of the difficulty in identifying these language deficits lies in the current assessment measures which are designed only to evaluate language structure or form (Copland et al., 2000a). Emerging evidence from fMRI studies also support the notion of a basal ganglia dysfunction in language generation. Crosson et al. (2003) used fMRI to evaluate the frontal, basal ganglia, and thalamic structures engaged in language generation. Results indicated that components of a left pre-SMA-dorsal caudate-ventral anterior thalamic loop were activated during word generation from rhyming or category cues but not non-sense syllable generation. They concluded that the loop was involved in word retrieval. Their findings also suggest a specific role for the basal ganglia in expressive language performance. Although it has been hypothesized that diseases of the basal ganglia affect expressive language performance, it is unknown if diseases of the basal ganglia influence language use. Therefore, the purpose of this study is to evaluate the language form and language use of narrative discourse production in individuals with basal ganglia disease and to compare their abilities with individuals absent of basal ganglia disease. Individuals with PD were chosen for study because they have a disease positioned in the
4 basal ganglia and exclusive of significant cortical involvement. In addition, narrative discourse is a language skill requiring the production of complex verbalizations that follow a predictable pattern of form and use (Chapman et al., 1992). Therefore, an analysis of narrative discourse will provide the opportunity to assess macrolinguistic and microlinguistic aspects of expressive language form and use after basal ganglia disease.
CHAPTER 2 REVIEW OF THE LITERATURE The Basal Ganglia The basal ganglia are a collection of nuclei typically associated with motor performance. The primary components of the basal ganglia are the striatum, the pallidum, the substantia nigra and the subthalamic nucleus. Although expressive language disorders due to basal ganglia lesions have been described, there is a lack of consensus on the role that the basal ganglia has in language functioning (Brunner, Kornhuber, Seemuller, Suger, & Wallesch, 1982; Copland et al., 2000a, 2000b, 2000c; Crosson, 1992a, 1992b, Crosson et al., 2003; Crosson, Zawacki, Brinson, Lu, & Sadek, 1997; Fromm, Holland, Swindell, & Renmuty, 1985; Hillis et al., 2002; Liang et al., 2001; Nadeau & Crosson, 1997; Radanovic & Scaff, 2003; Robin, 1990; Weiller et al., 1993). As a result, many feel that despite being described as related to language, linguistic deficits following lesions to the basal ganglia result primarily from coexisting cortical damage, not the basal ganglia damage itself (Han et al., 2002; Hillis et al., 2002, 2004; Love et al., 2002; Nadeau & Crosson, 1997; Radonnovic & Scaff, 2003; Weiler et al., 1993). In addition, reports of expressive language disturbances following diseases of the basal ganglia such as Parkinsonâ€™s disease have been few and generally inconclusive (Berg et al., 2003; Illes et al., 1988, 1989; Murray, 1999, 2000; Murray & Lenz, 2001). Although the basal ganglia has a relationship with the cerebral cortex via cortical-basal ganglia-thalamus-cortical loops, it is not entirely clear how the basal ganglia complements cortical functioning for expressive language (Parent & Hazrati, 1995). As 5
6 such, a definitive role of the basal ganglia in expressive language functioning continues to be questioned. However, an emerging body of evidence suggests that expressive language disorders may in fact occur following disorders of the basal ganglia. Basal Ganglia Anatomy and Circuitry The basal ganglia consist of a collection of grey matter structures deep in the brain below the cerebral cortex and surrounding the thalamus and hypothalamus (Young & Penney, 2002). Basal ganglia structures include the caudate nucleus, putamen, globus pallidus (internal and external segments), the subthalamic nucleus, and substantia nigra (pars compacta and pars reticulata). The striatum (caudate and putatmen) are the main entry points for the cerebral cortex and thalamus with principal output occurring through the globus pallidus internal segment (GPi) and substantia nigra pars reticulata (SNr). According to Wichmann & DeLong (2003), projections from motor, somatosensory and premotor cortex end in the motor portion of the striatum, the postcommissural putamen. In addition, they note that prefrontal cortical areas project to the caudate nucleus and precomissural putamen while limbic cortices, the amygdala and hippocampus terminate in the ventral striatum. Further, the subthalamic nucleus (STN) is also a termination point for cortical projections from the primary motor cortex (afferents to dorsolateral STN), premotor and supplementary cortex (afferents to the medial third of STN). Additional inputs to the striatum and STN occur from intralaminar nuclei and centromedian nuclei of the thalamus (Wichmann & DeLong, 2003). Subsequently, basal ganglia output is directed through the GPi and SNr to the frontal lobes via the thalamus as well as other brainstem structures. Despite extensive knowledge of these structures, many questions remain about how this region functions (Young & Penney, 2002).
7 In order to clearly understand the potential role the basal ganglia has in expressive language functioning, we must first consider the subcortical circuitry and its relationship to the cerebral cortex. The basal ganglia is thought to receive input from most of the cerebral cortex and then project back to the frontal areas of the cortex via parallel pathways. Alexander, DeLong, & Strick (1986) proposed the multiple cortical-subcortical-cortical circuits exist that are functionally segregated while at the same time having similar anatomical structures. They suggest that these circuits consist of cortical and subcortical structures including the cerebral cortex, striatum, globus pallidum and thalamus. They also note that within this base loop, five distinct circuits exist, which consists of the following: motor circuit, oculomotor circuit, dorsolateral prefrontal circuit, lateral circuit, and anterior cingulate circuit. Each circuit is thought to engage different regions of the basal ganglia and output to different areas of the frontal lobe as follows: the motor circuit is focused upon precentral motor fields; the oculomotor circuit on the frontal eye fields; the prefrontal circuits on dorsolateral prefrontal and lateral orbitofrontal cortex; the limbic circuit on anterior cingulate and medial orbital frontal cortex (Alexander, Crutcher, & DeLong, 1990). The motor circuit has been the most extensively studied and correlated to specific aspects of motor movement. Further, it has been hypothesized that in addition to the five noted circuits, more parallel circuits could exist (Alexander et al., 1986). Alexander et al. (1990) expanded upon his original model and included additional evidence that suggests that each circuit contains highly specialized channels and perhaps sub-channels that allow parallel and multilevel processing of multiple variables concurrently. In addition, each circuit contains a direct and indirect pathway. The direct
8 pathway passes from the striatum directly to one of the basal ganglia output nuclei, while the indirect pathway includes an intermediate relay through the external pallidum and subthalamic nuclei. Therefore, the net result of the direct pathway is a disinhibition of the thalamus while the net result of the indirect pathway is an increased inhibition of the thalamus (Alexander et al., 1990). The end result of the direct and indirect pathways is opposing effects on the influence of the thalamus on the cortex and subsequently cortical control. Basal Ganglia Models and Signal Enhancement The emergence of computational models has begun to provide new insights into the manner in which the basal ganglia operates (Gurney, Prescott, & Redgrave, 1998; Mink, 1996; Nambu et al., 2000, 2002; Prescott, Gurney, Montes-Gonzalez, Humphries, & Redgrave, 2002). Subsequently, a more definitive role in language functioning has been hypothesized. In regards to motor function, Nambu, Tokuno, & Takada (2002) and Nambu et al. (2000) proposed that the basal ganglia operates via a dynamic â€œcenter surroundâ€ mechanism by way of the direct and indirect motor pathways. This process serves to inhibit competing motor programs while facilitating the desired motor program. In the latest model, Nambu et al. (2002) proposed that the basal ganglia operates via three pathways: hyperdirect, direct and indirect. The hyperdirect system activates when a movement is about to be initiated by cortical mechanisms and inhibits areas of the thalamus and cortex. Therefore, it serves as a primary inhibition of selected and competing signals. The direct pathway then activates the desired signal while inhibiting all competing programs. Finally, the indirect pathway serves as the mechanism to inhibit all competing signals or programs that are also attempting to be activated and expressed. Ultimately only the desired motor program is initiated. Prescott et al. (2002) further
9 described this process and defined the role of the basal ganglia as an â€œaction selection deviceâ€. He noted that it is the responsibility of the â€œaction selection deviceâ€ to select one â€œactionâ€ from a multitude of competing â€œactionsâ€, each potentially carrying a request for an action to be taken. Further support for a specific role of the basal ganglia in â€œaction selectionâ€ was provided by Mink (1996) when he suggested that the basal ganglia serves as a mechanism for â€œfocused selection and inhibitionâ€. In â€œfocused selection and inhibitionâ€, the basal ganglia acts as a brake for motor programs in the cortex and thalamus. Thus, when a motor program is initiated, the basal ganglia output neurons of the competing motor programs increase their firing rate, resulting in a â€œbrakeâ€ for the competing programs. As a result, the desired motor program is released while the competing programs are halted. In addition to the braking mechanism proposed by Mink, others have proposed that the basal ganglia circuitry serves as a system that creates a contrast between stronger and weaker signals (Gurney, Prescott, & Redgrave, 1998). Subsequently, the enhancement of the desired signal may arise through recurrent reciprocal inhibition of competing signals. This process occurs as all â€œbidsâ€ for an activity receive continuous input, forming a common currency for each bid. Each request is then compared via the internal circuitry of the basal ganglia, which in turn determines the â€œwinnerâ€, or signal that is ultimately processed or initiated. As a result, the â€œlosersâ€ remain inhibited (Gurney et al., 1998). The concept of â€œfocused selection and inhibitionâ€ has typically been applied to motor function, however recent evidences suggest that the basal ganglia influences word retrieval and ultimately expressive language in the same manner (Copland et al., 2000b, 2000c; Copland, 2003; Crosson et al., 2003).
10 The enhancement of internal signals in the basal ganglia for the selection of one program over another or as Mink defined â€œfocused selectionâ€ is thought to primarily occur via the neurotransmitter dopamine. Dopamine is believed to play a critical role in the tuning of the activity of cortico-subcortico-cortico circuits in its contribution to â€œdrive of actionâ€ and the ability to perform â€œcontinuous adaptationâ€ (Nieoullon, 2002). It has also been suggested that dopamine modulates target neuron responses by increasing the signal-to-noise ratio of neurons firing in relationship to baseline neuronal firing (Cohen, Braver, & Brown, 2002). Two receptor types reside in the striatum: D1 and D2, with D1 receptors having an excitatory effect and D2 receptors having an inhibitory effect (Riederer, Gerlach, & Foley, 2002). Therefore, dopamine has an excitatory effect on D1 receptors in the striatum and inhibitory effects on D2 receptors. In addition to having contrasting influences on different receptor types, dopamine also interacts with acetycholine (Ach) in the process of modulating neurons in the striatum. Rolls, Thorpe, Boytim, Szabo, & Perrett (1984) analyzed the effects of iontophoretic application of dopamine on the responsiveness of the neurons of one key component of the basal ganglia, the striatum, to their normal inputs in the behaving monkey. Results indicated that dopamine enhanced the magnitude of the neuronal responses to arm movements relative to their spontaneous activity. It was thought that the effect of dopamine was to increase the signal to noise ratio of the task related firing neuron by holding spontaneous background activity to a reduced level. Further, results suggested that dopamine might act to influence transmission by increasing the magnitude of the responses or by setting the threshold. Supporting such evidence, Chiodo and Berger (1986) used iontophoretic techniques to examine the effects of dopamine on other
11 neurotransmitters in the striatum of rats. They noted that at certain levels, dopamine served to modulate the efficacy of other neurotransmitters resulting in overall enhancement the firing rates of striatal neurons. Additional studies of rats by Pierce and Rebec (1995) and Kiyatkin and Rebec (1996) also suggest that dopamine release may enhance information transmission through the basal ganglia. However, questions remain as to whether the basal gangliaâ€™s exact role via dopamine interactions is to inhibit competing stimuli as described by Mink (1996) or to primarily enhance or â€œboostâ€ the desired signal for response and output (Stout, Wylie, Simone, & Simers, 2001; Taylor and Saint-Cyr, 1995). In support of Mink (1996) and Nambu et al. (2000, 2002), Nieoullon and Coquerel (2003) proposed that dopamine might also act as a â€œbrain conductorâ€ in regards to organization of behavior. They noted that dopamine is in a position in the forebrain to â€œtuneâ€ activity of the multiple cortico-subcortico-cortico loops and therefore contribute to the drive of action and continuous adaptation of behavior. Further, they concluded that dopamine has a critical role in integrative processes. Dopamine has the ability to modulate the activity of the parallel and segregated cortico-subcortico-cortico circuits not only at the basal ganglia and cortical levels, but also within specific basal ganglia structures themselves. Therefore, some suggest that dopamine in conjunction with prefrontal cortex systems are critical for the control of thought and behavior (Cohen et al., 2002). To date, it is unclear if focused selection and inhibition and signal enhancement via dopamine are individual components of basal ganglia function or a coordination of two distinct processes. Therefore, further study of the basal gangliaâ€™s
12 integrative influence of cognitive processes is needed, particularly those that support language. Fronto-Basal Ganglia Connections and Intentional Behavior In addition to its role in signal enhancement and action selection, the basal ganglia has widespread connections to the frontal lobe, which plays a critical role in intentional behavior. Intention is the selection of one action for initiation among multiple competing actions (Crosson, et al., 2005). The intentional system is believed to consist of the dorsolateral and medial frontal lobe in conjunction with the anterior cingulate gyrus and basal ganglia loops (Heilman, Watson & Valenstein, 2003). Intention is thought to influence language by facilitating word retrieval and speech initiation (Crosson et al., 2005). Therefore, studies of basal ganglia disorders and language production must consider: (1) the extensive number of basal ganglia connections with the frontal lobe, (2) the role of the frontal lobe in intentional behavior such as language and (3) the potential influence of a disruption of fronto-basal ganglia connections on expressive language. Basal ganglia loops that are critical to cognitive tasks such as the generative aspects of expressive language, begin and target the frontal lobes (Middleton & Strict, 2000). Many individuals with frontal lobe injuries exhibit a limited or decreased ability to plan short and long term future behaviors such as language (Damasio & Anderson, 2003). In addition, there is a relationship between planning deficits in individuals with frontal lobe injuries and the temporal sequencing demands of planning (Owen, 1997; Sirigu et al., 1995). However, the relationship between frontal lobe injury, planning deficits and expressive language is not entirely clear at this time. Alexander (2002) proposed that language impairments that occur following frontal lobe damage could be quantified as action planning deficits specific to language use. He noted that a frontal type language
13 impairment could be characterized as â€œdisrupted action planningâ€ resulting in a fundamental impairment of complex, goal directed intentional behavior. He also proposed that the overall action plan includes the selection and use of lexical items as part of the process. Therefore, a language â€œaction planâ€ includes the ability to simultaneously generate and monitor simultaneously the complex task, or in this case â€œexpressive language,â€ that will unfold in the future (Alexander, 2002). As a result, it has been suggested that faulty â€œaction plansâ€ or â€œdrive of actionâ€ result in expressive language disorders (mutism and adynamic aphasia) following frontal lobe damage or a disruption of the fronto-basal ganglia circuit (Costello & Warrington, 1989; Gold et al., 1997; Robinson, Blair & Cipolotti, 1998). While basal ganglia disorders can occur in the absence of frontal lobe damage, it has been reported that damage specific to frontal lobe-basal ganglia circuitry may also affect cognitive processes that are generally thought to be influenced by frontal function (Cools, Stephanova, Barker, Robbins, & Owen, 2002; Dagher, Owen, Boecker, & Brooks, 2001; Owen, Doyon, Dagher, Sadikot, & Evans, 1998). This can occur following a disruption of the multiple and extensive reciprocal connections that exist between the frontal lobe and basal ganglia. Therefore, generative or intentional tasks such as expressive language may be affected as a result of only disrupted frontal lobe-basal ganglia circuitry as opposed to damage specific to the frontal lobe itself. However, it is not entirely clear how basal ganglia diseases that affect basal ganglia circuitry, influence intentional language behavior. To further investigate this issue, Owen et al. (1998) examined subcortical blood flow changes during tasks known to involve subcortical circuitry. They reported reduced
14 cerebral blood flow in the basal ganglia in individuals with PD, a disease of the basal ganglia, during planning tasks. They concluded that basal ganglia dopamine depletion disrupts the normal pattern of outflow in individuals with PD, subsequently interrupting the normal transmission of information through fronto-basal ganglia circuitry. Therefore, they hypothesized that the ability of the basal ganglia to modulate cortical excitatory input to the striatum is reduced. This reduction results in an abnormal influence on the globus pallibus internal segment (GPi) via the direct and/or indirect pathways during planning or spatial working memory tasks. As a result, abnormal processing of neuronal activity is present in the basal ganglia. They concluded that â€œfrontal cognitiveâ€ deficits observed in individuals in the early stages of basal ganglia diseases are the result of this circuitry disruption. These findings have critical implications for expressive language, since significant planning and online executive control are required for the development of logical, coherent, and cohesive verbal output. Dagher et al. (2001) observed similar findings and concluded that disruption of fronto-basal ganglia circuitry resulted in â€œfrontal typeâ€ deficits in individuals with PD. They suggested that cognitive deficits in PD could be the result of inhibition of the pre-frontal cortex neurons due to GPi overactivity within the dorsolateral cortico-striatal loop. They also proposed that the observed cognitive deficits could result from an alteration in the pattern of activity in the basal ganglia rather than the net output of the basal ganglia itself. Disruptions of basal ganglia circuitry may occur following basal ganglia diseases and result in a reduction in its ability to serve as the â€œbrain conductorâ€ (Nieoullon & Coquerel, 2003). Interestingly, the basal gangliaâ€™s role as â€œbrain conductorâ€ closely parallels that of the frontal lobes role in executive functioning (Owen et al., 1995). Owen
15 et al. (1995) compared individuals with PD on and off medications with patients with frontal lobe incisions on tests of planning and spatial working memory. Results indicated similar planning deficits in individuals with PD in the â€œoffâ€ medication state compared to those with frontal lobe incisions. However, they noted that there were some minor differences in the manner in which planning function in PD and frontal incisions was impaired. In their test of planning, those participants with frontal lobe lesions required more attempts to identify solutions during more challenging problems. However, their response latencies were not significantly prolonged. In contrast, the participants with PD required prolonged response latencies although their planning accuracy was preserved. As a result, they hypothesized that accuracy and efficiency of planning is mediated by mechanisms within the fronto-striatial circuitry, while speed of thinking may be modulated by neurotransmitter systems that innervate the cortex, caudate nucleus or both. These findings agree with Morris et al. (1998) and Taylor, Saint-Cyr, & Lang (1986) who also suggested distinctions in the exact anatomical location of fronto-basal ganglia disruptions in tasks that require self directed planning. Therefore, the basal ganglia and frontal lobe may perform some parallel and integrative functions related to planning. This would suggest that the fronto-basal ganglia circuitry itself is critical to planning and executive behavior. In addition, disruptions of the frontal lobe, basal ganglia or fronto-basal ganglia circuitry could potentially contribute to decreased planning behavior and influence multiple cognitive processes such as expressive language. Fronto-Basal Ganglia Connections and Word Retrieval Findings from Owen et al. (1995) suggest a critical relationship between the frontal lobe and basal ganglia for word retrieval. Recent studies also indicate that the frontal lobe and basal ganglia have a direct influence on expressive language via word retrieval
16 (Crosson et al., 2003; Damasio & Tanel, 1993). Findings from studies of word retrieval by Damasio and Tranel (1993) suggest that words such as objects and actions are represented in different mental lexicons; each vulnerable to brain damage in specific locations. They propose that a large cerebral network, which includes the left frontal cortices, is critical for the mediation of systems essential for word retrieval of verbs. They suggest that the frontal lobes serve to mediate the stage between the processing of a concept and vocalization of verb word forms. More importantly, they note that systems that mediate access to verbs are anatomically close to those that support concepts of movement and relationship to space-time. These findings are in contrast to systems that mediate nouns, which are close to systems that support concepts for concrete entities. The relationship between basal ganglia disorders and verb retrieval deficits is not entirely clear. However, since the basal ganglia is frequently related to movement deficits, this hypothesis may have some relationship to word retrieval in diseases or lesions of the basal ganglia. Additional study of verb naming by Grossman (1998) suggests that verb-naming deficits might reflect an underlying executive system deficit that prohibits the ability to mentally coordinate and manipulate the wide range of information associated with the verb. Peran et al. (2003) noted similar deficits in verb production in individuals with PD and proposed that verbal production deficits result from an impairment of accessing linguistic and mental representations of action. Finally, Baldo and Shimamura (1998) studied individuals with frontal lobe lesions and noted that the frontal lobes contribute significantly to the retrieval of lexical-semantic knowledge and thus result in reduced word retrieval when damaged. An earlier study by Shimamura (1994) proposed that
17 these deficits might in fact be associated with a failure in inhibiting and filtering inappropriate or extraneous information processing. Therefore, during word fluency tasks, some of the reduction in ability may be related to an inability to filter through multiple choices for the correct word selection. A number of imaging studies of word generation have also indicated a significant role for the frontal lobe and basal ganglia in word retrieval (Crosson et al., 1999, 2001, 2003, 2005). Crosson et al. (2003) noted frontal lobe and basal ganglia activity during generation of words from a rhyming or category cue but not during nonsense syllable generation. Their findings indicate that the fronto-basal ganglia loop is involved in word retrieval from pre-existing lexical representations. Parkinsonâ€™s Disease and Language Much of our understanding of language deficits due to basal ganglia disorders or disease can be attributed to the Alexander et al. (1986) proposal of the cortico-striato-pallido-thalamo-cortico circuits. While no specific circuit has been identified to facilitate expressive language, the dorsolateral prefrontal or â€œcognitive circuitâ€ has been most often used to explain language deficits due to basal ganglia disorders. Therefore, expressive language disorders due to basal ganglia involvement have been attributed to a disruption of the â€œcognitive circuitâ€ (Copland, 2003; Copland et al., 2000a, 2000b; Gold et al., 1990; Mega & Alexander 1994). Parkinsonâ€™s disease is one disorder of the basal ganglia of which language disorders have been hypothesized. Idiopathic Parkinsonâ€™s disease (PD) is a progressive, neurodegenerative disorder that results from a deficiency of the neurotransmitter dopamine in the basal ganglia. Dopamine is manufactured in one of the basal ganglia nuclei, the substantia nigra, and is transmitted via nigrostriatal tracts to the striatum
18 (Duffy, 1995). Dopamine is thought to be a modulatory neurotransmitter that acts as a reinforcement signal or serves to enhance signal strength of neurons in the striatum (Riederer, Gerlach, & Foley, 2002). As a result of the dopamine deficiency in PD, there is an imbalance in the involvement of the direct and indirect systems of the basal ganglia. Therefore, there is an overactivation of the indirect system due to a reduction of dopamineâ€™s influence on the Gpi. This overactivation results in a greater inhibition of the indirect system. This occurs in concurrence with an underactivation of the direct system, resulting in a failure of the direct systems ability to disinhibit the thalamus and activate frontal cortical areas. PD is characterized by such motor symptoms as bradykinesia, rigidity, tremor and loss of natural reflexes (Golbe, 1998). These deficits occur as a result of an imbalance of dopamine and other neuro-transmitters in the basal ganglia that are important for motor control. Additional symptoms may include visospatial dysfunction, memory disorders, executive functioning deficits, as well as mood and behavioral changes (Dubois & Pillon, 2002). Parkinsonâ€™s disease has a significant influence on speech production and is typically associated with hypokinetic dysarthria. Hypokinetic dysarthria is characterized by deficits primarily in voice, articulation and prosody but may include deficits in respiratory, phonotory, resonatory and atriculatory levels of speech (Duffy, 1995). While motor speech disorders are typically associated with PD, it has generally been noted that individuals with PD do not present with any of the classical aphasia syndromes (Grossman, 1999). However, receptive language impairments have been reported in individuals with PD. Disorders of sentence processing have been frequently noted in PD and are generally regarded as the result of limitations in cognitive resources
19 such as selective attention and information processing (Friederici, Kotz, Werheid, Hein, & von Cramon, 2003; Grossman et al., 2002a; Grossman, Lee, Morris, Stern, & Hurtig, 2002b; Grossman, 1999; Murray & Stout, 1999; Ullman et al., 1997). Grossman (1999) noted that as individuals with PD exhaust their cognitive resources, they have greater difficulty understanding sentences. He concluded that reductions in brain dopamine levels limit the distribution of cognitive resources and influence comprehension skills. Further, Grossman (1999) indicated that the interruption of the fronto-striatal system significantly influences sentence processing in individuals with PD. In addition, while many individuals with PD frequently perceive main information in spoken discourse, they exhibit difficulty processing detailed or implied information (Murray & Stout, 1999). As a result, these deficits may significantly influence the professional and social lives of individuals with PD (Murray & Stout, 1999) Despite these findings, expressive language disorders have generally not been associated with PD. However, a number of studies of language functioning following PD and the emergence of new models of basal ganglia functioning suggest that PD may have greater influence on expressive language than previously reported. Verbal fluency is a test frequently used to provide a baseline measure of word retrieval. Verbal fluency consists of a time generated sequence of multiple response alterations under constrained search conditions requiring retrieval of words based upon phonemic or semantic criteria (Henry & Crawford, 2004). Auriacombe et al. (1993) noted that individuals with PD exhibit deficits in verbal fluency due to lexical retrieval impairment. They found that selective impairment in category naming was present in non-demented PD participants while their performance on other fluency tasks was quantitatively similar to that of the
20 non-diseased control subjects. Henry and Crawford (2004) completed a meta-analysis of sixty-eight studies, including more than 4600 participants, to evaluate the sensitivity of test of verbal fluency in individuals with PD relative to normal controls. Results indicated that individuals with PD exhibited greater impairment on semantic fluency than phonemic fluency tasks. They concluded that since those with PD had similar scores on the Boston Naming Test that imposes minimal demands on cognitive speed and effortful retrieval, the reduction in verbal fluency most likely represented a deficit in semantic memory. While fluency deficits have typically been accounted for by a reduction in language supporting cognitive processes, it may be hypothesized that decreased naming fluency in PD may be related specifically to lexical retrieval. Extensive study of verbal fluency in individuals with PD has suggested multiple causes of the deficits in fluency tasks. Troster, Woods, Fields, Hanisch, & Beatty (2002) proposed that cortico-basal ganglia circuits are important for set shifting, cognitive flexibility, and initiating and maintaining effective word search and retrieval. Therefore, individuals with diseases of the basal ganglia exhibit a reduced ability to complete fluency verbal fluency tasks due to the aforementioned cognitive deficits. Results from studies by Peran et al. (2003) support Trosterâ€™s findings and they propose that difficulties with verb generation in PD reflected frontal dysfunction. They suggested that individuals with PD have difficulty shifting between noun and verb retrieval, a task that requires significant cognitive flexibility. They concluded that cognitive flexibility is typically reduced in individuals with fronto-striatal dysfunction. Finally, Piatt, Fields, Paolo, Koller, & Troster (1999) proposed that deficits in â€œactionâ€ verbal fluency, which is
21 frequent in individuals with PD, reflect frontostriatal neuropathology and/or neurochemical deterioration that occurs in PD. Early studies of language following PD completed by Illes et al. (1988) compared the acoustic and linguistic features of individuals with PD to normal controls. Although linguistic measures differentiated PD from the normal controls in a) silent hesitations per minute, b) abnormally long silent hesitations, c) words per silent hesitation, d) open class phrases, e) optional open class phrases, and f) decreases in the number of interjections, the differences in language structure were attributed primarily to an adaptation to the disease process rather than language specific impairment. Recently, studies of language in individuals with PD have noted that â€œhigh level expressive language deficitsâ€ are frequently present (Berg et al., 2003; Murray & Lenz, 2001; Murray, 2000) however the underlying cause of such deficits remains a source of debate. Berg et al. (2003) noted that individuals with PD exhibit deficits in processing implied information. These deficits resulted in a reduced ability to make inferences and to analyze and recreate sentences. In addition, their performances on high-level language assessments are strongly correlated with cognitive performance. It is unclear whether their language performance resulted from disruption of other cognitive processes supporting language or specific disruption of language functioning itself. Murray and Lenz (2001) examined the syntactic changes in spoken discourse in patients with Huntingtonâ€™s (HD) and PD to explore possible relationships between their syntactic changes and concomitant cognitive and motor changes. Despite finding deficits in productive syntax, they concluded that changes in language production abilities reflect concomitant cognitive and speech impairments rather than specific language deficits.
22 Further, Murray (2000) concluded that the spoken language abilities of individuals with PD are related to a variety of neuropsychological and motor speech changes. Although the concepts of focused selection and inhibition have primarily been used to describe motor programming via the basal ganglia, similar modeling has been used to hypothesize the role of the basal ganglia in expressive language in individuals with PD. Studies by Copland et al. (2000b, 2000c) and Copland (2003) suggest that the basal ganglia serves as an action selection device via the cortico-striato-thalamo-cortico pathway during language functioning. Copland (2000b) evaluated a group of individuals with PD and basal ganglia stroke with priming tasks and concluded that damage to the basal ganglia resulted in decreased priming for word retrieval tasks. The lexical ambiguity priming tasks included in the studies were based on the exhaustive access theory (Simpson, 1984). This theory suggests that when ambiguous words are presented, all possible meanings are brought on line for consideration and selection. Subsequently, a stage of meaning selection or inhibition occurs via attentional or controlled processing. Thus, priming tasks allow inferences to be made regarding semantic memory, its access, and the spread activation within the lexical-semantic network of interconnected lexical items and semantic concepts (Collins & Loftus, 1975). In addition, attentional mechanisms are also activated and required to sustain proper activation of the semantic targets allowing the investigator to hypothesize about the integrity of the attentional system. As noted in the computational models, Copland (2000b, 2000c) concluded that the basal ganglia serves to enhance selected language signals (words, language segments) while inhibiting competing signals. Also, he noted that the process of enhancement and inhibition of language signals is disrupted following
23 damage or disease to the basal ganglia as proposed in motor models (Gurney et al., 1998; Mink, 1996; Nambu et al., 2002, 2000; Prescott et al., 2002). More specifically, he suggested that damage to the basal ganglia resulted in a disruption of the fronto-striatal mechanisms that mediate attention allocation and strategy formation for language. Therefore, â€œaction selectionâ€ may be temporally disrupted as evidenced by studies that have resulted in decrease priming ability as the interstimulus intervals (ISIs) increases. It has also been noted that individuals with PD exhibit alterations in executive-type behavior that require continual updating of information (Nieoullon, 2002). Nieoullon proposed that such deficits may be related to the working memory that contributes to reasoning by storing information for a short period of time, while taking into account rapid changes in the external context. While this hypothesis has also been applied to motor behavior, it may be proposed that similar deficits may occur during the word retrieval process and disrupt the generation of expressive language. As a result, language deficits due to basal ganglia disorders may be related to an inability to maintain attention and/or memory processes for the efficiency and/or effectiveness of the action selection system for language. Further, it may be hypothesized that the end result of these disrupted attentional and memory processes that influence language, is a reduction in the efficiency of word retrieval strategies and ultimately expressive language output. Language Assessment of Language Disorders of the Basal Ganglia Recent evidence suggests much of the difficulties characterizing language disorders following basal ganglia diseases are the result of inadequate methodologies used to assess language. Since most studies of language following basal ganglia disorders have consisted of â€œoff-lineâ€ measures of language form such as confrontational naming and word list generation, the multiple processes that are required (linguistic and non
24 linguistic) to complete such tasks are not properly considered (Copland, 2000b). More specifically, Copland (2000b) suggested that language disorders of the basal ganglia are not due to deficits in semantic representations but reflect attentional breakdowns in retrieval and manipulation of semantic information. Therefore, these disruptions may not be easily identifiable. In addition, deficits in language use are generally not evaluated and considered. As a result, traditional approaches to the assessment of language disorders after basal ganglia disorders have failed in their ability to consistently capture the disruptions in expressive language. Therefore, the literature provides support for an analysis of the discourse with individuals with basal ganglia damage. Discourse analyses may further characterize expressive language form and language use after such basal ganglia disease or lesions. Discourse is a high level skill that places greater demands on word retrieval, working memory, and attentional control as sentences are developed, while maintaining online evaluation of output. Prior studies of discourse have concluded that language productivity, language efficiency and language coherence differ when comparing brain injured individuals with normal populations (Glosser & Deser, 1990; Wilson & Proctor, 2000; Wilson, Smith & Proctor, 2001). Therefore, an evaluation of the discourse of individuals with basal ganglia damage will allow the investigator to study the expressive language system at which time the greatest demands are being placed on the system. Further, this may provide a means through which to evaluate the contribution of the basal ganglia in language and hypothesize its specific role in expressive language production. This would contrast the aforementioned studies of language production (Illes et al., 1998; Illes, 1989) which suggested that language deficits are an intrinsic part of the disease
25 process, while offering no specific neurolinguistic or mechanism focused hypothesis of the contribution of the basal ganglia to the observed language deficits. Discourse Studies Discourse is the production of complex and structured verbal output that follows an anticipated form (Chapman et al., 1992). Discourse requires some knowledge of what the listener knows and expects (Alexander, 2002). Alexander (2002) suggests that some forms of discourse are complex verbal schemas, which represent plans of action in the verbal domain on a much larger scale than the individual sentence. Discourse has been traditionally elicited by using picture description tasks (Brenneise-Sarshad, Nicholas, & Brookshire, 1991; Correia, Brookshire, & Nicholas, 1990; Doyle et al., 1998; Nicholas & Brookshire, 1995; Potechin, Nicholas, & Brookshire, 1987). Discourse obtained from picture descriptions typically includes only people, objects and actions that may not be related by inference (Olness, Ulatowska, Wertz, Thompson, & Auther, 2002). Therefore, such tasks may not be representative of normal communicative functioning. Studies of discourse by Shadden, Burnette, Eikenberry, & DiBrezzo (1991) and Smith, Heuerman, Wilson, & Proctor (2003) support this notion, as participants spoke more during narrative discourse tasks than pictured activity descriptions. On the other hand, narrative discourse allows the speaker the ability to express temporal progression, establish and maintain personal reference and highlight certain events above others, leading to a more representative pattern of normal communication (Ulatowska & Olness, 1997). Because traditional tests of language appear not to detect or adequately characterize the nature of language deficits in individuals with basal ganglia involvement, an analysis of narrative discourse may allow
26 more consistent identification of communication deficits associated with Parkinsonâ€™s disease. Narrative discourse is constructed via cognitive processes that enable the speaker to produce a temporally organized flow of verbal information in a coherent and cohesive manner. Therefore, discourse organization consists of the interaction of the cortical and subcortical language structures and other cognitive processes supporting language. If this notion is correct, we might conclude that one source of cognitive disruption influencing the linguistic system may be the basal ganglia. This may occur following damage to the basal ganglia due to reduced efficiency and effectiveness in its role in â€œaction selectionâ€. Further, it has been suggested that the basal ganglia and frontal cortex have a role in the temporal order of sequences of events (Beiser & Houck, 1998). Beiser and Houck (1998) developed computational models of cortical-basal ganglionic functioning to study visual processing in the cortico-basal ganglia circuits. However, they suggested that circuits linking the basal ganglia, thalamus and cortex have the capacity to encode other sensory or internal events in a temporal fashion via alternate cortical-basal ganglia loops. Therefore, these findings suggest that the temporal aspects of discourse may require an efficiently functioning basal ganglia to maintain adequate flow of coherent, cohesive and temporally guided information. As a result, temporal disruptions in expressive language may be evident in a comparison of the discourse of individuals with Parkinsonâ€™s disease when compared to normal controls. One aspect of narrative discourse that relies on temporal organization is global coherence. Coehlo and Flewellyn (2003) define global coherence as the relationship of the meaning or content of an utterance to the general topic. It is not known how the basal ganglia influences global coherence during narrative
27 discourse. It could be hypothesized that fronto-basal ganglia disruption may potentially disrupt coherence maintenance during narratives. It is expected that a comparison of individuals with PD to normal controls could capture such disruption. In addition to coherence, discourse contains characteristics of cohesion, the linking of meaning across sentences thorough the use of cohesive markers (Halliday & Hasan, 1976). Cohesion is a semantic concept in that it refers to relations of meaning that exist within a passage, spoken or written, and define it as a passage (Halliday & Hasan, 1976). Glosser (1993) notes that discourse cohesion occurs via linguistic devices that are used to index the interconnections of multiple segments of the discourse. More specifically, cohesion occurs in narrative discourse via the use of cohesive markers, which are words that direct the listener to information found outside individual sentences (Halliday & Hasan, 1976). Therefore, cohesion is a semantic relationship between an element in the passage and some other element that is crucial to its interpretation. A cohesive marker creates a tie with the information found outside the sentence thus establishing a meaning relationship across sentences within the passage (Coelho, 1995). Van Leer and Turkstra (1999) noted that ambiguous intersentential meanings occur when the information that completes a markerâ€™s meaning is not readily apparent. As a result of this disrupted cohesion, functional communication is compromised and an increased listener effort is required to interpret the discourse. Prior studies of discourse production have suggested that disrupted cohesion may reflect impaired lexical retrieval rather than impaired intersentential organization (Gloser & Deser, 1990). Therefore lexical retrieval must be considered when more clearly defining the root cause of reduced use of complete cohesive ties.
28 Halliday and Hasan (1976) proposed a methodology for cohesion analysis and noted that the basic concept that is employed in analyzing cohesion of a passage is that of the cohesive tie. The tie includes the cohesive element in addition to that which is presupposed by the cohesive element. Five categories of cohesive elements or markers were defined by Halliday and Hasan (1976): Reference, Substitution, Ellipsis, Conjunction and Lexical. Reference consists of personal, demonstrative and comparative pronouns (e.g. The car belongs to â€œhimâ€). Substitution is a relation in the wording rather than meaning. Substitutions are alternate words used in the place of a repetition of a particular item (e.g. My pencil is broken. I need a new â€œoneâ€). Ellipsis is the omission of an item (e.g. Did you hear the local news? No, only the weather). Conjunctions are cohesive indirectly as they express certain meanings that presuppose the presence of other discourse components (e.g. The game was over at three. After the game we went for ice cream). Lexicals or general nouns are cohesive by selection of vocabulary (e.g. James ran into the street. The moving car didnâ€™t seem to scare the man). Van Leer and Turkstra (1999) suggested that in order to objectively quantify the cohesiveness of narrative discourse and determine how well the speaker can maintain meaning across discourse, the cohesive adequacy must be evaluated. Liles (1985) proposed a method based on Halliday and Hasanâ€™s (1976) procedure to address cohesive adequacy in discourse. According to the Liles procedure, cohesive ties are classified as complete, incomplete or erroneous. Only cohesive ties in which the referent is easily found in the preceeding test is deemed complete. Further, cohesive ties are judged incomplete when the information is not found in the test and erroneous if the listener if guided to ambiguous information.
29 Currently, there are no established norms for cohesive adequacy in the narrative discourse of non-brain injured adults. However, studies including normal controls suggest a high rate of cohesive adequacy in those participants absent of brain injury. Coelho (2002) evaluated the percentage use of cohesive ties during a story retelling task with a group of adult traumatically brain injured adults and non-brain injured adults serving as controls. He noted that the normal controls produced complete ties 94% of the time while the traumatically brain injured adults produced ties 90% of the time. In addition, a study of individuals with traumatic brain injury (TBI) by Liles, Coelho, Duffy, & Zalagens (1989) included twenty-three normal controls ages (18-22) that produced complete cohesive ties at a rate of 98%. While there have been studies of language production (Illes et al., 1988; Illes, 1989; Murray & Lenz, 2001; Murray, 2000), there are no known studies of the role the basal ganglia plays in discourse coherence or the use of cohesive markers. It would be expected that the linkage of meaning among sentences is a cognitive skill potentially influenced by the basal ganglia. If this hypothesis is correct it would expected that cohesion and coherence may be disrupted following disorders of the basal ganglia and disruption of the cortico-striato-pallido-thalamo-cortico system. It would also be expected that these disruptions in the use of cohesive markers can be captured in a comparison between the discourse of individuals with Parkinsonâ€™s disease and normal controls. Therefore, the purpose of this study is to evaluate the language form and language use of narrative discourse of individuals with diagnosis of idiopathic PD and compare their narrative discourse abilities to those without evidence of disease. Comparisons of
30 language productivity, coherence and cohesion will be made to evaluate language form and use. Individuals with idiopathic PD were chosen because they are homogeneous in regards to their disease of the basal ganglia, therefore making them an ideal population to study. In addition, they have a disorder of the basal ganglia that may influence their expressive language performance. Further, individuals with PD have a reduction in the neurotransmitter dopamine in the basal ganglia where dopamine concentrations are greatest in the brain (Skeel et al., 2001). It should be noted that dopamine depletion is relatively low in the frontal lobe in individuals with PD making it unlikely that frontal lobe dysfunction alone could account deficits that may occur in high level tasks such as narrative discourse (Cools, Stefanova, Barker, Robbins, & Owen, 2002). As a result, reductions in dopamine levels would be expected to primarily influence signal enhancement in the basal ganglia rather than the cerebral cortex. Therefore, the influences of the basal ganglia on language, a cognitive task thought to primarily be limited to the left cerebral cortex can be evaluated. In addition, the basal ganglia has extensive connections with the frontal lobe, which also has a significant influence on intentional tasks such as expressive language. As a result, basal ganglia disorders may disrupt frontal-basal ganglia functions and result in impairments of learned cognitive operations or â€œaction plansâ€ needed for the mobilization of complex language or intentional behavior such as discourse (Alexander, 2002). Therefore, since the basal ganglia has a widespread connections to the frontal lobe and the frontal lobe has a significant contribution to intentional and generative aspect of language, the temporal and generative characteristics of narrative discourse may provide insights into the manner in which disruption of the basal ganglia and frontal lobe
31 connections influence narrative discourse production. More specifically, the evaluation of narrative discourse in subjects with idiopathic PD can potentially reveal how the disruption of fronto-basal ganglia pathways manifest in temporally based tasks such as narrative discourse. More importantly, since the creation of narrative discourse requires support from intentional and executive systems, a disruption of these systems should influence expressive language. Finally, the evaluation of narrative discourse could provide insights into the basal gangliaâ€™s role in the intentional language system and how a disruption of the basal ganglia manifests in language productivity, global coherence and cohesion. Research Questions and Hypotheses Since it has been hypothesized that the basal ganglia: (1) is part of the intentional system, (2) performs â€œaction selectionâ€ and (3) serves to enhance signals via dopamine in a temporally sensitive manner and language productivity, coherence and cohesion are temporally distributed in narrative discourse my research questions are as follows: Research Questions 1. Do individuals with PD exhibit decreased language productivity during narrative discourse production relative to normal controls as measured by duration of narratives produced, number of communication units and number of words. 2. Do individuals with PD require increased verbal cuing to produce narrative discourse relative to normal controls? 3. Do individuals with PD exhibit decreased global coherence during narrative discourse production relative to normal controls? 4. Do individuals with PD exhibit decreased language cohesion during narrative discourse production tasks relative to normal controls? 5. Do individuals with PD exhibit a progressive decline in language productivity across time relative to normal controls?
32 6. Do individuals with PD require increased verbal cuing across time during narrative discourse relative to normal controls? 7. Do individuals with PD exhibit a progressive decline in global coherence across time relative to normal controls? 8. Do individuals with PD exhibit a progressive decline in cohesion across time relative to normal controls? Hypotheses Specific hypotheses for comparison of individuals with PD and normal controls based upon the stated research questions are as follows: 1. Individuals with PD will exhibit decreased language productivity relative to normal controls as measured by total duration of narratives produced. A comparison of the same time interval will reveal and a reduction in the following: a) total number of communication units and b) total number of words. 2. Individuals with PD will require increased verbal cuing relative to normal controls to generate equivalent durations of narrative discourse. 3. Individuals with PD will exhibit decreased average global coherence relative to normal controls. A comparison of the same time intervals will reveal a reduction in global coherence ratings. 4. Individuals with PD will exhibit decreased average language cohesion relative to normal controls. A comparison of the same time interval will reveal a reduction in the following: a) total of cohesive markers and b) total percentage adequate cohesive marker. 5. Individuals with PD will exhibit a progressive decline in language productivity relative to normal controls. A comparison of each one-minute interval will demonstrate a progressive decline in: a) total number of communication units and b) number of words per communication units relative to normal controls. 6. Individuals with PD will require a greater number of verbal cues relative to normal controls. A comparison of each one-minute interval will demonstrate a progressive increase in the number of verbal cues required to generate equivalent durations of narrative discourse. 7. Individuals with PD will exhibit a progressive decline in global coherence relative to normal controls. A comparison of each one-minute interval will demonstrate a progressive decline in global coherence ratings relative to normal controls. 8. Individuals with PD will exhibit a progressive decline in cohesion relative to normal controls. A comparison of each one-minute interval will demonstrate a
33 progressive decline in: a) total number of cohesive markers and b) percentage use of complete cohesive markers.
CHAPTER 3 METHOD Subjects Subjects consisted of 12 patients diagnosed with idiopathic Parkinsonâ€™s disease recruited from the North Florida/South Georgia (NF/SG) Veterans Health System Movement Disorders Clinic. Twelve age and education matched neurologically intact subjects recruited from NF/SG Veterans Health System and surrounding areas served as the control group. Each subject gave written consent to participate after full disclosure of the studyâ€™s purpose, risks, and potential benefits. The study was reviewed and monitored by the VA Subcommittee for Clinical Investigation, VA Research and Development Committee and University of Florida Institutional Review Board. Potential subjects were excluded from the study if they had history of prior stroke, dementia, brain tumor, or head trauma as identified in the medical history. Each potential subject had a least a seventh grade education, functional hearing for normal conversation, functional vision for reading tasks and spoke English as their primary language. All subjects demonstrated expressive language skills within intact range for normal conversation. All subjects were male and right handed. A board certified neurologist specializing in movement disorders evaluated all subjects with PD. Each subject presented with a minimum of 3 of 4 cardinal features of PD (resting tremor, rigidity, bradykinesea, postural instability) and had no history of deep brain stimulation. The neurologist rated each subject with the Hoehn & Yahr Staging Scale for PD (Hoehn & Yahr, 1967). Scores on the Hoehn & Yahr Staging Scale range 34
35 from 0 (No signs of disease) to 5 (wheelchair bound or bed ridden unless aided). Each PD subject was also classified by predominate Parkinsonâ€™s disease feature (tremor vs. rigidity) Test Instruments Mini Mental Status Exam (MMSE) Each subject was screened for general cognitive ability with the MMSE (Folstein, Folstein, & McHugh, 1975) prior to participating in the study. Scores of 25 or below were considered an indication of dementia or significant cognitive impairment and resulted in those subjects being excluded from the study. Boston Naming Test (BNT) Each subjectâ€™s general naming ability was assessed with the BNT (Kaplan, Goodglass, & Weintraub, 1983). A non-standard administration was completed as each subject was presented all 60 black and white pictures and instructed to give the name of each. In the event the subject was unable to name the picture he was then give a verbal cue indicating a specific feature of the picture or how the item was used. If the feature cue did facilitate the name of the item, the subject was given a phonemic cue or the sound of the initial portion of the picture name. If the phonemic cue did not facilitate the name of the picture, the subject was given multiple choices to identify the picture name. Scores on the BNT range from 0-60. Wechsler Memory Scale â€“ Logical Memory I (WMS-LMI) Each subjectâ€™s auditory immediate memory was assessed with the WMS-LMI (Wechsler, 1997). Each subject was orally presented 2 short stories with the second story being presented twice for a total of 3 story presentations. Subjects were asked to retell each of the stories from memory. Scoring on the WMS-LMI is based upon recalling 25
36 predetermined ideas defined as critical to each story. One point is given for each story element that is repeated by the individual being tested. Scores on the WMS-LMI range from 0-75 for the 3 stories. Procedure Standardized Assessments PD subjects were identified in the NF/SG Veterans Health System Movement Disorders clinic. Each subject with PD was administered the MMSE as part of their clinic visit. Eleven of the 12 individuals with PD were seen in their homes for participation in the study with the final subject being seen at the Malcolm Randall VA Medical Center following his clinic visit. Subjects were seen in the â€œoffâ€ state of their parkinsonian medications in which patients were seen prior to their first daily dose of anti-parkinsonian medication, with their last dosage at least 12 hours prior to study participation. Four of the 12 subjects were newly diagnosed PD and had no history of parkinsonian medication use at the time of the study. One subject had a 4-day history of medication use. All other subjects had at least a six-month history of medication use. Each subject was administered the BNT and WMS-LMI at the time of the study. Control subjects were also seen either in their homes or at the NF/SG Veterans Health System/Malcolm Randall VA Medical Center at the time of the study. The MMSE, BNT and WMS-LMI were administered at the time of the study. The control group was matched to the PD group on the basis of age, ethnicity, education and gender. Narrative Discourse Data Collection Each subject was instructed to discuss 3 topics, which included the following: (1) a typical day, (2) a memorable vacation and (3) his/her family. Each subject was instructed to discuss each topic for a minimum of 3 minutes and in the event that he stopped before
37 3 minutes he would be cued to continue. The subject would be cued with the following: â€œTell me more about thatâ€. Each subject was reinstructed on the task before eliciting each topic sample. The order of presentation of the 3 topics was counterbalanced across subjects. Specific instructions for the narrative task are included in Appendix A. A Sony VN-480 PC digital voice recorder and electret condenser microphone was used to record each subject enrolled in the study. The microphone was attached to the left side of the subjectâ€™s shirt approximately 3 inches from the mouth. The investigator provided the subject the instructions for each sample followed by a restatement of the topic. Audio-taping began at the point when the topic was restated. A short sample of each recording was reviewed at the time of the study to ensure that the quality of the recording was adequate for further analysis. Audio-taped samples of narratives collected from each of the twelve participants with PD were evaluated by a judge independent of the project for dysarthria. Each sample was rated on a 5-point scale to determine the stage of functional limitation of dysarthria on speech production (Yorkston et al., 1999). Ratings ranged from 1 (No detectable disorder) to 5 (No functional speech). (See Appendix C for summary of stages) Analysis of Narrative Data Transcription and segmentation All recorded samples were downloaded from the Sony VN-480 PC digital voice recorder into a Dell Pentium IV desktop computer with software designed for the digital recorder. The software program created individual audio files that were saved to the computerâ€™s memory. Each file was then re-named with the participantâ€™s study identification number for future analysis. The investigator then transcribed the first 3 minutes all language samples verbatim using Microsoft WORD 2000. A transcription
38 form was developed for the project (See Appendix B). The transcription form was designed for the following: 1) segment productions in 1-minute intervals, 2) score global and local coherence, 3) identify and score cohesive ties and 4) identify verbal cues. Each sample was divided into communication units and each communication unit was transcribed individually. A communication unit was defined as the shortest allowable independent clause and related dependent clauses (Hunt, 1965). A communication unit is the equivalent of a simple sentence. Individual communication units were defined primarily by syntax, however prosodic and semantic features were used at times when the unit could not be determined entirely by syntax (Glosser, 1993). In instances where the location of coordinating conjunctions such as â€œandâ€, â€œbutâ€ and â€œorâ€ was unclear, their prosodic feature determined their final location at the beginning or ending of the communication unit. All words unintelligible to the investigator were excluded from the analysis. During transcription, each 1-minute interval was noted on the transcription form. The audio-file for each narrative sample visually displayed the real time for each sample and this was used to identify the time intervals. In the event that a subject initiated a communication unit prior to a 1-minute interval break but continued past the minute interval point, the entire communication unit was transcribed as part of the prior 1-minute sample. In addition, all verbal cues that were provided were also noted with the word CUE on the transcript at the point in the sample at which the cue was provided. The total time of the narrative production was also recorded on the transcription form. The total time was calculated from the participantâ€™s initiation of the narrative to the ending of the last uncued utterance.
39 Productivity analysis Following segmentation of communication units, the number of communication units included in each sample 3-minute sample was counted. The number of words included in each sample was also counted. Each individual word was counted with contractions counting as 2 words. Dialectical variations such as â€œkindaâ€, â€œwannaâ€ and â€œgonnaâ€ were also counted as 2 words while variations such as â€œemâ€ (them) were counted as 1 word. Filler words such as â€œyou knowâ€ â€œuhmâ€ or â€œahâ€ were excluded from the communication unit word count. Immediate and exact word or phrase repetitions were excluded from the word and word per communication unit count. After all words were identified and counted, the total numbers of words were recorded on the transcription form. Coherence analysis Each communication unit within each 3-minute narrative was assigned a rating for global coherence. Global coherence is defined as the relationship of the meaning of the context of each communication unit to the established topic of the narrative (Gloser and Deser, 1990). In this study narrative topics included: (1) a typical day, (2) a memorable vacation and (3) discussion of family and all communication units should relate to the specified narrative topic in order to maintain global coherence. The rating scale was originally developed by Glosser & Deser (1990) and later modified by Van Leer & Turkstra (1999) (Appendix D). Global coherence was rated on a 5-point scale with the higher numbers indicating a higher degree of coherence. The numerical rating of global coherence for each communication unit was written on the transcription form.
40 Cohesion analysis Each communication unit was evaluated for use of cohesive markers. Cohesive markers within three categories (Reference, Conjunction, and Lexical) as defined by Liles (1985) were identified (Appendix E). Liles procedure was based on a system developed by Halliday & Hasan (1976) to evaluate language cohesion. Although five types of cohesive markers were identified by Halliday & Hasan, subsequent work evaluating discourse cohesion noted that typically only three markers are used frequently enough for statistical analysis (Van Leer & Turkstra, 1999). Each cohesive marker was circled on the specified transcript. Following identification of cohesive markers, each was judged for the adequacy of its cohesive tie. Cohesive ties were specified as complete, incomplete or erroneous as defined by the Liles procedure (1985). Cohesive ties were judged complete when the referent could be easily found in the preceding discourse. Incomplete ties were defined as cohesive markers in which the referent could not be identified in the discourse or was not evident in the context. Erroneous ties were judged as such when multiple referents could be identified in the discourse therefore making the marker ambiguous (See Appendix E for definitions). Each tie was assigned a numerical value for cohesive adequacy as follows: 1-complete tie, 2-incomplete ties and 3-erroneous tie. The numerical value for each cohesive tie was recorded on the transcription form. The number of ties in each sample and the percentage of complete ties were calculated. Scoring reliability Three trained raters participated in the project to establish reliability of identifying and scoring productivity, coherence and cohesion of the narrative. One trained rater analyzed 100% of the samples and served as the primary rater for the project. Two
41 additional trained raters independently analyzed 11 or 72 randomly selected narratives collected equivalent to (15%) of the total sample. The reliability of scoring of the narratives for the primary rater was established for identification of total narrative production time, number of communication units, number of words, number of cohesive ties, percent correct cohesive ties and ratings of global coherence. Intra-class correlation coefficients were completed to analyze inter-rater agreement among the 3 raters. The ratings of the primary rater were used in the statistical analysis of the project.
CHAPTER 4 RESULTS Demographic Comparisons Table 1 lists demographic data and descriptive statistics for participants in the study. Two-tailed t-test with an alpha level of .05 revealed no significant differences between the PD group and control group in age or education. Five subjects with PD completed the study within 1 week of receiving a diagnosis or PD and received a Park year score of 0. The remaining 7 participants park years ranged from 2 â€“ 14 years. Subjects with PD received Hoehn & Yahr scores of 2 with 7 participants receiving a score of 2 and 5 receiving a score of 3. Park years and Hoehn & Yahr ratings for the PD subjects are included in Table 1. Table 4-1. Demographic Data for Parkinsonâ€™s Subjects and Controls PD subjects Controls Variable M SD M SD T p Age 71.8 13.2 72.6 13.5 -0.14 >.05 Education 12.0 1.3 12.8 2.8 -0.94 >.05 Park Years 3.6 4.6 Hoehn & Yahr 2.4 .5 Note. Park Years = # of years since PD subjects initially diagnosed with PD. Park Years ranged 0-14. Hoehn & Yahr scores ranged 2-3. 42
43 Cognitive and Language Comparisons Table 2 lists comparisons of the PD subjects and controls on the 3 measures of cognitive and language ability. Two-tailed t-test with an alphas level of .05 revealed no significant differences between the PD group and control group on the MMSE, BNT, and WMS-LMI. Table 4-2. Cognitive and Language Measures for Parkinsonâ€™s Subjects and Controls PD subjects Controls Variable M SD M SD t p BNT 52.8 6.7 51.8 8.4 .31 >.05 MMSE 28.6 1.4 28.8 1.7 -.26 >.05 WMS-LMI 27.5 11.5 30.6 14.4 -.58 >.05 Note. BNT scores ranged 33-60. MMSE scores ranged 26-30. WMS-LMI ranged 9-60. Dysarthria Ratings Two-tailed t-test with an alphas level of .05 revealed a significant difference between the PD group (M = 2.2, SD .72) and control group (1.3 SD .62) on the dysarthria rating, t (22), p=.003. Scores ranged from 1 â€“ 3 for each group. The control group had one subject with a score of 3, 1 with score of 2 and the remaining subjects received a score of 1. The PD group had 4 subjects receiving a score of 3, 6 receiving a score of 2 and 2 receiving a score of 1. Scoring Reliability Intra-class correlation coefficients (ICC) were calculated by using a two-way mixed model with repeated measures to evaluate scoring agreement among the raters for total narrative production time, number of communication units, number of words, number of words per communication unit (WPCU), number of cohesive ties, percent correct cohesive ties and percent high coherence. Results indicated a high agreement with the
44 primary rater for identifying communication units, words, cohesive ties and moderate agreement on percent correct cohesive ties and coherence. ICC values are included in Table 3. Table 4-3. Intraclass Correlation Coefficients for Scoring Variable ICC Communication Units .99 Words .99 Cohesive Ties .91 Percent Correct Ties .83 Coherence .80 Time .99 Productivity The results of all variables will compare PD subjects to normal controls. An alpha level of .05 (2-tailed) was used for all statistical analyses. Mixed model analysis was used for all statistical tests unless otherwise stated. The first comparison (narrative production time) considered the total duration of each narrative produced. Subsequent analyses reported are comparisons of the first 3 minutes of each narrative sample. Narrative Production Time The total narrative production time of PD subjects with PD was compared to the control subjects across narratives. Results from a 2 (group) X 3 (narrative) comparison indicated a main effect for group [F(1, 42) = 9.16, p=.004] and a group X narrative interaction [F(2, 206) = 3.73, p=.026]. The effect of narrative was not significant [F(2, 206 = 1.33, p=.267]. The controls subjectsâ€™ total production time was greater than the PD subjects across each narrative with the interaction occurring on the discussion of a typical day. Controls subjects produced shorter typical day narratives relative to memorable
45 vacation and family while PD subjects produced longer narratives relative to the other narrative. The mean production time for the 2 groups across the 3 narratives is reported Table 4 and displayed in Figure 1. Table 4-4. Mean Narrative Production Time (Minutes) Narrative Typical Day Memorable Vacation Family PD 3.65 (.74) 3.36 (.49) 3.54 (.35) Controls 4.43 (1.51) 5.12 (3.03) 5.17 (3.09) Narrative Production Time33.544.555.5Typical DayVacationFamilyMinutes PD Controls Figure 4-1. Average Narrative Production Verbal Cuing The total number of cues required to produce the narrative samples were collapsed across the 3 narratives for comparison. Two-tailed t-test with an alphas level of .05 revealed a significant difference between the PD group (M =3.75, SD 3.07) and control group (1.58 SD, 1.78) in the total number of cues required for narrative production, t(22), p=.046.
46 PD subjects required a total of 45 cues across the 72 narrative samples produced. Three cues (7%) were required during minute 1, 13 cues (29%) during minute 2, and 29 (64%) during minute 3. Of these totals, one PD subject required 4 cues to complete one of his 3-minute narratives, 1 PD subjects required 3 cues to complete two of his 3-minute narratives, 1 PD subject required 3 cues to complete one of his 3-minute narratives and 6 PD subjects required 2 cues to produce one of their 3-minute narrative. Each of the twelve PD subjects required at least 1 verbal cue during at least 1 of their 3 narratives produced. Control subjects required a total of 18 cues across the 72 narrative samples produced. Two cues (11%) were required during minute 1, 6 cues (33%) during minute 2, and 10 (56%) during minute 3. Of these totals, one control subject required 4 cues to complete one of his 3-minute narratives and 2 PD subjects required 2 cues to produce one of their 3-minute narrative. Seven of the twelve control subjects required at least 1 verbal cue during at least 1 of their 3 narratives produced. The total number of cues required to elicit the narrative samples are depicted in Figure 2. Verbal Cues05101520253035Min 1Min 2Min 3 PD Controls Figure 4-2. Total Number of Verbal Cues for Each Group During Each 1-minute Interval
47 Communication Units The total number of communication units produced by PD subjects for the initial 3-minutes of each narrative was compared to the control subjects across narratives. Results from a 2 (group) X 3 (narrative) comparison indicated a main effect for narrative [F(2, 192) = 9.56 , p=.000] and a group X narrative interaction [F(2, 192) = 3.90, p=.022]. The effect of group was not significant [F(1, 23) = .99, p=.330]. The PD subjectsâ€™ used a greater, although not statistically significant number of communication units relative to the controls across narratives with the interaction occurring on the discussion of typical day and memorable vacation. The mean number of communication units produced by the 2 groups is reported in Table 5 and displayed in Figure 3. Table 4-5. Mean Number of Communication Units Narrative Typical Day Memorable Vacation Family PD 35.9 (7.6) 35.3 (9.0) 34.8 (7.4) Controls 33.9 (6.0) 34.0 (6.4) 30.5 (7.4) Communication Units2728293031323334353637Typical DayVacationFamily PD Controls Figure 4-3. Average Number of Communication Units Produced for Each Narrative
48 The total number of communication units produced by PD subjects for each 1-minute interval was compared to the control subjects across narratives for a total of 3 minutes. Results of a 2 (group) X 3 (narrative) X 3 (time in 1 minute intervals) comparison indicated main effects approaching significant for group [F(1, 24) = 3.62, p=.069] and narrative [F(2, 168) = 2.7, p=.069. The main effect for minute was non-significant [F(1, 24) = .37, p=.551]. Two interactions were non-significant; group X narrative [F(2, 168) = 1.03, p=.359] and narrative X minute [F(2, 168) = .93, p=.398]. The group X minute interaction approached significance [F(1, 24) = 3.55, p=.072] indicating variability in communication units produced at 1 minute intervals. The PD group increased the number of communication units during minute 2 on typical day narrative while controls decreased. The second interaction occurred during memorable vacation narrative and PD subjects decreased from minute 1 to minute 2 while the normal controls increased during the same interval. Finally, during the family narrative, the PD subjects decreased their number of communication units from minute 2 to minute 3 while the controls maintained a constant rate of production during the same interval. The mean numbers of communication units for each 1-minute interval are displayed in Figure 4. Word Production The total number of words produced by PD subjects for the initial 3-minutes of each narrative was compared to the control subjects across narratives. Results from a 2 (group) X 3 (narrative) comparison indicated a main effect for narrative [F(2, 192) = 78.97, p=.000]. The PD subjectsâ€™ and control subjects produced significantly less words during the family narrative relative to typical day and memorable vacation. The effect of group [F(1, 24) = .00, p=.974] and the group X narrative interaction was non significant
49 [F(2, 194) = 2.04, p =.132]. The mean number of words for each narrative are reported Table 6 and displayed in Figure 5. Communication Units Across TimeFor Each Narrative91011121314Min1Min2Min3Communication Units Typical Day-PD Typical Day-Control Vacation-PD Vacation-Control Family-PD Family-Control Figure 4-4. Average Number of Communication Units 1-minute intervals. Table 4-6. Mean Words Narrative Typical Day Memorable Vacation Family PD 398.8 (67.2) 388.3 (66.8) 354.0 (58.3) Controls 407.3 (114.3) 386.0 (73.0) 344.8 (102.4) Word Production310320330340350360370380390400410420Typical DayVacationFamilyWords PD Controls Figure 4-5. Average Number of Words Produced for Each Narrative
50 Coherence The results of all variables will compare PD subjects to normal controls. An alpha level of .05 (2-tailed) was used for all statistical analyses. Mixed model analysis was used for all statistical tests unless otherwise stated. All analyses reported are comparisons of the first 3 minutes of each narrative sample. Communication units receiving a coherence score of (4-5) were considered highly coherent while those receiving a score of (3) were considered moderately coherent. Those communication units receiving a score of (1-2) were considered to have low coherence. A minimal number of communication units were rated as moderate coherence (score of 3) therefore all communication units rated as moderate coherence were combined with low coherence. The dependent variable used for the coherence analysis was the percentage high coherence. The average percent high coherence produced by PD subjects was compared to the control subjects across narratives. Results from a 2 (group) X 3 (narrative) comparison indicated a non-significant main effect for group [F(1, 23) = 1.58, p=.221] and a main effect for narrative [F(2, 192) = 9.20, p=.000]. The PD and control groups differed in percent high coherence on the typical day and family narratives. A group X narrative interaction was significant [F(2, 192) = 5.08, p=.007]. The PD subjects produced a lower percent high coherence on the typical day narrative relative to the memorable vacation and family narratives. The mean percentage high coherence ties produced by the 2 groups is reported in Table 7 and displayed in Figure 6. The percent high coherence by PD subjects was compared to the control subjects for each 1-minute across narratives for a total of 3 minutes. Results from a 2 (group) X 3 (narrative) X 3 (time in 1 minute intervals) comparison indicated a non-significant main
51 effects for group [F(1, 290) = .505, p= .478], time [F(1, 17) = 2.01, p = .174] and narrative [F(2, 295) = 1.15, p=.316. A narrative X minute interaction was significant [F(2,295) = 4.08, p .018]. Both groups produced significantly higher global coherence when producing the typical day narrative during minute-1 relative to minutes 2 and 3. The final 2 interactions were non-significant which included group X minute [F(1,17) = .06, p = .810] and group X narrative [F(2, 295) = 2.41, p=.091]. The coherence distribution was highly skewed to the high values and these results should be interpreted with caution. The mean percent high coherence for each one-minute interval is displayed in Figure 7. Table 4-7 Percentage High Coherence Narrative Typical Day Memorable Vacation Family PD 79.4 (23.6) 91.3 (12.9) 89.5 (10.2) Controls 91.4 (17.3) 92.6 (15.2) 93.9 (11.4) Percentage High Coherence707580859095Typical DayVacationFamilyPercent PD Controls Figure 4-6. Percent High Coherence for Each Narrative
52 Percent High Coherence Across Time for Each Narrative708090100Minute 1Minute 2Minute3Percent High Coherence Typical Day-PD VacationPD Family-PD Typical Day-Control Vacation-Control Family-Control Figure 4-7. Percent High Coherence 1-minute intervals Cohesion The results of all variables will compare PD subjects to normal controls. An alpha level of .05 (2-tailed) was used for all statistical analyses. Mixed model analysis was used for all statistical tests unless otherwise stated. All analyses reported are comparisons of the first 3 minutes of each narrative sample. Cohesive Ties The total number of cohesive ties produced by PD subjects for the initial 3-minutes of each narrative was compared to the control subjects across narratives. Results from a 2 (group) X 3 (narrative) comparison indicated a non-significant effect for group [F(1, 24) = .010, p =.919]. A main effect for narrative [F(2, 192) = 5.61, p=.004] and a group X narrative interaction [F(2, 194) = 10.48, p =.000] was significant. The main effect for narrative indicated that the PD and control subjectsâ€™ had relative differences in the number of cohesive ties produced on the typical day and memorable vacation narratives in comparison to the family narrative. The group X narrative interaction indicated that the PD group produced a significantly lower number of cohesive ties on the memorable
53 vacation narrative relative to typical day and family narratives while the control group produced a significantly higher number of cohesive ties on the memorable vacation narrative relative to the typical day and family narratives. The mean number of cohesive ties produced by the 2 groups is reported in Table 8 and displayed in Figure 6. The Percentage Distribution of each cohesive tie type (Reference, Lexical, Conjunction) is displayed in Figure 9. Table 4-8. Cohesive Ties Narrative Typical Day Memorable Vacation Family PD 61.6 (20.5) 59.6 (21.0) 65.0 (17.5) Controls 55.4 (16.7) 67.7 (19.1) 61.3 (17.5) Cohesive Ties5052545658606264666870Typical DayVacationFamilyNumber of Ties PD Controls Figure 4-8. Average Number of Cohesive Ties Produced for Each Narrative The total number of cohesive ties produced by PD subjects for each 1-minute interval was compared to the control subjects across narratives for a total of 3 minutes. Results from a 2 (group) X 3 (narrative) X 3 (time in 1 minute intervals) comparison indicated a non-significant main effects for group [F(1, 55) = .04, p=.850] and narrative
54 [F(2, 209) = 1.62, p=.201. A significant main effect for minute [F(1, 310) = 17.69, p=.000] and a group X narrative [F(2, 209) =3.31, p =.038) interaction was observed. Both groups produced a higher number of cohesive ties during minute-2 across narratives. In addition, the control group produced a higher number of cohesive ties relative to the PD group on the memorable vacation narrative. Group X minute [F(1, 310) = .46, p = .499] and narrative X minute interactions [F(2, 209) = .29, p =.748] were non-significant. The mean number of cohesive ties produced for each one-minute interval is displayed in Figure 10. Percentage Distribution of Ties05101520253035404550ReferenceLexicalConjunctionsPercent PD Controls Figure 4-9. Percentage Distribution of Each Cohesive Tie type for Each Group Cohesive Ties Across Time for Each Narrative12141618202224262830Min1Min2Min3Cohesive Ties Typical Day-PD Typical Day-Control Vacation-PD Vacation-Control Family-PD Family-Control Figure 4-10. Average Number of Communication Units 1-minute intervals
55 Total Percent Correct Cohesive Ties The total percent correct use of cohesive ties produced by PD subjects for the initial 3-minutes of each narrative was compared to the control subjects across narratives. Results from a 2 (group) X 3 (narrative) comparison indicated a significant main effect for group [F(1, 81) = 5.891, p=.017] and a main effect for narrative approaching significance [F(2, 192) = 2.90, p=.057]. The control subjects produced a higher percentage complete cohesive ties relative to the PD subjects across narratives. A group X narrative interaction also approached significance [F(2, 192) = 2.670, p=.072] indicating that the PD group produced a significantly lower percent correct cohesive ties on the memorable vacation narrative relative to typical day and family. The mean percentage correct use of cohesive ties produced by the 2 groups is reported in Table 9 and displayed in Figure 11. The percent correct use of cohesive ties produced by PD subjects for each 1-minute interval was compared to the control subjects across narratives for a total of 3-minutes. Results from a 2 (group) X 3 (narrative) X 3 (time in 1 minute intervals) comparison indicated a non-significant main effects for group [F(1, 24) = 1.40, p= 1.000], time [F(1, 24) = .014, p = .908] and narrative [F(2, 168) = .34, p=.714]. All interactions were non-significant which included group X minute [F(1,24) = .00, p = .969], group X narrative [F(2, 168) = .67, p=.513] and narrative X minute [F(2, 168) = .78, p=.458]. The mean percent correct cohesive ties produced for each one-minute interval is displayed in Figure 12.
56 Table 4-9. Percentage Correct Cohesive Ties Narrative Typical Day Memorable Vacation Family PD 94.7 (3.9) 92.3 (6.4) 94.8 (5.7) Controls 97.7 (3.5) 96.9 (2.5) 96.4 (8.2) Percent Correct Cohesive Ties8990919293949596979899Typical DayVacationFamilyPercent PD Controls Figure 4-11. Average Percent Correct Cohesive Ties for Each Narrative Percent Correct Cohesive Ties Across Time for Each Narrative9092949698100Min1Min2Min3Percent Typical Day-PD Typical Day-Control Vacation-PD Vacation-Control Family-PD Family-Control Figure 4-12. Average Percent Correct Cohesive Ties 1-minute intervals
57 Further inspection of the cohesive ties judged as incomplete or erroneous indicated that the PD subjects produced 136 ties judges as incomplete or in error compared to 63 by the controls subjects. Sixty-nine percent of all ties judged incomplete or erroneous were of reference type for both groups. The total number of ties judges as incomplete or erroneous is depicted in Figure 13. Incomplete and Erroneous Ties051015202530354045ReferenceLexicalConjunctionsReferenceLexicalConjunctionReferenceLexicalConjunctions Typical DayVacationFamily PD Controls Figure 4-13. Distribution of Incomplete and Erroneous Cohesive Ties Parkinsonâ€™s Disease Within Group Comparisons PD Demographic Data Table 10 lists demographic data and descriptive statistics for PD subjects categorized by Hoehn & Yahr scores. Two tailed Mann-Whitney test with an alpha level of .05 revealed no significant differences between PD subjects at Hoehn & Yahr Stage 2 vs Stage 3 in mean age and level of educations. Park Years did approach significance.
58 Table 4-10. Comparison of Demographic Data for PD Subjects Hoehn & Yahr 2 Hoehn & Yahr 3 Variable M SD M SD z p Age 69.6 16.1 74.6 8.5 -.16 .87 Education 12.1 1.7 11.8 .4 -.87 .38 Park Years 1.4 2.6 6.8 5.3 -1.86 .06 Note. Park Years = # of years since PD subjects initially diagnosed with PD. Park Years ranged 0-14. PD Cognitive and Language Comparisons Table 11 lists comparisons of the PD subjects on the 3 measures of cognitive and language ability. Two tailed Mann-Whitney test with an alpha level of .05 revealed no significant differences between PD subjects at Hoehn & Yahr Stage 2 vs Stage 3 on the MMSE, BNT, and WMS-LMI. Table 4-11. Cognitive and Language Measures for Parkinsonâ€™s Subjects and Controls Hoehn & Yahr 2 Hoehn & Yahr 3 Variable M SD M SD z p BNT 52.3 9.1 53.4 2.1 -.66 .51 MMSE 28.1 1.8 29.2 .45 -.79 .43 WMS-LMI 25 10.3 31 13.4 -1.22 .22 Note. BNT scores ranged 33-60. MMSE scores ranged 26-30. WMS-LMI ranged 9-43. PD Dysarthria Ratings Two tailed Mann-Whitney test with an alpha level of .05 revealed no significant differences between PD subjects at Hoehn & Yahr Stage 2 ( M =2.0) vs Stage 3 ( M 2.1) on dysarthria ratings [z = -.18, 0=.859]. Scores ranged from 1 â€“ 3 for each group. Two Hoehn & Yahr Stage 2 subjects had a rating of 3, 4 had a rating of 2 and 1 subject had a
59 rating of 1. Two Hoehn & Yahr Stage 3 subjects had a rating of 3, 2 had a rating of 2 and 1 had a rating of 1. Narrative Comparisons Two tailed Mann-Whitney test with an alpha level of .05 revealed no significant differences between PD subjects at Hoehn & Yahr Stage 2 vs Stage 3 on average total narrative time [z = -1.11, p = .27], total words [z =-1.99, p = .83], percent high coherence [z = -1.09, p = .28], and percent correct use cohesive ties [z = -.22, p = .83]. Significant differences were noted on average total number of communication units [z = -1.99, p = .046] and average number of cohesive ties [z = -1.99, p = .046]. These results should be interpreted with caution due to the limited number of subjects in each group. The results of narrative comparisons between Hoehn & Yahr 2 and 3 subjects are included in Table 12. Table 4-12. Comparison of Narrative Measures for PD Subjects Hoehn & Yahr 2 Hoehn & Yahr 3 Variable M SD M SD z p Time 3.5 .2 3.6 .2 -1.11 .268 Communication Units 33.3 1.5 36.3 .6 -1.99 .046 Words 375.3 15.9 379.3 28.1 -.218 .83 Cohesive Ties 57.3 .6 65.0 5.6 -1.99 .046 Percent Correct Ties 94.3 2.6 94.3 1.5 -.218 .827 Percent High Coherence 83.7 5.9 90.7 6.1 -1.09 .268
CHAPTER 5 DISCUSSION To explore the role of the basal ganglia in expressive language performance, this study evaluated language productivity, global coherence and cohesion of narrative discourse in individuals with idiopathic Parkinsonâ€™s disease. A comparison of the narrative discourse samples of individuals with PD to age matched controls was completed to evaluate the relative contribution of the basal ganglia to narrative discourse production. Basic intra-group comparisons of individuals at 2 different stages of PD were also completed as a consideration of the relative contribution of the disease state. I will discuss each of the 3 primary domains (productivity, coherence, cohesion) evaluated and conclude with an integration of each to draw final conclusions. Productivity The hypothesis that individuals with PD would exhibit decreased language productivity relative to normal controls was supported in part by the results of this study. In regards to overall average time of narrative production, individuals with PD produced narratives of shorter duration relative to the control subjects. However, a comparison of the first 3-minutes of the each narrative, demonstrated slightly decreased but non-significant differences in communication units and words produced. It is unclear if the comparative samples would have yielded different results if a longer duration of sample was elicited and compared. Despite the fact that many of the participants were very early in the disease state, their ability to produce narratives of 3-minutes or greater duration in the absence of multiple verbal cues seemed very unlikely. 60
61 Total Narrative Comparisons Each of the PD subjects required at least 1 verbal cue during the study to produce one of the 3-minute narratives. A number of factors must be considered when evaluating these mixed results (total time vs 3-minutes) and determining their significance. First, studies of individuals with PD have generally concluded that a reduction of motor speech performance is the primary factor influencing expressive language performance (Illes et al., 1988; Murray & Lenz, 2001; Murray, 2000). Therefore, it may be expected that a reduction in motor speech production would be the cause of the total time differences. However, the participants of this study were generally early in the disease process and presented with mild overall ratings of motor speech performance. Scores ranged from 1-3 on the Stages of Functional Limitations in Dysarthria Scale (Yorkston & Beukelman, 1999) with 8 of the 12 subjects receiving a score of either 1 or 2. A trained speech-language pathologist independent of the project evaluated the narrative samples for motor speech performance. The evaluator was aware that the samples were obtained from individuals with PD and therefore the subtle changes of speech that are not typically observed by the untrained listener were identified. Therefore a gauge of the maximum probable influence of motor speech on speech production on this task was obtained. Based upon the identified scores, the PD subjects did not exhibit significant motor speech influences compared to the control subjects, as the most obvious speech production observation was low volume and monotone voice. An alternate explanation of these results would suggest a greater role of the basal ganglia in expressive language particularly in regards to signal enhancement, intention, action planning and temporal encoding. Nieoullon and Coquerel (2003) proposed that the basal ganglia might serve as a â€œbrain conductorâ€ and parallel frontal lobe functioning
62 in executive functioning. If this hypothesis is correct, a disruption of fronto-basal ganglia circuitry as observed in individuals with PD may manifest in expressive language performance in terms of language productivity. More specifically, a reduction in dopamine levels in the basal ganglia, which are known to be critical to basal ganglia performance, could result in decreased signal enhancement of internal basal ganglia circuitry and subsequently negatively influence the sequence of events critical to language production. First, it would be expected that decreased dopamine levels as a result of Parkinsonâ€™s disease, would alter the normal performance of the intentional system. Since all narratives were collected in the â€œoffâ€ medication state, it could be hypothesized that this reduction could influence intention for expressive language. The first and most obvious observation of a potential disruption of this system, decreased initiation time, was not accounted for in the study design and as a result was not quantitatively analyzed. After the presentation of the instructions, the normal controls used the completion of the instructions as a signal to begin the narrative task or they asked if they should be begin at that time. In contrast, the PD subjects frequently acknowledged an understanding of the instructions but did not begin without additional instruction to initiate the narratives. In regards to motor function, Heilman et al. (2003) noted that individuals with intentional disorders might fail to respond to a stimulus even when they are aware of it. Although, the observation of response delay was not quantified in this study and was not reported as a verbal cue, it may be suggestive of a reduction in the intentional systemâ€™s ability to initiate and generate language. Further qualitative analysis of these observations is needed to evaluate this issue.
63 Secondly, The PD subjects required double the number of verbal cues to produce the minimum 3-minute narrative samples when compared to the controls. While the control subjects also needed verbal cues, the number and juncture in the narrative generally distinguished the groups. Controls subjects were typically cued in the final minute of the narratives, and oftentimes this correlated with their asking if they had reached the 3-minute minimum. Since the investigator responded by providing the elapsed time, this was considered a verbal cue as defined in the instructions. In contrast, the PD subjects frequently stopped abruptly in their narratives, oftentimes at a critical juncture of the narrative. For example, they might begin a narrative regarding a memorable vacation and after they reached the juncture in the narrative at which time they had reached their destination, they might end the narrative with â€œthatâ€™s about itâ€. It was obvious to the investigator that additional information was available as well as that they had not reached the 3-minute minimum. Therefore, a verbal cue was provided. In some cases, the PD subjects were able to continue the narratives to the 3-minute minimum time point. However, at times, multiple verbal cues were required to reach the 3-minute minimum. While cues were needed by each group, the cuing pattern in regards to type and location may also suggest a disruption in the overall language â€œaction planâ€ to reach the goal of expressing a sequence of events or actions necessary to construct a narrative. Alexander (2002) defined â€œaction planâ€ as the ability to generate and monitor simultaneously a complex task that will unfold in the future. Therefore, while it was obvious to the listener (investigator) that the narrative plan was not complete, it was not as obvious to the PD subjects. Nieoullon (2002) noted that individuals with PD may
64 require external feedback to establish a strategy of action for tasks requiring sequential behavior. This need for external feedback appears to correlate with the increased need of verbal cues to complete the narrative tasks in this study. He also noted that when explicit information is available they demonstrate no deficits in completing even complex tasks. Additionally, it has been noted that individuals with PD exhibit difficulties with time estimation (Harrington, Haaland, & Hermanowicz, 1998; Nieoullon, 2002). Nieoullon (2002) noted that individuals with PD frequently underestimate time intervals. While these observations were in relationship to motor functioning, they correlate with the observations of this study in terms of total time of discourse production. The PD subjects may have exhibited difficulty estimating a 3-minute time interval, resulting in discontinued narratives due to perceived completion of the minimum time period. Further, Beiser and Houck (1998) suggested that one role of the basal ganglia was in temporal encoding of sensory or internal events. Therefore, a reduced functioning basal ganglia may influence expressive language performance by decreasing its ability to sequence and maintain adequate flow information. This would result in disruptions with the basal gangliaâ€™s interactions with the frontal lobes and other cortical areas zones responsible for language. The influence of this disruption on expressive language may result in incomplete narratives as observed in this study. Further study of these hypotheses would be required to distinguish the influences of defect time estimation strategies versus disrupted temporal encoding in relationship to the findings of this study. Finally, Heilman et al., (2003) proposed that the basal ganglia may have a role in intentionally guided attention. This concept was hypothesized as the intentional systems ability to initiate and maintain attention to a task. A disruption in the intention system as
65 previously suggested may appear during intentional tasks requiring sustained attention such as narrative discourse. Therefore, the reductions in narrative production time may be related to reduced intentionally guided maintenance of cognitve resources to the task (narrative discourse) as well as the action plan. This would contrast deficits in temporal estimation (Nieoullon, 2002) or reduced temporal encoding as proposed by Beiser & Houck (1998). Findings from Brown and Marsden (1988) also support this notion of reduced intentionally guided attention. They proposed that individuals with PD have impairments in the supervisory attentional system. As a result, they exhibit difficulty with tasks requiring internal attentional control processes. They also note that when task demands are within their available resources, they may not show deficits. However, when their available attentional resources are exceeded, as in cases that require internal cues, deficits are noted. Three-Minute Comparisons A comparison of the PD subjects and controls subjects on equivalent 3-minute intervals yielded no statistically significant differences in communication units and words produced. Further, a comparison of each 1-minute interval for a total of 3 minutes also resulted in non-significant results. However, it should be noted that the comparison of communication units across each 1-minute interval approached significance and further investigation is warranted. It has been proposed that signal enhancement and action selection are critical features of the basal ganglia that influence word retrieval (Baldo & Shimamura, 1998; Crosson et al., 2003; Owen et al., 1995). Therefore, it was expected that a reduction in communication units and total words would occur when comparing equivalent time intervals. Owen et al. (1995) suggested that a critical relationship existed between the frontal lobes and basal ganglia for word retrieval. As a result, disruptions in
66 fronto-basal ganglia circuitry secondary to PD could potentially influence the word retrieval process. The lack of differences in communication units and words produced were surprising and could suggest that fronto-basal ganglia disruptions that influence linguistic processing may occur later in the disease process and minimally affect word retrieval at Hoehn & Yahr Stages 2 & 3. However, the interaction results were of interest and should be discussed in regards to story structure. Although a comparison of total communication units and words during the 3-minute yielded non-significant results, each group demonstrated a reduction in communication units and words for the narrative relating to their family. Since no specific directions were provided as to the manner in which the narrative should be constructed, it provided an opportunity for each individual to discuss their topics in various formats. Typical day narratives were generally told in either a temporal order beginning in the morning and proceeding throughout the day. Memorable vacation narratives were told in a temporal sequence of the specifics of the vacation or in a random order to highlight the specifics of the vacation. Finally, narratives regarding family were told in multiple formats including: chronological order, sibling focused, spouse and children focused or child/grandchildren focused with emphasis on their accomplishments. Additionally, the PD subjects had a greater reduction in the number of communication units for the family narrative relative to typical day and memorable vacation and in comparison to the normal controls. It may be that the multiple potential formats for narrative generation also created increased possibilities for disruptions of language action plans. This suggestion emerges from the observation that control subjects typically stated their format early in the narrative providing the investigator with
67 an indication of the focus of their narrative as opposed to the PD subjects who frequently attempted multiple formats within narratives to achieve the 3-minute goal. These findings would also suggest subtle deficits in initial organization and narrative plan maintenance. Instead of selecting a format early in the narrative and sustaining it until completion, the PD subjects chose a more haphazard plan to achieve the 3-minute requirement. Therefore, they did not consistently verbalize narratives with logical flow. This would be a viable explanation if the PD subjects consistently told their typical day narratives and memorable vacation narratives based upon a temporal order of events or in chronological order of occurrence. A review of the 1-minute intervals seems to support this claim. Individuals with PD demonstrated a general decline in their narratives describing a typical day. However, during their family narratives they had an increase in communication units from minute 1 to minute 2 while declining to minute 3. The increase then decrease suggests faulty action planning or lack of overall plan for goal completion since other narratives were told with similar number of communication units at each 1-minute interval. The most frequent pattern for typical day was to steadily decline, which may correlate with the increased cuing observed in minute 3. It was also surprising that a discussion of family would result in a decline of communication units in both groups as it was thought that this narrative would elicit the greatest duration and productivity. Coherence The hypothesis that individuals with PD would exhibit decreased global coherence relative to normal controls was not supported statistically by the results of this study. Coherence provided the greatest challenge for statistical analysis as the distribution of coherence ratings were highly skewed. Due to the lack of rating of (3) or moderate
68 coherence, the sample consisted of high and low coherence ratings. Although the results were not statistically significant, their clinical relevance is of importance. Overall, the control group was rated higher in terms of global coherence relative to the PD subjects. A comparison of overall coherence indicated that the PD subjects were rated as highly coherent 80-91% of the time while the control subjects were rated highly coherent 91-94% across the three narratives. Global coherence is the relationship between the content and the general topic. These differences between the groups are minor but they have clinical importance. PD subjects were highly coherent approximately 80% of the time during their typical day narratives. As a result, they were off topic 20% of the time therefore reducing the meaning and clarity of their narratives. Narratives describing their typical day seemed to provide greater opportunity for off topic remarks concerning relevant individuals in their lives, which were of importance, but not directly related to their typical dayâ€™s experience. Both groups demonstrated variability in overall coherence and coherence across specified time intervals. However, the higher average percent high coherence among the controls suggests that as they went off topic, they were able to return to the original topic independently. These findings suggest differences in language use during narrative construction. As a result, they exhibited higher overall average scores relative to the PD subjects. These results also suggest that a language action plan for narrative completion is also required for coherence and that the basal ganglia might have some influence on this skill. It may be more suitable to suggest that its role is more of an executive nature in terms of language use or self-monitoring the task of narrative generation. Therefore, the control subjects are more likely to have retained this skills resulting in a greater ability to
69 self-monitor the specified goal and return independently to the language plan of which they had diverted. Ulatowska and Chapman (1994) suggested that discourse formulation is an intellectual task that requires the understanding and manipulation of linguistic information and involves cognitive processes essential for information organization. In addition, they note that successful discourse production relies on correct structure of information. They further suggest that a macrostructure exists, that are cognitive processes serving primarily as organizational devices of meaning. This might suggest that the basal ganglia has a role in maintaining this overall structure, as it has a role in executive functioning for general topic maintenance and self monitoring. The results of this study do not provide statistically significant evidence of disruption of global coherence, however the high level of verbal cuing needed for narrative construction may have influenced those results. PD subjects required greater verbal cues and while the cues served to complete the narratives, they may have provided cuing to return to the original topic. The cues were designed to have minimal influence on the narratives. However, in retrospect, they may have provided the external influence to maintain attention to the original topic. A qualitative analysis of investigator cuing patterns may provide additional evidence of a relationship between verbal cuing and global coherence. Beiser and Houckâ€™s (1998) hypothesis also seems to have explanatory value here as temporal encoding may be expected to play a role in global coherence. It may be hypothesized that a disruption in temporal encoding would result in disrupted sequencing or language action planning. In the event of temporal coding disruption, it would be expected that a higher incidence of reduced global occurrence might emerge as language
70 action plans are disrupted. As a result, the PD subjects may be more likely to go off topic after the disruption and cue to restart. It should also be noted that the enrolled PD subjects were early in the disease process and may have presented with greater cognitive resources to maintain high coherence. Further research is needed with individuals at all disease states to carefully evaluate the relationship between level of a cognitive functioning and global coherence. A final factor that must be given consideration was the lower intra-class correlation coefficients on global coherence. Since global coherence was calculated as the percentage of communication units rated as highly coherent, the number of communication units present in each sample greatly impacted these results. There was greater variability among the raters with subjects with fewer communications units. The global coherence scores of subjects, who produced a higher number of communication units, were less influenced by lower coherence ratings than those with fewer communication units. These differences must be considered in future studies and warrants consideration for a larger sample size and possibly a longer narrative sample to obtain a greater number of communication units. In addition, the highly skewed values towards high coherence create difficulty in suggesting a more definite role for the basal ganglia in the maintenance of global coherence. Further, since both groups demonstrated variability in global coherence, strongly suggesting a hypothesized role for the basal ganglia with the current results would be in error. Therefore, the basal gangliaâ€™s role in global coherence should be investigated further with larger sample size and multiple narrative samples.
71 Cohesion The hypothesis that individuals with PD would exhibit decreased language cohesion relative to normal controls was partially supported by the results of this study. As noted with global coherence, some comparisons of cohesion between individuals with PD and controls were not statistically significant however they were clinically important. Cohesive Ties Overall, the PD group produced a greater number of cohesive ties during the typical day and family narratives, however there were not statistically significant group differences on total number of ties. A comparison of the distribution of the 3 types of ties revealed that individuals with PD and the controls used relative, lexical and conjunctions ties in an equivalent distribution. Further, a comparison of individual 1-minute intervals indicated that the PD subjects were generally consistent in the number of ties produced for each interval in contrast to the controls who had an increase from minute 1 to minute 2 followed by a reduction during the third minute during the family and typical day narratives. This may suggest that discussion of family and vacations would include primarily family members whose names would be specified in the first minute followed by a increase of reference or personal pronoun use during minute 2 as the speaker referred back to the individuals previously named. Percent Correct Cohesive Ties A comparison of percent correct use of cohesive ties revealed that ties produced by individuals with PD were less likely to be judged as complete compared to normal controls. The overall percent correct use of cohesive ties for the PD subjects ranged from 92.3% to 94.7% while the percent correct for the control subjects was 96.4% to 97.7%. While the results may seem minimal, they agree with prior studies of cohesion that have
72 evaluated percent correct use of cohesive ties and have reported rates of 94%-98% for normal controls (Coelho, 2002; Liles et al., 1989). It is unclear if 94% represents a threshold of adequate cohesiveness. It should be noted that average global cohesive values were similar between the two groups, however the standard deviations revealed some further differences. When considering average scores 1 standard deviation below the mean, the average global coherence scores for all narratives produced by the PD subjects were 90% or below. In contrast, only one of the controls subjects presented with only narrative below 94% average global coherence (88% for family). Surprisingly, while there were group differences in total percent correct ties, the PD subjects did not demonstrate a reduction in percent correct cohesive ties over time as hypothesized. A review of all ties judged as incomplete or in error revealed that the PD subjects produced 136 incomplete or error ties compared to 63 for the control subjects. Sixty nine-percent of all ties judged as incomplete or in error were of the reference type for both groups. This is of importance since reference type cohesive ties direct the listener to the identity of the thing or thing to which the reference refers (Halliday & Hasan, 1976). The use of personal pronouns was particularly evident in the family and memorable vacation narratives as the subjects described events which family members. Consequently, a high number of personal pronouns such as â€œheâ€, sheâ€, himâ€, and â€œherâ€ were present during the narratives. In instances of high personal pronoun use, the clarity of the narrative was decided by the listenerâ€™s ability to determine to whom the pronouns were referring. Ulatowska, Allard, & Chapman (1990) noted that reference is significantly important to discourse as it connects lower and higher levels of language and that a
73 disruption in reference words such as pronouns may result in impaired discourse. Further, they note that reference is particularly susceptible to disruption due to the complexity of the reference system. It is unclear why lexicals and conjunctions were less susceptible to reduced cohesive adequacy in the narratives. It might be argued that general nouns and conjunctions create less opportunity for ambiguity when used compared to pronouns resulting in increased accuracy of use. It is also unknown why some normal controls produced incomplete and/or erroneous cohesive ties or whether these ties represent a distinct narrative speaking style. These results may offer additional insights into studies of word fluency (Auriacombe et al., 1993; Henry & Crawford, 2004) and word retrieval (Copland et al., 2000a; Crosson et al., 2003) that have suggested a specific role for the basal ganglia. It may be possible that the basal ganglia modulates word retrieval by serving as an executor for selecting words based upon grammatical use. This hypothesis is based upon evidences from studies of verb retrieval that suggest an executive type role for the fronto-basal ganglia system. First, Grossman (1998) proposed that verb retrieval was governed by an executive system that coordinated and manipulated a wide range of verb choices. Additionally, Shimamura (1994) noted that deficits following frontal lobe injury may result from poor inhibiting and filtering extraneous information during the word retrieval process. It may be suggested that the reference system is organized in the same manner and therefore requires an executive role for correct reference selections. As a result, diseases of the basal ganglia may disrupt its role in word retrieval due to the inability to filter extraneous information, or in this case alternate word selections. However, the application of this
74 principle in regards to the role of the basal ganglia and word retrieval will require additional study of reference use in individuals with PD. Studies by Ullman et al. (2005, 1997) and Ullman, (2004, 2001a, 2001b) also propose a distinct role of the basal ganglia in word selection based upon grammatical structure. Ullman et al. (2001b) proposed a declarative/procedural model to distinguish between structures that create the relationship between form and meaning. He suggested that two distinct but integrated memory systems are present that give language expressive power (Ullman et al., 1997). First, the declarative memory system consists of the storage of facts and events or a memorized word specific knowledge that relies primarily on the temporal lobes (Ullman, 2004). In contrast, the procedural memory system provides rules or mental grammar for the combination of lexical items. The procedural system is thought to rely on frontal, basal ganglia, parietal and cerebellar structures (Ullman, 2004). This â€œdual systemâ€ distinguishes memory stores for words from those for syntax. However, the author notes that the model does not assume that all parts of the two systems subserve language (Ullman, 2001a). Further, the model suggests that other cognitive and computational systems may also be involved in lexicon and grammar. This model is in contrast to the â€œsingle systemâ€ or connectionist models that propose that language depends on a single computational system with broad neuroanatomical representation. This distributed system is thought to facilitate learning via adjustments in the weights of neural connections (Rumelhardt, McClelland, & PDP Research Group, 1986). The importance of this model in relationship to the findings of this study, arise from the hypothesis that the procedural system is rooted in frontal-basal ganglia
75 structures and is specialized for sequences particularly grammatical rules for syntax and morphology (Ullman, 2001b). Studies of the declarative/procedural model have primarily addressed regular and irregular verb form use. Ullman (2001a) noted that this model may be important for grammatical-structure building for the sequential and hierarchical combination of stored verb forms. Further, studies of individuals with PD indicated that they generally exhibited greater difficulties with regular verb forms, which are thought to be rule governed by the procedural system. This is in opposition to disorders such as Alzheimerâ€™s disease in which they have greater difficulty with irregular verbs forms that are thought to be stored in the declarative memory system. There are two key issues in Ullmanâ€™s hypotheses that may have some relevance to disruptions in the referencing system. First, Ullman (2001a) suggests that the procedural system acts to retrieve verbs from a hierachical system of verbs based upon specific rules. It may be argued that the referencing system operates in the same manner in that references are selected based upon exophoric (situational reference) versus endophoric (contextual reference) reference (Halliday & Hasan, 1976). Halliday & Hasan (1976) further noted that endophoric references may be either anaphoric (to preceeding the text) or cataphoric (to following text) and that references can be distinguished for their different uses. They also note that exophoric references contribute to the creation of the text and only endophoric references are cohesive. Exophoric references do not name specific things, they only signal that references must be made in the context. For example, if a subject began a narrative or portion of a narrative with the pronoun â€œtheyâ€ without previous specification of who he was referring, the meaning of the reference would be unclear. Therefore, incomplete and error ties result from exophoric
76 referencing. Incomplete and erroneous ties also point to the complexity of the reference system in terms of the multiple choices available and a hierarchical structure of reference types. The hierarchical structure of references is created by the three types of references (personal, demonstrative and comparative) which are distinguishable by semantic category, grammatical function and class (Halliday & Hasan, 1976). Secondly, there appears to be a rule-governed system for reference or pronoun selection, which distinguishes contextual, and situations references. It is clear that endophoric referencing is required for language cohesion and rules for pronoun selection govern the successfulness of cohesion. Therefore, it could be hypothesized that other word categories in addition to verbs may be rule governed and have the potential to be disrupted in the declarative/procedural model. This would suggest that a disruption of the procedural portion of the model, which relies on frontal-basal ganglia circuitry, might manifest in other word forms such as pronouns, following PD. Since verbs were not evaluated in this study it is unclear if the original model as described by Ullman et al. (1997) would yield such predictions for verbs as well as pronouns. To support this argument, a review of word fluency and retrieval studies of individuals with basal ganglia disorders note that they have generally focused upon retrieving words from semantically related categories. These words are generally used in the same manner in higher levels language structure such as sentences and discourse. If this notion is correct, the higher rate of disruption of percent correct ties, particularly reference, would correlate with these findings and the basal gangliaâ€™s hypothesized role in word selection for grammatical use. For example, errors in reference selection such as â€œheâ€ versus â€œtheyâ€ to direct a listener to a previously specified person, may represent a
77 word use defect as opposed to a selection error due related to semantics. Therefore at a base level, the basal ganglia would serve as mechanism for word selection based on use. These results may serve as some indication of the â€œaction selectionâ€ capabilities for word retrieval by the basal ganglia. In the event that the basal gangliaâ€™s capabilities in action selection are reduced, word use errors could potentially occur. Further, since the PD subjects in this study were primarily early in their disease process, it is unclear if fronto-basal ganglia circuitry was disrupted to a significant level to produce the hypothesized error patterns. Also, similar errors were noted in the control group narratives, although at a lower rate. This finding is not clearly accounted for in this hypothesized model and further study of these issues is warranted before a more definitive conclusion can be drawn. Finally, as noted with global coherence, lower intra-class correlation coefficients were noted on percent complete ties agreement. Since percent complete ties global coherence was calculated as the percentage of cohesive ties that were rated as complete, the number of communication units present in each sample greatly impacted these results. There was greater variability among the raters with subjects with fewer communications units. The percent complete ties for subjects who produced a higher number of communication units, were less influenced by incomplete and erroneous ties than those with fewer communication units. These differences must be also considered in future studies and warrants consideration for a larger sample size and possibly a longer narrative sample to obtain a greater number of communication units. Intra-Group Comparisons Disease state as measured by the Hoehn & Yahr scale was not considered in the primary analysis. However, intra-group comparisons indicated no significant differences
78 in (1) motor dysarthria ratings, (2) total time, (3) total words, (4) percent high coherence, and (5) percent correct use of cohesive ties among the PD subjects. Significant intra-group differences were noted on total number of communication units and total number of cohesive ties. These results were surprising as the Hoehn and Yahr level 3 subjects had a slightly greater number. However, these findings were suggestive of a homogenous group of PD subjects and therefore the primary emphasis of this study was PD and control group comparisons. General Discussion Prior studies of language function in with PD have suggested that language deficits are primarily due to a disruption of underlying motor speech and neuropsychological changes. However, the results of this study suggest alternative causes. The general hypotheses of this study proposed that individuals with basal ganglia disorders such as PD, exhibit deficits in language form and use as measured by less productivity, lower global coherence and decreased cohesiveness when producing narrative discourse. While the findings of the study were mixed and inconclusive, they offer suggestions of a potential role of the basal ganglia in language and at a minimum hypotheses that can be tested in future research. The findings of this study offer information regarding expressive language functioning in individuals with PD that is clinically significant even in the absence of statistical significance. Additionally, these findings suggest that the basal gangliaâ€™s role in signal enhancement and action selection for word retrieval in conjunction with language action planning, intention, temporal encoding and intentionally guided attention, may have a greater influence on expressive language performance than previously reported. Further, since the neurotransmitter dopamine potentially serves as a signal enhancer to facilitate the aforementioned processes, the
79 study of individuals with PD who have a deficient of dopamine, can provide additional explanations for the linguistic deficits observed in PD. Many previous studies had concluded that language deficits in individuals with PD result from motor speech and other cognitive processes that support language. However, the results of this study suggest that future studies of the basal ganglia and its role in expressive language must consider a multitude of cognitive-linguistic factors as well as a number of promising models of basal ganglia function and its contribution to language. As in most studies, there were a number of weaknesses that must be considered when evaluating the merits of the study results. First, a small sample size including participants early in the PD disease progression was used. However, despite this small sample size, significant results were obtained on many of the variables. A larger sample size including individuals with equivalent tremor and rigid predominant features as well as the full range of Hoehn & Yahr disease states would provide additional information about language form and use issues. However, the significant amount of time and resources needed for the language analyses must be considered with larger sample sizes. Second, while the study proposed to evaluate macrolinguistic and microlinguistic aspects of language form, the measures selected were not able to adequately evaluate language form as originally intended. An additional analysis of syntactic complexity and a type token ratio of word usage could provide additional information about language form. Third, alternate explanations such as depression and apathy were not measured in the participants and each could potentially affect the narrative production. Finally, the difficulties in inter-rater agreement on coherence and cohesive adequacy must be
80 addressed in future studies and longer narrative samples may partially improve these difficulties.
CHAPTER 6 TREATMENT CONSIDERATIONS Traditional approaches to treatment of individuals with Parkinsonâ€™s disease (PD) have generally focused upon motor performance. The findings of this study suggest that in addition to motor speech changes, language disruptions may occur early in individuals with PD and should be addressed. Speech-language pathologists have frequently relied on treatment philosophies paralleling physical therapy without regard for a critical consideration of the influences of the subcortical mechanisms that govern motor speech control. The findings of this study may suggest an alternative approach to treatment designs which emphasize a greater role for the basal ganglia and a greater level of language disruption in individuals with PD. Crosson et al. (2005) proposed a novel language treatment based upon a hypothesized model consisting of an integration of theories of intention and expressive language. In contrast to this model of treatment development, many currently practiced and emerging treatments are not based upon such models and are developed primarily as a result of anecdotal evidences. In this chapter I will propose some treatment paradigms based upon the findings of this study. While there are no treatments specific to linguistic deficits in PD, the proposed treatments here will have implications for language as well as traditional motor speech disorders observed in individuals with PD. General Considerations The most notable communication disorder following PD is hypokinetic dysarthria. Hypokinetic dysarthria is characterized by deficits primarily in voice, articulation and 81
82 prosody but may include deficits in respiratory, phonotory, resonatory and atriculatory levels of speech (Duffy, 1995). Specific speech deficits may include monopitch, reduced stress, monoloudness, imprecise consonants, inappropriate silences, short rushes, harsh voice quality, breathy voice (continuous), pitch level and variable rate. This collection of symptoms is typically quantified as a dysarthria type, however some of these motor speech characteristics can be observed in other disorders with frontal lobe involvement. In regards to traumatic brain injury (TBI), moderate to severe TBI survivors frequently present with some of the characteristics of hypokinetic dysarthria although the deficits are traditionally characterized as executive functioning deficits. They are characterized as such because they appear to result from a reduction in the ability of the individual to self-monitor their verbal output. Since these deficits typically resolve in parallel with recovery of cognitive functioning, they are generally not considered to be primarily motoric. Additionally, similar deficits can be present in frontal lobe stroke and are traditionally characterized as components of the dysexecutive syndrome. Therefore, since the basal ganglia has extensive reciprocal connections with the frontal lobe, the primary site of injury causing these deficits in the aforementioned populations, are we simply referring to a similar deficit in PD? Traditional models of dopamine depletion in the basal ganglia note that depletion occurs earliest in the putamen in individuals with PD. Since the putamen is the primary input to the motor circuit of the basal ganglia, the earliest observable and treated speech-language symptoms observed in individuals with PD are typically considered motor speech. As such, they are quantified as hypokinetic dysarthria. However, in the early stages of PD prior to the emergence of motor speech deficits, it is unclear how the â€œcognitiveâ€ or dorsolateral-frontal circuit is
83 potentially disrupted. Therefore, it is unknown if the observed â€œspeechâ€ changes are truly motoric or if they represent a coexisting reduction in the executive system, specifically self monitoring of the quality of verbal output. Additionally, it is unclear if minor linguistic deficits are present in early PD but masked by the features of hypokinetic dysarthria. Traditional models of expressive language functioning have not consistently and carefully considered the coexisting role of the basal ganglia in action planning, intention, temporal encoding and intentionally guided attention. Therefore, it is unknown if these processes may have a greater influence on expressive language performance that previously reported. The factors hypothesized and discussed in this study, may offer a greater understanding of the language deficits noted in PD and offer alternative insights into novel treatments. Models of Basal Ganglia Function and Treatment Development The role of the basal ganglia in executive control and in expressive language performance is uncertain. As a result, language has generally been ignored in the remediation of individuals with PD. Therefore, future treatment studies in individuals with PD should consider the basal gangliaâ€™s role in fronto-executive-intentional behavior and the potential influences on expressive language. Recent studies in aphasia have suggested that short term intensive treatments are efficacious for language recovery and measurable neuroplasticity can be achieved. While plasticity is typically reserved for the cerebral cortex, improvement in the strength of basal ganglia-cortical connections could conceivably improve speech and language deficits observed in PD. Therefore, treatment designs must emphasize models of language action planning, language intention, temporal encoding for language and intentionally guided attention for language.
84 Exo-evoked Cuing Strategies The preliminary results of this study suggest that exo-evoked cuing strategies may facilitate increased language production in individuals with PD. Compensatory strategies for the hypothesized deficits in language action planning, temporal encoding for language and intention for expressive language may improve initiation and maintenance deficits observed in expressive language. Further, the basic principles of exo-evoked cuing may also have implications for motor speech. Currently walkers and canes exist that provide visual cuing for deficits in ambulation freezing resulting in improved and safe ambulation. The provision of external visual cuing for motor initiation is the basic principle underlying the development of these devices. Therefore, similar strategies may emphasize visual and auditory cuing to facilitate verbal initiation and maintenance of language productions. Currently, there are many alerting devices used primarily for memory. Similar devices may be used for expressive language and serve as an external executive device for communication. Concept Manipulation Nadeau & Gonzalez-Rothi (2004) proposed that concept manipulation may have a declarative component that represents a set of skills. It may be hypothesized that breakdowns in the narrative action plan as observed in narrative generation in this study, may serve as a potential skills amenable to treatment. This hypothesis is based on the notion that narrative generation is guided by a global language action plan for narrative completion. Nadeau & Gozalez-Rothi (2004) hypothesized concept manipulation in regards to phonological therapy and cleft constructions, however, the basic principles underlying their approach may have some implications for long term language action planning and narrative discourse construction. Finally, they note that extensive practice
85 of a variety of constructions is required for improvement of the hypothesized concepts. The manipulation of language action planning is a dramatic attempt to extend the concepts proposed by Nadeau & Gonzalez-Rothi and significant review of this hypothesized treatment direction will be required. Final Considerations Treatment designs should also consider the presentation of interventions early versus late in the disease process to investigate the potential for slowing the effects of the disease progression. Early interventions may also target language and other cognitive processes prior the dramatic decline in motor function that is commonly observed. Treatment studies for individuals with PD must also consider and attempt to integrate the following considerations: disease state, disease duration, predominant features (tremor vs rigidity), indo-evoked vs exo-evoked cuing mechanisms, apathy and depression among others. Finally, traditional treatment principles such as intensity, frequency, duration as well as compensation versus remediation must be factored into the treatment design and proposed hypotheses.
CHAPTER 7 CONCLUSIONS The role of the basal ganglia in language functioning remains a source of debate and intrigue for those who treat individuals with disorders of the basal ganglia, as well as those who have devoted their research careers to understanding its complex role in higher cortical functioning. The findings of this study should be evaluated with caution particularly due to the small sample size and the lack of control of a number of variables that may potentially indirectly influence expressive language functioning. However, these findings suggest that additional study is needed to clarify the role of the basal ganglia in generative/intentional behaviors such as expressive language. Studies of individuals with PD have generally focused upon motor speech changes while omitting language related issues. However, emerging models of basal ganglia functioning suggest that it might have a greater role in language functioning. The findings of this study suggest other factors related to basal ganglia functioning that are independent of motor functioning could influence expressive language performance. Additional study is needed to understand the basal gangliaâ€™s parallel role as central executor and its potential influence on language form and language use in narrative discourse. Therefore, the findings of this study should serve as a catalyst for future studies. The results of this study may offer a number of research possibilities for future studies to increase our understanding of the role of the basal ganglia in expressive language performance. A number of theoretical factors must be considered in the 86
87 design of future studies. Comparisons of disruption of basal ganglia function between individuals with PD and those who have sustained a basal ganglia stroke should be completed. These comparisons can evaluate disruption to the structure resulting from disease (PD) thought to encompass the entire basal ganglia as opposed to infarction, which is typically limited to the striatum. Additional studies should also consider the proposed role of the basal ganglia proposed by this study in relationship to connectionists models of language functioning. Models on intention must also be considered in relationship to models of language functioning. Alternate explanations such as depression and apathy must be controlled in future studies. Finally, as imaging techniques advance, concurrent imaging and expressive language tasks should be designed. Ultimately, it is our responsibility to improve our understanding of the basal gangliaâ€™ role in expressive language in order to improve the treatments that we offer those with diseases and disorders of the basal ganglia. Currently, most interventions are pharmacological and it is unclear if behavioral interventions can facilitate improvements in expressive language performance and/or if behavioral treatments may serve as a supplement to established treatments. Future studies should consider these issues in addition to the complexities of basal ganglia function.
APPENDIX A INSTRUCTIONS FOR NARRATIVE TASK I am going to ask you to discuss three events or topics. I want you to discuss each event or topic with as much detail as you possibly can. I would like for you to speak about each event or topic for at least three minutes. It is okay for you to speak longer than three minutes and feel free to speak as long as you would like. If you stop before three minutes have elapsed, I will ask you to â€œTell me moreâ€. After you have finished speaking on each event or topic I will ask you â€œIs that all?â€ 88
APPENDIX B TRANSCRIPTION FORM ID: Words Communication Unit GC LC Ties Total Time 89
APPENDIX C DYSARTRIA STAGES Stage 1 No detectable disorder Stage 2 Obvious speech disorder with intelligible speech Stage 3 Reduction in speech intelligibility Stage 4 Natural speech supplemented by augmented devices Stage 5 No functional speech Stages of Functional Limitations in Dysarthria as defined by Yorkston et al., (1999) 90
APPENDIX D GLOBAL COHERENCE RATINGS Score of 5 â€“ The utterance provided substantial information related to the general topic Score of 4 â€“ The utterance contains multiple clauses, with one clause relating directly to the topic and the other relates indirectly. Score of 3 â€“ The utterance provides information possibly related to the general topic or is an evaluation statement without providing substantial information or the topic must be inferred from the statement. Score of 2 â€“ The utterance contains multiple clauses, with one clause possibly related to the general topic and one does not. Score of 1 â€“ The utterance is unrelated to the general topic or is a comment on the discourse. Coherence ratings as defined by Van Leer and Turkstra (1999). 91
APPENDIX E DEFINITION OF COHESIVE MARKERS AND COHESIVE ADEQUACY Cohesive Markers Reference: personal pronouns, demonstrative pronouns and comparative pronouns are cohesive. The identity of the thing or class or things referred to is found in the preceding or following text. Conjunctions: conjunctions are cohesive indirectly as they express meaning that presuppose the presence of other discourse components. They specify the relation of content between sentences. Lexicals: general nouns are cohesive by selection of vocabulary. Cohesive adequacy Complete tie: a tie is complete if the referent can be found and defined with no ambiguity. Incomplete tie: a tie is incomplete if the referent can be found and defined with no ambiguity. Erroneous: a tie is judged erroneous if the listener is guided to ambiguous or erroneous information. Definitions and ratings as defined by Van Leer and Turkstra (1999). 92
APPENDIX F DEMOGRAPHIC DATA AND COGNITIVE SCORES FOR ALL SUBJECTS ID AGE ED MMSE WMS BNT H&Y PY TYPE P1 85 12 26 12 53 2 0 tremor P2 74 12 30 43 52 3 0 tremor P3 84 11 29 9 53 3 4 tremor P4 78 12 29 37 51 3 7 rigid P5 64 14 27 36 60 2 .5 rigid P6 84 12 26 13 33 2 0 tremor P7 69 14 30 25 56 2 0 tremor P8 61 12 29 28 55 3 9 rigid P9 76 12 29 38 56 3 14 tremor P10 83 9 30 34 50 2 0 tremor P11 64 12 29 20 58 2 7 tremor P12 40 12 29 35 56 2 2 tremor C1 64 14 30 53 55 C2 79 12 30 24 55 C3 71 14 30 32 57 C4 73 7 27 15 50 C5 70 16 30 38 60 C6 84 16 26 19 53 C7 73 12 29 31 47 C8 83 9 27 17 36 C9 64 12 30 33 60 C10 84 14 30 30 52 C11 89 12 26 15 36 C12 38 16 30 60 60 93
LIST OF REFERENCES Alexander, M.P. (2002). Disorders of language after frontal lobe injury: evidence for the neural mechanisms of assembling language. In D. T. Stuss & R. T. Knight (Eds.) Principles of Frontal Lobe Function (pp. 159-167). Oxford: Oxford University Press. Alexander, G.E., Crutcher, M.D., & DeLong, M.R. (1990). Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, "prefrontal" and "limbic" functions. Progress in Brain Research, 85, 119-46. Alexander, G.E., DeLong, M.R., & Strick, P.L. (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience, 9, 357-81. Alexander, M.P., Naeser, M.A., & Palumbo, C.L. (1987). Correlations of subcortical CT lesion sites and aphasia profiles. Brain, 110, 961-991. Auriacombe, S., Grossman, M., Carvell, S., Gollomp, S., Stern, M.B., & Hurtig, H.I. (1993). Verbal fluency deficits in Parkinson's disease. Neurosychology, 7(2), 182-192. Baldo, J.V., & Shimamura, A.P. (1998). Letter and category fluency in patients with frontal lobe lesions. Neuropsychology, 12(2), 259-267. Beiser, D.G., & Houk, J.C. (1998). Model of cortical-basal ganglionic processing: encoding the serial order of sensory events. Journal of Neurophysiology, 79(6), 3168-3188. Berg, E., Bjornram, C., Hartelius, L., Laakso, K., & Johnels, B. (2003). High-level language difficulties in Parkinsonâ€™s disease. Clinical Linguistics & Phonetics, 17(1), 63-80. Brenneise-Sarshad, R., Nicholas, L.E., & Brookshire, R.H. (1991). Effects of apparent listener knowledge and picture stimuli on aphasic and non-brain-damaged speakers' narrative discourse. Journal of Speech and Hearing Research, 34(1), 168-176. Brown, R.G., & Marsden, C.D. (1988). Internal versus external cues and the control of attention in Parkinsonâ€™s disease. Brain, 111(2), 323-345. Brunner, R.J., Kornhuber, H.H., Seemuller, E., Suger, G., & Wallesch, C. (1982). Basal ganglia participation in language pathology. Brain and Language, 16, 281-299. 94
95 Chapman, S.B., Culhane, K.A., Levin, H.S., Harward, H., Mendelsohn, D., EwingCobbs, L., Fletcher, J.M., & Bruce, D. (1992). Narrative discourse after closed head injury in children and adolescents. Brain and Language, 43(1), 42-65. Chiodo, L. A., & Berger, T.W. (1986). Interactions between dopamine and amino acid induced excitation and inhibition in the striatum. Brain Research, 375(1), 198-203. Coelho, C.A. (2002). Story narratives of adults with closed head injury and non-brain injured adults: influence of socioeconomic status, elicitation task, and executive functioning. Journal of Speech Language and Hearing Research, 45(6), 1232-48. Coelho, C.A. (1995). Discourse production deficits following traumatic brain injury: a critical review of the recent literature. Aphasiology, 9(5), 409-429. Coelho, C., & Flewellyn, L. (2003). Longitudinal assessment of coherence in an adult with fluent aphasia: a follow-up study. Aphasiology, 17(2), 173-182. Cohen, J.D., Braver, T.S., & Brown, J.W. (2002). Computational perspectives on dopamine function in prefrontal cortex. Current Opinion in Neurobiology, 12(2), 223-9. Collins, A.M., & Loftus, E.F. (1975). A spreading-activation theory of semantic processing. Psychological Review, 82, 407-428. Cools, R., Stefanova, E., Barker, R. A., Robbins, T. W., & Owen, A. M. (2002). Dopaminergic modulation of high-level cognition in Parkinson's disease: the role of the prefrontal cortex revealed by PET. Brain, 125(Pt 3), 584-594. Copland, D.A. (2003). The basal ganglia and semantic engagement: potential insights from semantic priming in individuals with subcortical vascular lesions, Parkinsonâ€™s disease, and cortical lesions. Journal of the International Neuropsychological Society, 9, 1041-1052. Copland, D.A., Chenery, H.J., & Murdoch, B.E. (2000a). Persistent deficits in complex language function following dominant nonthalamic subcortical lesions. Journal of Medical Speech-Language Pathology, 8(1), 1-14. Copland, D.A., Chenery, H.J., & Murdoch, B.E. (2000b). Processing lexical ambiguities in word triplets: evidence of lexical-semantic deficits following dominant nonthalamic subcortical lesions. Neuropsychology, 14(3), 379-390. Copland, D.A., Chernery, H.J., & Murdoch, B.E. (2000c). Understanding ambiguous words in biased sentences: evidence of transient contextual effects in individuals with nonthalamic subcortical lesions and Parkinsonâ€™s disease. Cortex, 36, 601-622. Correia, L., Brookshire, R.H., & Nicholas, L.E. (1990). Aphasic and non-brain-damaged adults' descriptions of aphasia test pictures and gender-biased pictures. Journal of Speech and Hearing Disorders, 55(4), 713-720.
96 Costello, A.L., & Warrington, E. K. (1989). Dyanamic aphasia. The selective impairment of verbal planning. Cortex, 25, 103-144. Crosson, B. (1999). Subcortical mechanisms in language: lexical-semantic mechanisms and the thalamus. Brain and Cognition, 40(2), 414-438. Crosson, B. (1992a). Subcortical Functions in Language and Memory. New York: The Guildford Press. Crosson, B. (1992b). Is the striatum involved in language. In G. Vallar, S.F. Cappa & C.W. Wallasch (Eds.) Neuropsychological Disorders Associated with Subcortical Lesions (pp 268-293). New York: Oxford University Press. 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., Moore, A.B., Gopinath, K., White, K.D., Wierenga, C.E., Gaiefsky, M.E., Fabrizio, K.S., Peck K.K., Soltysik, D., Milsted, C., Briggs, R.W., Conway, T.W., & Gonzalez-Rothi, L.J. (2005). Role of the right and left hemispheres in recovery of function during treatment of intention in aphasia. Journal of Cognitive Neuroscience, 17(3), 392-406. 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 Neurosceince, 13(2), 272-283. Crosson, B., Zawacki, T., Brinson, G., Lu, L., & Sadek, J.R. (1997). Models of subcortical functions in language: current status. Journal of Neurolinguistics, 10(4), 277-300. Dagher, A., Owen, A.M., Boecker, H., & Brooks, D.J. (2001). The role of the striatum and hippocampus in planning: a PET activation study in Parkinson's disease. Brain, 124(5), 1020-1032. Damasio, A.R., & Anderson, S.W. (2003). The Frontal Lobes. In K. M. Heilman & E. Valenstein (Eds.) Clinical Neuropsychology (4th ed., pp. 404-446). New York: Oxford. Damasio, A.R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed neural systems. Proceedings of National Academy of Science U S A, 90(11), 4957-4960.
97 Doyle, P.J., McNeil, M.R., Spencer, K.A., Goda, A.J., Cottrell, K., & Lustig, A.P. (1998). The effects of concurrent picture presentations on retelling of orally presented stories by adults with aphasia. Aphasiology, 12(7/8), 561-574. Dubois, B., & Pillon, B. (1988). Cognitive deficits in Parkinsonâ€™s disease. Journal of Neurology, 244(1), 2-8. Duffy, J. R. (1995). Motor Speech Disorders: Substrates, Differential Diagnosis and Management. St. Louis: Mosby. Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). Mini-mental state: a practical method for grading the state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198. Friederici, A.D., Kotz, S.A., Werheid, K., Hein, G., & von Cramon, D.Y. (2003). Syntactic comprehension in Parkinsonâ€™s disease: investigating early automatic and late integrational processes using event-related brain potentials. Neuropsychology, 17(1), 133-142. Fromm, D., Holland, A.L., Swindell, C.S., & Renmuty, O.M. (1985). Various consequences of subcortical stroke. Prospective study of 16 consecutive cases. Archives of Neurology, 42, 943-950. Glosser, G. (1993). Discourse patterns in neurologically impaired and aged populations. In H.H. Browell and Y. Joanette (Eds.) Narrative discourse in neurologically impaired and normal aging adults (pp.191-211). San Diego: Singular Publishing. Glosser, G., & Deser, T. (1990). Patterns of discourse production among neurological patients with fluent language disorders. Brain and Language, 40, 67-88. Golbe, L. I. (1998). Epidemiology of movement disorders. In J. Jankovic & E. Tolosa (Eds.) Parkinson's disease and movement disorders (pp. 119-132). Baltimore: Williams & Wilkins. Gold, M., Nadeau, S.E., Jacobs, D.H., Adair, J.C., Gonzalez-Rothi, L.J., & Heilman, K.M. (1997). Adynamic aphasia: a transcortical motor aphasia with defective semantic strategy formation. Brain and Language, 57, 374-393. Grossman, M. (1999). Sentence processing in Parkinsonâ€™s disease. Brain and Cognition, 40, 387-413. Grossman, M. (1998). Not all words are created equal. Category-specific deficits in central nervous system disease. Neurology, 50(2), 324-325. Grossman, M., Lee, C., Morris, J., Stern, M.B., & Hurtig, H.I. (2002b). Assessing resource demands during sentence processing in Parkinsonâ€™s disease. Brain and Language, 80, 603-616.
98 Grossman, M., Zurif, E., Lee, C., Pranther, P., Kalmanson, J., Stern, B., & Hurting, H.I. (2002a). Information processing speed and sentence comprehension in Parkinsonâ€™s disease. Neuropsychology, 16(2), 174-181. Gurney, K.N., Prescott, T.J., & Redgrave, P. (1998). The basal ganglia viewed as an action selection device. In the Proceedings of the Eighth International Conference on Artificial Neural Networks. Skovde, Sweden, 1033-1038. Halliday, M., & Hasan, R. (1976). Cohesion in English. London: Longman Group Han, M., Kang, D., Bae, H., Oh, G., Jeong, S., & Roh, J. (2003). Aphasia in striatocapsular infarction may be explained by concomitant small cortical infarctions of cortical language zones. Stroke, 34, 259. Harrington, D.L., Haaland, K.Y., & Hermanowicz, N. (1998). Temporal processing in the basal ganglia. Neuropsychology, 12(1), 3-12. Heilman, K., Watson, R., & Valenstein, E. (2003). Neglect and related disorders. In K. Heilman & E. Valenstein (Eds.) Clinical Neuropsychology (4th ed., pp. 296-346). New York: Oxford University Press. Henry, J. D., & Crawford, J. R. (2004). Verbal fluency deficits in Parkinson's disease: a meta-analysis. Journal of the International Neuropsychological Society, 10(4), 608-622. Hillis, A.E., Barker, P.B., Wityk, R.J., Aldrich, E.M., Restrepo, L., Breese, E.L., & Work, M. (2004). Variability in subcortical aphasia is due to variable sites of cortical hypoperfusion. Brain and Language, 89, 524-530. Hillis, A.E., Wityk, R.J., Barker, P.B., Beauchamp, N.J., Gailloud, P., Murphy, K., Cooper, O., & Metter, E.J. (2002). Subcortical aphasia and neglected in acute stroke: the role of subcortical hypofusion. Brain, 125, 1094-1104. Hoehn, M.H., & Yahr, M.D. (1967). Parkinsonism: onset, progression and mortality. Neurology, 17, 427. Illes, J. (1989). Neurolinguistic features of spontaneous language production dissociate three forms of neurodegenrative disease: Alzheimerâ€™s, Huntingtonâ€™s, and Parkinsonâ€™s. Brain and Language, 37, 628-642. Illes, J., Metter, E.J., Hanson, W.R., & Iritani, S. (1988). Language production in Parkinsonâ€™s disease: acoustic and linguistic considerations. Brain and Language, 33, 146-160. Kaplan, E., Goodglass, H., & Weintraub, S. (1983). Boston naming test scoring booklet. Philadelphia: Lea & Febiger.
99 Kiyatkin, E.A., & Rebec, G.V. (1996). Dopaminergic modulation of glutamate-induced excitations of neurons in the neostriatum and nucleus accumbens of awake, unrestrainted rats. Journal of Neurophysiology, 75(1), 142-153. Liang, C.L., Chang, H.W., Lee, T.C., Liliang, P.C., Lu, C.H., & Chen, H.J. (2001). Early prediction of aphasia outcome in left basal ganglia hemorrhage. Acta Neurologica Scandinavica, 103, 148-152. Liles, B.Z., Coelho, C.A., Duffy, R.J., & Zalagens, M.R. (1989). Effects of elicitation procedures on the narratives of normal and closed head-injured adults. Journal of Speech and Hearing Disorders, 54(3), 356-66. Love, T., Swinney, D., Wong, E., & Buxton, R. (2002). Perfusion imaging and stroke: a more sensitive measure of the brain bases of cognitive deficits. Aphasiology, 16(9), 873-883. Mega, M.S., & Alexander, M.P. (1994). Subcortical aphasia: the core profile of capsulostriatal infarction. Neurology, 44, 1824-1829. Middleton, F.A., & Strick, P.L. (2000). Basal ganglia and cerebellar loops: motor and cognitive circuits. Brain Research Brain Research Reviews, 31(2-3), 236-250. Mink, J.W. (1996). The basal ganglia: focused selection and inhibition of competing motor programs. Progress in Neurobiology, 50, 381-425. Morris, R.G., Downes, J.J., Sahakian, B.J., Evenden, J.L., Heald, A., & Robbins, T.W. (1988). Planning and spatial working memory in Parkinson's disease. Journal of Neurology, Neurosurgery and Psychiatry, 51(6), 757-766. Murray, L.L. (2000). Spoken language production in Huntingtonâ€™s and Parkinsonâ€™s diseases. Journal of Speech, Language, and Hearing Research, 43, 1350-1366. Murray, L.L., & Lenz, L.P. (2001). Productive syntax abilities in Huntingtonâ€™s and Parkinsonâ€™s diseases, Brain and Cognition, 46, 213-219. Murray, L.L., & Stout, J. C. (1999). Discourse comprehension in Huntington's and Parkinson's disease. American Journal of Speech-Language Pathology, 8, 137-148. Nadeau, S.E., & Crosson, B. (1997). Subcortical aphasia. Brain and Language, 58, 355 402. Nadeau, S.E., & Gonzalez-Rothi, L.J. (2004). Rehabilitation of language disorders. In J. Ponsford (Ed.) Cognitive and Behavioral Rehabitation: From Neurobiology to Clinical Practice (pp. 129-174). New York: The Guilford Press. Nambu, A., Tokuno, H., Hamada, I., Kita, H., Imanishi, M., Akazawa, T., Ikeuchi, Y., & Hasegawa, N. (2000). Excitatory cortical inputs to pallidal neurons via the subthalamic nucleus in the monkey. Journal of Neurophysiology, 84, 289-300.
100 Nambu, A., Tokuno, H., & Takada, M. (2002). Functional significance of the cortico-subthalamo-pallidal hyperdirect pathway. Neuroscience Research, 43, 111-117. Nicholas, L.E., & Brookshire, R H. (1995). Presence, completeness and accuracy of main concepts in the connected speech of non-brain-damaged adults and adults with aphasia. Journal of Speech and Hearing Research, 38(1), 145-156. Nieoullon, A. (2002). Dopamine and the regulation of cognition and attention. Progress in Neurobiology , 67(1), 53-83. Nieoullon, A., & Coquerel, A. (2003). Dopamine: a key regulator to adapt action, emotion, motivation and cognition. Current Opinion in Neurology, 16 (Suppl 2), S3-9. Olness, G.S., Ulatowska, H.K., Wertz, R.T., Thompson, J.L., & Auther, L.L. (2002). Discourse elicitation with pictorial stimuli in African Americans and Caucasians with and without aphasia. Aphasiology, 16(4/5/6), 623-633. Owen, A.M. (1997). Cogntive planning in humans: neuropsychological, neuroanatomical, and neuropharmacological perspectives. Progress in Neurobiology, 53, 431-450. Owen, A.M., Doyon, J., Dagher, A., Sadikot, A., & Evans, A.C. (1998). Abnormal basal ganglia outflow in Parkinson's disease identified with PET. Implications for higher cortical functions. Brain, 121 (5), 949-965. Owen, A.M., Sahakian, B.J., Hodges, J.R., Summers, B.A., Polkey, C.E., & Robbins, T.W. (1995). Dopamine-dependent frontostriatal planning deficits in early Parkinson's disease. Neuropsychology, 9(1), 126-140. Parent, A., & Hazrati, L. (1995). Functional anatomy of the basal ganglia I. The cortico-basal ganglia-thalamo-cortical loop. Brain Research Reviews, 20, 91-127. Peran, P., Rascol, O., Demonet, J.F., Celsis, P., Nespoulous, J.L., Dubois, B., & Cardebat, D. (2003). Deficit of verb generation in nondemented patients with Parkinson's disease. Movement Disorders, 18(2), 150-156. Piatt, A.L., Fields, J.A., Paolo, A.M., Koller, W.C., & Troster, A.I. (1999). Lexical, semantic, and action verbal fluency in Parkinson's disease with and without dementia. Journal of Clinical and Experimental Neuropsychology, 21(4), 435-443. Pierce, R.C., & Rebec, G.V. (1995). Iontophoresis in the neostriatum of awake, unrestrained rats: differential effects of dopamine, glutamate and ascorbate on motor and nonmotor-related neurons. Neuroscience, 67(2), 313-324. Potechin, G.C., Nicholas, L.E., & Brookshire, R.H. (1987). Effects of picture stimuli on discourse production by aphasic patients. In R. H. Brookshire (Ed.) Clinical Aphasiology Proceedings (pp. 216-220). Minneapolis: BRK Publishers.
101 Prescott, T.J., Gurney, K., Montes-Gonzalez, F., Humphries, M., & Redgrave, P. (2002). The robot basal ganglia: action selection by an embedded model of the basal ganglia. In L.F.B. Nicholson & R.L.M. Faull (Eds.) Basal Ganglia VII (pp. 349-356). New York: Plenum Press. Radanovic, M., & Scaff, M. (2003). Speech and language disturbances due to subcortical lesions. Brain and Language, 84, 337-352. Riederer, P., Gerlach, M., & Foley, P. (2002). Neurotransmitters and pharmacology of the basal ganglia. In J.J. Jankovic & E. Toloso (Eds.) Parkinson's Disease and Movement Disorders (pp. 23-37). Philadelphia: Lippincott Williams & Wilkins. Robin, D., & Scheinberg, S. (1990). Subcortical lesions and aphasia. Journal of Speech and Hearing Disorders, 55, 90-100. Robinson, G., Blair, J., & Cipolotti, L. (1998). Dynamic aphasia: an inability to select between competing verbal responses? Brain, 121, 77-89. Rolls, E. T., Thorpe, S. J., Boytim, M., Szabo, I., & Perrett, D. I. (1984). Responses of striatal neurons in the behaving monkey. Effects of iontophoretically applied dopamine on normal responsiveness. Neuroscience, 12(4), 1201-1212. Rumelhart, D.E., McClelland, J.L., & PDP Research Group. (1986). Parallel distributed processing: explorations in the microstructures of cognition. Vol. 1. Foundations. Cambridge: MIT Press. Shadden, B. B., Burnette, R. B., Eikenberry, B. R., & DiBrezzo, R. (1991). All discourse tasks are not created equal. Clinical Aphasiology, 20, 327-341. Shimamura, A. (1994). Memory and frontal lobe function. In M. Gazzinga (Ed.) The Cognitive Neurosciences (pp. 803-813). Cambridge: MIT Press. Simpson, G. B. (1984). Lexical ambiguity and its role in models of word recognition. Psychological Bulletin, 96(2), 316-340. Sirigu, A., Zalla, T., Pillon, B., Grafman, J., Dubois, B., & Agid, Y. (1995). Planning and script analysis following prefrontal lobe lesions. Annals of N Y Academy of Science, 769, 277-288. Skeel, R.L., Crosson, B., Nadeau, S.E., Algina, J., Bauer, R.M., & Fennell, E.B. (2001). Basal ganglia dysfunction, working memory, and sentence comprehension in patients with Parkinson's disease. Neuropsychologia, 39(9), 962-971. Smith, R., Heuerman, M., Wilson, B.M., & Proctor, A. (2003). Analysis of normal discourse patterns. Brain and Cognition, 53(2), 368-371.
102 Stout, J., Wylie, S., Simone, P., & Simers, E.R. (2001). Influence of competing distractors on response selection in Huntington's disease and Parkinson's disease. Cognitive Neuropsychology, 18(7), 643-653. Taylor, A.E., & Saint-Cyr, J.A. (1995). The neuropsychology of Parkinson's disease. Brain and Cognition, 28(3), 281-296. Taylor, A.E., Saint-Cyr, J.A., & Lang, A.E. (1986). Frontal lobe dysfunction in Parkinson's disease. The cortical focus of neostriatal outflow. Brain, 109 (5), 845-883. Troster, A.I., Woods, S.P., Fields, J.A., Hanisch, C., & Beatty, W.W. (2002). Declines in switching underlie verbal fluency changes after unilateral pallidal surgery in Parkinson's disease. Brain and Cognition, 50(2), 207-217. Ulatowska, H.K., Allard, L., & Chapman, S. (1990). Narrative and procedural discourse in aphasia. In Y. Joanette and H. Brownell (Eds.) Discourse Ability and Brain Damage: Theoretical and Impirical Perspectives (pp. 180-199). New York: Springer Verlag. Ulatowska, H.K., & Chapman, S.B. (1994). Discourse macrostructure in aphasia. In R. Bloom, L.K. Obler, S.DeSanti, & J.S. Ehrlich (Eds.) Discourse Analysis and Applications: Studies in Adult Clinical Populations (pp. 29-46). New Jersey: Lawrence Erlbaum Associates. Ulatowska, H.K., & Olness, G.S. (1997). Some observations on narratives by aphasics and their contribution to narrative theory. Journal of Narrative and Life History, 7, 259-264. Ullman, M. T. (2004). Contributions of memory circuits to language: the declarative/procedural model. Cognition, 92(1-2), 231-270. Ullman, M.T. (2001a). A neurocognitive perspective on language: the declarative/procedural model. Nature Reviews Neuroscience, 2(10), 717-726. Ullman, M. T. (2001b). The declarative/procedural model of lexicon and grammar. Journal of Psycholinguistic Research, 30(1), 37-69. Ullman, M.T., Corkin, S., Coppola, M., Hickok, G., Growdon, J.H., Koroshetz, W.J., & Pinker, A. (1997). A neural dissociation within language: evidence that the mental dictionary is part of declarative memory, and that grammatical rules are processed by the procedural system. Journal of Cognitive Neuroscience, 9(2), 266-276. Ullman, M.T., Pancheva, R., Love, T., Yee, E., Swinney, D., & Hickok, G. (2005). Neural correlates of lexicon and grammar: evidence from the production, reading, and judgment of inflection in aphasia. Brain and Language, 93(2), 185-238.
103 Van Leer, E., & Turkstra, L. (1999). The effect of elicitation task on discourse coherence and cohesion in adolescents with brain injury. Journal of Communication Disorders, 32(5), 327-48. Wechsler, D. (1997). Wechsler Memory Scale III. New York: Psychological Corporation. Weiller, C., Willmes, K., Reiche, W., Thron, A., Isensee, C., Buell, U., & Ringelstein, E.B. (1993). The case of aphasia or neglect after striatocapsular infarction. Brain, 116, 1509-1525. Wichmann, T., & DeLong, M.R.. (2003). Functional neuroanatomy of the basal ganglia in Parkinson's disease. Advances in Neurology, 91, 9-18. Wilson, B. M., & Proctor, A. (2000). Oral and written discourse in adolescents with closed head injury. Brain and Cognition, 43(1-3), 425-429. Wilson, B. M., Smith, R., & Proctor, A. (2001). The validity of cognitive distance in oral and written discourse. Brain and Cognition, 46(1-2), 304-307. Yorkston, K. M., Beukelman, D. R., Strand, E. A., & Bell, K. R. (1999). Management of Motor Speech Disorders in children and Adults (2nd ed.). Austin: Pro-Ed. Young, A.B., & Penney, J.B.. (2002). Biochemical and functional organization of the basal ganglia. In J. Jankovic and E. Tolosa (Eds.) Parkinson's Disease and Movement Disorders (pp.1-10). Philadelphia: Lippincott Williams & Wilkins.
BIOGRAPHICAL SKETCH Charles Ellis, Jr., was born in Claxton, Georgia, in 1967. After earning a B.S. in communication sciences and disorders from The University of Georgia in Athens, Georgia, in 1990, he began graduate study. Mr. Ellis earned an M.S. in communication sciences and disorders from The University of Georgia in 1992. He began his doctoral study in rehabilitation science at the University of Florida in 2002. Mr. Ellis completed the current study as his dissertation research and earned his doctoral degree in 2005. 104